american professional constructor journal - june 2010

Journal of the American Institute of Constructors

OCTOBER 2009Volume 33 • Number 2

T H E A M E R IC A N PROF E S SIONA L C ON ST RU C TOR

IN THIS ISSUE:

Sources of High School Junior’s and Senior’s Perception of the Construction Industry

Guidance Counselors’ Knowledge and Perception of Careers in the Construction Industry

Issues in Building Commissioning

Final Construction Cost Prediction by Using the Earned Value Index Application Method

Emergence of Green Buildings in India: A Study of Sustainable Features and Occupant Satisfaction

Analysis of the Preferences for Specific Project Delivery System Utilized by Texas Public Universities

Measuring Contractor’s Success to Meet the Reasonable Prudent Contractor Standard When Claiming for a Type I Differing Condition: A Differing Site Condition Anthology

THE AMERICAN PROFESSIONAL CONSTRUCTORJournal of the American Institute of Constructors

IN THIS ISSUE:

A Performance-Based Incentive Program for Flexible Pavement

A Conceptual Framework: Teaching Reality-Based Construction Project Management via Constrainted Virtual Construction Environment

Construction Safety: A Lean Construction Research Agenda

Integrated Collaborative Experiences: Construction and Architecture Programs in the Same College

The Impacts of Undergraduate Construction Internships on Recruitment, Training, and Retention of Entry-Level Employees of the Construction Industry

Construction Education at Texas A&M University: A Comparative Longitudinal Study of Graduates

JUNE 2010Volume 33 •Number 3

AIC 2009/2010Officers & Directors

PRESIDENTMark E. Giorgi, AIC Erie Affiliates, Inc 29017 Chardon Street, Suite 200 Willoughby Hills, OH 44092-1405 440.943.5995 [email protected]

VICE PRESIDENTAndrew J. Wasiniak, AIC, CPC Walbridge Aldinger 4852 79th Ave., Plaza E Sarasota, FL 34243 813.334.5179 [email protected]

SECRETARYDavid Fleming, AIC, CPC, DBIA Sundt Construction, Inc 4425 West Airport Fwy, Suite 473 Irving, TX 75062 972.258.0500 [email protected]

TREASURERPaul W. Mattingly, AIC, CPC Bossemattingly Constructors, Inc. 2116 Plantside Drive Louisville, KY 40299-1924 502.671.0995 [email protected]

PURPOSEThe purpose of the American Institute of Constructors is to promote individual excellence throughout the related fields of construction.

MISSIONOur mission is to provide:

A qualifying body to serve the individual in construction, the Constructor, who has achieved a recognized level of professional competence;

Opportunities for the individual constructor to participate in the process of developing quality standards of practice and to exchange ideas;

Leadership in establishing and maintaining high ethical standards;

Support for construction education and research;

Encouragement of equitable and professional relationships between the professional constructor and other entities in the construction process; and

An environment to enhance the overall standing of the construction profession.

Journal of the American Institute of Constructors

AIC PAST PRESIDENTS

1971-74 Walter Nashert, Sr., FAIC

1975 Francis R. Dugan, FAIC

1976 William Lathrop, FAIC

1977 James A. Jackson, FAIC

1978 William M. Kuhne, FAIC

1979 E. Grant Hesser, FAIC

1980 Clarke E. Redlinger, FAIC

1981 Robert D. Nabholz, FAIC

1982 Bruce C. Gilbert, FAIC

1983 Ralph. J. Hubert, FAIC

1984 Herbert L. McCaskill Jr.,FAIC

1985 Albert L Culberson, FAIC

1986 Richard H. Frantz, FAIC

1987 L.A. (Jack) Kinnaman, FAIC

1988 Robert W. Dorsey, FAIC

1989 T.R. Benning Jr., FAIC

1990 O.L. Pfaffmann, FAIC

1991 David Wahl, FAIC

1992 Richard Kafonek, FAIC

1993 Roger Baldwin, FAIC

1994 Roger Liska, FAIC

1995 Allen Crowley, FAIC

1996 Martin R. Griek, AIC

1997 C.J. Tiesen, AIC

1998-99 Gary Thurston, AIC

2000 William R. Edwards, AIC

2001-02 James C. Redlinger, FAIC

2003-04 Stephen DeSalvo, FAIC

2005-06 David R. Mattson, FAIC

2007-09 Stephen P. Byrne, FAIC, CPC

The American Institute of Constructors | PO Box 26334 | Alexandria, VA 22314 | Tel: 703.683.4999 | Fax: 571.527.3105 | www.professionalconstructor.org

construction fields.

AIC 2009/2010 Board of DirectorsDIRECTOR (ELECTED 2009–2012) Robert W. Arnold, CPC ASCO Hardware Company, Inc 1409 Osage, Redfield, AR 72132 501.376.6858 [email protected]

DIRECTOR (ELECTED 2009–2012) Paul M. Byrne, AC 6411 Lange Cir., Dallas, TX 75214 214.878.1634 [email protected]

DIRECTOR (ELECTED 2008–2011) Dennis C. Bausman, AIC, CPC, PhD Clemson University 126 Lee Hall, Clemson, SC 29634 864.656.3919 [email protected]

DIRECTOR (ELECTED 2008–2011) Matthew A. Conrad, AC The Christman Company 208 Capitol Ave., Lansing, MI 48933 517.482.1488 [email protected]

DIRECTOR (ELECTED 2010-2013) Allen L. Crowley, Jr., AIC The Crowley Group, LLC 12434 Cedar Rd, Ste 12 Cleveland, OH 44106 216.231.1100 [email protected]

DIRECTOR (ELECTED 2010-2013) Steven A. DeSalvo, FAIC, CPC Turner Construction Company 2315 Iowa Ave., Cincinnati, OH 45206 513.363.0845 [email protected]

DIRECTOR (ELECTED 2008–2011) David Fleming, AIC, CPC, DBIA Sundt Construction,Inc 4425 W. Airport Fwy, Ste 473 Dallas, TX 75062 972.258.0500 [email protected]

DIRECTOR (ELECTED 2008–2011) E. Terence Foster, AIC, CPC, PhD, PE University of Nebraska — Omaha 1014 N. 67th Cir., Omaha, NE 68132 402.554.3273 [email protected]

DIRECTOR (ELECTED 2009–2012) Michael A. Garrett, AIC, CPC Megen Construction Company. Inc. 2060 Miles Woods Dr., Cincinnati, OH 45231 513.616.1414 [email protected]

DIRECTOR (ELECTED 2010-2013) Mark E. Giorgi, AIC Erie Affiliates, Inc 29017 Chardon St., Ste 200 Willoughby Hills, OH 44092-1405 440.943.5995 [email protected]

DIRECTOR (ELECTED 2008–2011) Mike W. Golden, AIC, CPC MW Golden Corporation P.O. Box 338, Castle Rock, CO 80104 303.688.9848 [email protected]

DIRECTOR (ELECTED 2009–2012) Mark D. Hall, AIC, CPC Hall Construction Co., Inc PO Box 770, Howell, NJ 07731 732.938.4255 [email protected]

DIRECTOR (ELECTED 2008–2011) David C. Jesme, AIC, CPC,DBIA Sundt Construction, Inc 1660 Hotel Circle N., Ste 400 San Diego, CA 92108 619.321.4822 [email protected]

DIRECTOR (ELECTED 2008–2011) Paul W. Mattingly, AIC, CPC BosseMattingly Constructors 1916 Redleaf Dr., Louisville, KY 40242 502.671.0995 [email protected]

DIRECTOR (ELECTED 2008–2011) David R. Mattson, FAIC, CPC, MCIOB D.R. Mattson, Inc. P. O. Box 27842, Tempe, AZ 85285 480.970.3334 [email protected]

DIRECTOR (ELECTED 2009–2012) Samone Melson Smoot Construction 140 Brandy Mill Dr., Pataskala, OH 43062 513.623.4459 [email protected]

DIRECTOR (ELECTED 2007–2010) Philip F. Moffitt, AIC, CPC Nabholz Construction Company 21 Old Forge Ct., Little Rock, AR 72227 501.217.5513 [email protected]

DIRECTOR (ELECTED 2010-2013) Hoyt Monroe, FAIC Clark Power Corporation P.O. Box 45188, Little Rock, AR 72214 501.558.4901 [email protected]

DIRECTOR (ELECTED 2010-2013) Bradley T. Monson, AIC, CPC Tierra Group 150 South Eagle View Dr. Durango, CO 81301 970.375.6416 [email protected]

DIRECTOR (ELECTED 2007–2010) Wayne J. Reiter, AIC, CPC, CPA Reiter Companies 110 East Polk St., Richardson, TX 75081 972.238.1300 [email protected]

DIRECTOR (ELECTED 2010-2013) Bradford L. Sims, AIC Western Carolina University Kimmel School of Construction Management & Technology 211 Belk Building, Cullowhee, NC 28723 828.287.2175 [email protected]

DIRECTOR (APPOINTED—N OHIO) Bernard J. Ashyk, Jr, AIC Shook, Inc. 10245 Brecksville Rd, P.O. Box 41020 Brecksville, OH 44141-0020 440.838.5400 [email protected]

DIRECTOR (APPOINTED—DALLAS) Kyle B. Potts, AC Turner Construction 17659 Millwood Pl., Dallas, TX 75287 214.876.6760 [email protected]

THE AMERICAN PROFESSIONAL CONSTRUCTOR

Volume 33, Number 3 June 2010

Third class postage paid at Alexandria, Virginia. The American Professional Constructor (ISSN 0146-7557) is the official publication of the American Institute of Constructors (AIC), P.O. Box 26334 Alexandria VA 22314. Telephone 703.683.4999, Fax 703.683.5480, www.professionalconstructor.org.

Subscription rates: United States and Canada $100 per year, all other countries $150 per year. Single copies $50

Published in the USA by the American Institute of Constructors Education Foundation, and copyrighted by the American Institute of Constructors.

This publication or any part thereof may not be reproduced in any form without written permission from AIC. AIC assumes no responsibility for statements or opinions advanced by the contributors to its publications. Views expressed by them or the editor do not represent the official position of the The American Professional Constructor, its staff, or the AIC.

The American Professional Constructor is a refereed journal. All papers must be written and submitted in accordance with AIC journal guidelines available from AIC. All papers are reviewed by at least three experts in the field.

Articles

A Performance-Based Incentive Program for Flexible Pavement ................................................................. 7Wayne Jensen, Ph.D., P.E., Bruce Fischer, M. Arch., AIA, and Timothy Wentz, MBA, P.E.

A Conceptual Framework: Teaching Reality-Based Construction Project Management via Constrained Virtual Construction Environment ........................................................................................... 19

Dr. Borinara Park, Dr. Ronald Meier

Construction Safety: A Lean Construction Research Agenda .................................................................... 31Abdelhamid, T.S., Mitropoulos, P., Schafer, D. and Howell, G.A.

Integrated Collaborative Experiences - Construction and Architecture Programs in the Same College ........................................................................................................................................ 68

Richard C. Ryan AIC, CPC, LEED AP and Kenneth F. Robson AIC.

The Impacts of Undergraduate Construction Internships on Recruitment, Training, and Retention of Entry-Level Employees of the Construction Industry ............................................................ 76

David Bilbo, Jose L. Fernández-Solís, Nathan Bohne, M.Sc., and Mohamad Waseem

Construction Education at Texas A&M University: A Comparative Longitudinal Study of Graduates ......................................................................................................................................... 86

David Bilbo, Ph.D., Jose L. Fernandez-Solis, Ph.D, Kristen M. Ramsey-Souder, M.Sc.

Reviewer/Publication Interest Survey ......................................................................................................... 101

Construction Certification ............................................................................................................................ 102

Code of Ethics ............................................................................................................................................... 103

7

A Performance-Based Incentive Program for Flexible Pavement

Wayne Jensen, Ph.D., P.E., Bruce Fischer, M. Arch., AIA, and Timothy Wentz, MBA, P.E.

ABSTRACT: The Nebraska Department of Roads (NDOR) currently employs an incentive system to reward contractors for quality of materials and workmanship documented during the construction of flexible pavement. The NDOR was inter-ested in creating an incentive system that would reward contractors for producing flexible pavement with good medium-term (one to three years) performance characteristics. After studying pavement warranties and quality control measures used by other agencies, plus consulting with the NDOR and regional pavement contractors, researchers developed an incentive system for flexible pavement that is based upon two medium-term performance parameters, rutting and flushing. Three years after construction, flexible pavement with rutting less than 4 mm and flushing over less than twenty percent of the pavement surface would qualify for a monetary incentive. Three projects built in 2002-03 where construction quality bonuses rang-ing from 3.7% to 5.8% (of the asphalt materials cost) were paid were subsequently analyzed. Based upon the performance data specified, only one of these projects would have qualified for the proposed incentive. An incentive paid three years after construction has been completed should exceed any incentive(s) paid immediately after construction. The value of the recom-mended monetary incentive was thus established at 6% of asphalt materials cost. A performance-based incentive has the potential to provide benefits similar to pavement warranties, best value contracts or performance-based contracting proce-dures with significantly fewer legal entanglements.

Key Words:Performance-based incentive, flexible asphalt pavement, rutting and flushing pavement.

INTRODUCTION

The current procurement system for asphalt pave-ment used by the Nebraska Department of Roads (NDOR) utilizes a competitive sealed proposal with

the contract being awarded to the lowest responsible bidder (NDOR, 2007). Cost control continues to play a fundamental role throughout the pavement construc-tion process. The contractor purchases materials from a supplier, who is often the lowest bidder. As materi-als move through the contractor’s equipment, binder and aggregate properties are tested by the contractor, by NDOR personnel and/or by independent quality control technicians.

JUNE 2010 — Volume 33, Number 3The American Institute of Constructors | PO Box 26334 | Alexandria, VA 22314 | Tel: 703.683.4999 | Fax: 571.527.3105 | www.professionalconstructor.org

WAYNE JENSEN received his Ph.D. in civil engineering from the University of Wyoming. He currently teaches materials and methods courses as an associate professor in the Construction Management and Civil Engineering Departments at the University of Nebraska – Lincoln. Dr. Jensen has extensive experience in project planning, coordination and administration from working for the U.S. Army Corps of Engineers.

PROFESSOR BRUCE FISCHER, M. Arch, is a registered architect, an International Codes Council certified building plans examiner and an associate professor in the Construction Management Program at the University of Nebraska – Lincoln. He has worked as a commercial plan reviewer, project manager, and project architect on a wide range of architectural projects ranging from hospitals and prisons to convention centers and sports facilities. PROFESSOR TIM WENTZ, MBA, teaches courses in environmental systems, design/build, mechanical estimating and mechanical project management in the Construction Management Department for the University of Nebraska - Lincoln. Prof. Wentz has 20 years of experience in the construction industry as a design engineer and a project manager in a wide spectrum of commercial and residential projects.


8 A Performance-Based Incentive Program for Flexible Pavement

Workmanship or material quality below contractual specifications results in monetary disincentives lev-ied against the contractor.

The NDOR currently has a system of incentives in-place to reward contractors for pavement qual-ity based upon indices measured at completion of construction. Problems resulting from sub-standard materials or poor construction techniques often do not become noticeable until one to three years after construction has been completed. The cur-rent system used by the NDOR lacks incentives to encourage contractors to use materials or construc-tion techniques that might significantly improve the medium-term performance of asphalt pavement.This research sought to create applicable standards and determine values of monetary awards for con-tractors producing flexible pavement with good medium-term (one to three years) performance characteristics. Several pavement parameters were evaluated to determine which were most reflective of medium-term asphalt pavement performance. The concept of awarding monetary incentives to contrac-tors based upon levels of pavement performance at various points during a pavement’s lifespan was examined. The performance incentive system de-veloped allows contractors to receive current quality incentives for pavement built to construction specifi-cations, while subsequently providing a future mone-tary incentive for contractors who produce pavement that meets or exceeds established quality standards for medium-term performance.

CURRENT NDOR CONSTRUCTION INCENTIVES/DISINCENTIVES

The current NDOR quality incentive programs for asphalt pavement is based upon pavement smooth-ness and quality of materials immediately after construction has been completed. Smoothness pro-visions can be found in Section 502.08 and Section 1028 of the NDOR’s Supplemental and Standard Specifications for Highway Construction. Examples of adjustment factors for pavement smoothness are illustrated in Table 1, while examples of payment adjustment factors for materials and workmanship

are shown in Tables 2-4.

If the initial profile index is 10.0 inches/mile (in/mi) or less and bump removal is required, a sec-ond profilogram is taken after the bumps have been removed. The percent of pay for a profile index is then based upon the second profilogram. If the initial profile index exceeds 7 in/mi, then, except for total removal and replacement, the maximum percent of pay after bump removal is limited to 100 percent. Percent pay is based on a second run of the profilogram after bump removal. Smoothness test-ing is paid for at the lump sum unit price specified in the contract. This price is considered to be full compensation for all smoothness testing set forth in the specification (NDOR, 2007).

Table 1. The NDOR’s payment adjustment schedule for asphalt pavement smoothness

Profile Index (inches/lane mile)

Percent of Contract Prices

0 to 2.0 inches 105.0

More than 2.0 to 4.0 inches 102.0






More than 10.0 inches Corrective work required(Source: Section 502 – Asphaltic Concrete Pavement Smoothness

from the NDOR Supplemental and Standard Specifications for Highway Construction)


Wayne Jensen, Ph.D; Prof. Bruce Fischer, M. Arch; Professor Tim Wentz, MBA 9

A pay factor for smoothness of the top layer of asphalt pavement is determined according to the fol-lowing formula:

A (1.05) + B (1.02) + C (1.01) + D (1.00) + E (0.98) + F (0.95) + G (0.90)PF =

A + B + C + D + E + F + G

where:

A = length of pavement with a profile index of 0 to 2.0B = length of pavement with a profile index greater than 2.0 to 4.0C = length of pavement with a profile index greater than 4.0 to 5.0D = length of pavement with a profile index greater than 5.0 to 7.0E = length of pavement with a profile index greater than 7.0 to 8.0F = length of pavement with a profile index greater than 8.0 to 9.0

G = length of pavement with a profile index greater than 9.0 to 10.0

Table 2, illustrates the NDOR’s pay factors for asphalt materials. Payment is based upon the top layer of the driving lane asphalt materials (asphaltic concrete) only. Pay adjustments are calculated based on 0.1 mile sections (NDOR, 2007).

Table 2: The NDOR’s pay factors for asphaltic materialPay Factor Specified Property

1.00 Upper Limit +1% to 10%

Lower Limit

0.95 Greater than +10% to +15%

Less than -10% to -15%







0.40 or Reject Greater than +30% Less than -30%* If the resultant pay factor for the material is less than 0.70, the material shall be rejected if not already used. If incorporated in any work that is judged to be unsatisfactory, the material shall also be rejected. * If the pay factor is less than 0.70 and the material has been incorporated in work that is allowed to remain in place, the pay factor for the material shall be 0.40. (Source: Table 503.01 A Asphalt Materials – Pay Factors from the NDOR Supplemental and Standard Specifications for Highway Construction)

Tables 3 and 4, illustrate current NDOR material pay factors that serve as disincentives for poor quality. Pay factors based upon pavement density attempt to measure quality of both materials and workmanship (NDOR 2007).

Table 3: The NDOR’s schedule for acceptance – density of compacted asphaltic concrete (first lot)Average Density (5 Samples, Percent of Voidless Density)

Pay Factor

Greater than 90.0 1.00Greater than 89.5 to 90.0 0.95Greater than 89.0 to 89.5 0.70

89.0 or Less 0.40 or Reject(Source: Table 1028.21 Acceptance Schedule Density of Compacted Asphaltic Concrete (First Lot) from NDOR Supplemental and Standard Specifications for Highway Construction)

Table 4: The NDOR’s schedule for acceptance – density of compacted asphaltic concrete (subsequent lots)

Average Density (Five Samples, Percent of

Voidless Density)

Pay Factor

Greater than 92.4 1.00Greater than 91.9 to 92.4 0.95Greater than 91.4 to 91.9 0.90Greater than 90.9 to 91.4 0.85Greater than 90.4 to 90.9 0.80Greater than 89.9 to 90.4 0.70

89.9 or Less 0.40 or Reject(Source: Table 1028.22 Acceptance Schedule Density of Compacted Asphaltic Concrete (Subsequent Lot) from the NDOR Supplemental and Standard Specifications for Highway Construction)

Acceptance and pay factors for Superpave special (Type SPS) pavement are based on compacted-in-place average density and various properties of the aggregate and mix (Table 5).

Table 5: The NDOR’s production tolerance for Superpave asphaltic concrete

Test Allowable Single Test Deviation from Specification

Voids in the Mineral Aggregate - 0.75% to + 1.25% from Min.Dust to Asphalt Ratio None

Coarse Aggregate Angularity - 5% below Min.Fine Aggregate Angularity - 0.50% below Min.

These tolerances are applied to the mix design specification values, not the submitted mix design targets. (Source: Table 1028.19 Production Tolerances of Superpave Asphaltic Concrete from Section 1028 of the NDOR’s Supplemental and Standard Specifications for Highway Construction, revised 3-22-04)



Acceptance and pay factors for Superpave Type SP1, SP2, SP3, SP4, SP4 Special and SP5 are based on single test air voids, running average air voids, compacted in place average density, and production tolerance pay factors (NDOR, 2004).

When there is a production tolerance pay factor pen-alty, the penalty percentage is subtracted from the percent pay for single test air voids for each sublot affected. These three individual pay factors are then multiplied by each other to determine a total pay fac-tor for each sublot [(750 tons) (680 Mg)]. When any single test result on the same mix property from two consecutive quality control samples falls outside the allowable production tolerances of Table 5, the mate-rial represented by these tests can either be accepted with a 20% penalty or rejected at the discretion of the Project Engineer (NDOR, 2004). Examples of Superpave air void tolerances and acceptance factors are shown in Table 6.

Table 6: The NDOR’s schedule for acceptance – asphaltic concrete air voids

Air voids test results

Moving average of four

Single test

Less than 1.5% Reject Reject1.5% to less than

2.0%Reject 50%

2.0% to less than 2.5%

50% or Reject 95%


90% 95%


100% 100%

3.5% to 4.5% 102% 104%Over 4.5% to 5.0% 100% 100%Over 5.0% to 5.5% 95% 95%Over 5.5% to 6.0% 90% 95%Over 6.0% to 6.5% 50% or Reject 90%Over 6.5% to 7.0% Reject 50%

Over 7.0% Reject Reject(Source: Table 1028.20 Acceptance Schedule Air Voids - Ndes of Superpave Asphaltic Concrete from section 1028 of the NDOR’s Supplemental and Standard Specifications for Highway Construction, revised 3-22-04)

OTHER AGENCY’S PERFORMANCE-BASED INCENTIVES/DISINCENTIVES

Although many proposed roles for performance standards go well beyond current highway construc-tion practices, performance standards for highway construction are nothing new. Because pavement smoothness is widely recognized as important from a standpoint of both user satisfaction (no one likes to drive on a rough road) and long-term performance (smooth roads last longer and are often of higher overall quality than rougher roads), performance standards for asphalt pavement smoothness have seen widespread use (Carpenter, et al. 2003). Most highway agencies use smoothness specifications of one form or another. These specifications establish target values for smoothness measured using stan-dard engineering test methods that are related to user perceptions. Many agencies include incentives and/or disincentives to encourage high levels of smooth-ness that result in reduced operating costs for high-way users and reduced maintenance costs for the owner agencies. Current performance standards for smoothness and the results obtained from specifying performance standards are illustrated by examples from Arizona, Virginia, and Kansas.

For new construction, Arizona has established a target International Roughness Index (IRI) value of 41, with smoothness expressed in inches per mile. Incentives are earned for values below 38 and disin-centives are assessed for values in excess of 48. For rehabilitation projects, the target, incentive, and dis-incentive values vary as a function of highway type, the nature of the work to be performed, and (in some cases) the smoothness of the existing pavement. Tar-get smoothness is 39 to 68, while incentive thresh-olds vary from 37 to 66 (target value minus 2) and disincentive thresholds range from 49 to 78 (target value plus 10). Typical pavement smoothness incen-tives paid by the Arizona DOT average approximate-ly $7,500 per lane mile or about $1.00 per square yard. Average contractors in Arizona produce IRI smoothness values in the middle thirties. Some very good contractors consistently achieve IRI smooth-ness values in the low thirties, with substantial areas of pavement in the twenties (Richter, 2004).



Virginia has smoothness provisions for new con-struction and maintenance resurfacing, with smooth-ness expressed as IRI in inches per mile. For new construction, 100% payment is awarded for an IRI between 55 and 70 inches/mile. Bonus payments are earned for achieving IRI values less than 55 inches/mile and penalties are incurred for IRI values greater than 70 inches/mile, to a maximum of zero payment for IRI values greater than 160 inches/mile. Cor-rective action by the contractor is required when the average IRI for a pavement section exceeds 100 inches/mile (Richter, 2004).

For maintenance resurfacing, a maximum 10% bonus based on the asphaltic concrete surface cost is possible for interstate highway sections with an IRI less than 45 inches/mile and for non-interstate roads with an IRI of less than 55 inches/mile. Full pay-ment is reserved for interstates with IRI from 55 to 70 inches/mile, while non-interstates must have an IRI between 65 and 80 inches/mile for full payment (Richter 2004).

Unlike new construction projects, most resurfac-ing projects in Virginia are tested prior to and af-ter construction of new pavement. Resurfacing projects can be either straight overlay or mill-and-replace. Before-and-after testing is used to deter-mine the amount of improvement in ride quality. If the contractor is able to improve the quality by 30% or more, he/she is guaranteed an incentive for smoothness. For new construction, the contractor can receive an incentive of up to 5% based on IRI results. The amount of the incentive is based the total quantity of all asphaltic concrete used. Mainte-nance resurfacing contracts allow up to a 10% bonus (Richter, 2004).

Virginia has been actively using a “ride special provision” since the late 1990s. Most of the ride data have been collected on maintenance resurfac-ing projects. With more than 150 projects tabulated in 2002, the IRI on new interstate pavement aver-aged 60 inches/mile. The 2002 average was 67 inches/mile on U.S. routes and 74 inches/mile on State routes. Over the last six years, the average IRI on the interstates has stabilized while ride quality

on non-interstate routes has continued to improve (Richter, 2004).

In addition to improved ride quality, Virginia has seen other benefits through use of performance based provisions. During the mix-design process, contrac-tors have developed mixes that better balance mix production costs and level of construction effort to achieve good quality field placement. These cus-tom mixes result in a smoother ride and pavement with higher density, less tendency to segregate, less permeability, and increased durability. When the “ride special provision” is applied to a project, more attention to detail is required throughout the paving process. Use of a materials transfer vehicle, continu-ous feed of material, no stopping of the paver, and proper rolling techniques are examples of techniques employed to improve ride quality. The use of the ride special provision provides monetary incentives to the contractor and longer lasting pavements for the taxpayer (Richter, 2004).

With smoothness expressed as a profile index in mil-limeters/kilometer (mm/km), Kansas specifications require an average profile index of 475 mm/km or less per 0.1 km section as measured with a Califor-nia-type profilograph (Richter 2004). An exception is made for ramps and acceleration and deceleration lanes where a profile index of 630 mm/km or less is required. In addition, flexible pavement within each section having high points with deviations greater than 10 mm in a length of 7.5 meters must be corrected regardless of the profile index. These efforts seem to be working, as asphalt pavement with smoothness between 0 and 160 mm/km increased from about 5% in 1991 to roughly 55% by 2001 (Richter, 2004).

Pay adjustments are based on the average profile index determined for each section prior to any cor-rective work (such as grinding). If the contractor elects to remove and replace the sections or overlay pavement to meet the smoothness specification, pay adjustments are based on the average profile index obtained after replacement or overlay. Table 7 shows the schedule used to adjust payments for flexible pavement quality in Kansas. Although



some fluctuation has occurred from year to year, an increase in the percentage of pavements built with high levels of smoothness (0 to 160 mm/km for flexible pavements) has been noted (Richter, 2004). There are two basic types of construction warran-ties, materials and workmanship or performance. Materials and workmanship warranties address the quality of pavement through usually up to one year after construction while a performance warranty ad-dresses pavement quality at some point in the future. Performance warranties are typically referred to as warranty specifications when applied to pavement (WSDOT, 2002).Table 7: Kansas schedule for adjusted payments – flexible pavements

Average Profile Index(mm/km per lane

per 0.1 km section)

Contract Price Adjustment (per 0.1 km section per lane)

110 or less +$100.00111 to 160 +$50.00161 to 475 0.00*476 to 630 0.00*

*Correct to 475 mm/km (630 mm/km for ramps, acceleration and deceleration lanes (Source: Richter 2004)

Almost all flexible pavement is covered by a short duration (usually one year) materials and workman-

ship warranty. This type of warranty assigns risk to the contractor for following transportation agency specifications in regards to materials and workman ship. If a problem or defect is detected within the warranty period, the transportation agency uses some type of forensic analysis to determine the cause. If it is determined that specification non-compliance dur-ing construction caused the problem, the pavement is repaired at the contractor’s expense. If unexpected traffic volume or unknown site conditions caused the problem, the transportation agency assumes financial responsibility for the repair cost. This type of war-ranty is almost universal, rarely collected on, and is usually covered by sureties at no additional charge to the contractor. A performance warranty, however, assigns a greater portion of risk to the contractor. Throughout the warranty period, the transportation agency continues to monitor pavement performance.

MANAGEMENT OF LONG-TERM PAVEMENT PERFORMANCE THROUGH WARRANTIES

Warranties are one type of performance specifica-tion that has received more attention in recent years. When using warranty specifications, a transportation agency specifies parameters for pavement perfor-

Figure 1: States that have used pavement warranties (source: FHWA 2006)



mance only; the contractor must warrant that level of performance for the pavement over the speci-fied time period. This warranty period normally extends two to seven years for asphalt pavement, and some warranties have been written for periods up to twenty years for concrete pavement (FHWA, 2006). During the warranty period, any defects at-tributable to construction practices or materials must be repaired at the contractor’s expense. States that have used or are currently using pavement warran-ties are shown in Figure 1. Pavement performance below contractually defined limits attributable to construction methods or materials must be remedied at the contractor’s expense. Because the contractor assumes greater risk, he/she is usually allowed to control many additional aspects of the construction process (FHWA, 2006).

For specifying transportation agencies, warranties represent progress over traditional end-result speci-fications because warranties enumerate specific stan-dards for actual pavement performance rather than material characteristics that are only indicative of material performance. Table 8 shows an example of performance standards developed by the Indiana De-partment of Transportation. A warranty specification is more capable of aligning the sometimes compet-ing influences of economic incentives, innovation, customer requirements and pavement quality. This alignment, when achieved, allows market forces and economics, rather than construction specifications alone, to drive pavement quality (NCHRP, 2001).

Although warranty specifications are being used in other countries, most notably in Western Europe, they have been used only sparingly in the United States. There are several reasons for this. First, U.S. paving contractors have been very reluctant to as-sume greater risk. Second, the Federal Government places certain legal restrictions on warranty use. Third, performance testing requires further research so that methods can be proven accurate and test re-sults can be used to legally invoke warranty clauses. Finally, the surety industry may play the largest role

in deciding the extent to which performance based incentives will be adopted in the United States. Table 8: Indiana DOT pavement performance thresholds for a five year warranty specification

ParameterThreshold Value

(contractor must take action above this value)

IRI 2.1 m/km (133 inches/mile)Rut depth 9 mm (0.375 inches)

Surface Friction average of 35 but no single sec-tion < 25

Transverse Cracking Severity 2 (as defined by the Indiana DOT)

Longitudinal Cracking 5.5 m (18 ft.) per 152.5 m (500 ft.) section

(Source: Washington State DOT, 2002)

State laws commonly limit a transportation agency’s risk by requiring a contractor to be bonded. A bond-ing agency may or may not be willing to accept the risk associated with a two to seven year performance warranty. A surety is especially wary when the contractor is allowed little to no input to pavement design and no control over post-construction pave-ment use (WSDOT, 2002).

A few state transportation agencies have used both asphalt concrete and Portland cement concrete pave-ment warranties for many years. Under a pavement-warranty specification, quality is measured by the actual performance of the pavement as opposed to the properties of pavement materials and meth-ods of construction. Pavement warranties require the contractor to guarantee the post-construction performance of the pavement. The shifting of post-construction performance risk from a state highway agency to a contractor is perceived to reduce pre-mature pavement failure, reduce cost, and increase pavement quality (TRB, 2005). However, for most contractors to feel comfortable with assuming the greater risk associated with a pavement warranty, some type of monetary incentive must be provided.Change in asphalt pavement quality resulting from the use of pavement warranties has been mixed. The Wisconsin DOT indicated a significant quality



increase and overall cost reduction from use of five year performance warranties for asphalt concrete pavements (TRB, 2005). However, an evaluation of three year workmanship and materials warranties completed by the Colorado DOT showed no dis-cernible impact on pavement quality or cost (TRB, 2005).

DEVELOPMENT OF A PAVEMENT PERFORMANCE INCENTIVE SYSTEM

According to the National Cooperative Highway Research Program, performance-related standards should be based on attributes that are related to the actual performance of the product through a validat-ed quantitative model. Standards should incorporate sampling and testing procedures whose combined costs are consistent with the importance of the qual-ity benefit being sought, and make the contractor’s payment dependent on how close the product comes to the level of acceptable quality (Volokh, 1996).

Parameters used to measure the quality of long-term pavement performance must be understood by both construction personnel and the owner’s qual-ity control technicians. Guidelines with regard to which parameters should be evaluated for inclusion in the NDOR’s performance-based incentive pro-gram specified that additional testing should not be required. Parameters should be comprised of one or more performance indicators currently measured by the NDOR. The NDOR measures a variety of pavement performance indicators including vari-ous cracking indices, International Roughness Index (IRI), Present Serviceability Index (PSI), Nebraska Serviceability Index (NSI), and rutting. Flushing

is also being documented when a problem with a particular segment is evident. NDOR further speci-fied that selected parameters were to be correlated to reflect an acceptable level of pavement performance after the appropriate interval of time. The research team initially proposed two different sets of param-eters, one for conventional flexible pavement and the other for Superpave pavement illustrated in Table 9. Parameters in Table 9 were subsequently discussed with representatives from several regional paving contractors, including Dobson Brothers, Hawkins Construction and Werner Construction at the Univer-sity of Nebraska.

Comments from the contractor representatives indi-cated that four years was an excessive period of time before deciding whether or not pavement met qual-ity standards. Most believed that even long-term performance of asphalt pavement could be reliably predicted based upon specific parameters measured after two or three years. Based on the contractor’s extensive experience with numerous flexible pave-ment projects, variation in asphalt binder content had little to no effect on medium-term pavement performance. Contractor representatives agreed that a medium-term monetary incentive would have to be greater than any initial quality incentive to adequate-ly compensate a contractor for the extended period of time until the incentive was received.

Researchers next met with NDOR representatives from the Materials and Research and the Construc-tion Division. NDOR personnel informed the researchers that all asphalt pavement contracted (by the NDOR) must now meet Superpave specifica-tions, so traditional asphalt could be deleted from

Table 9: Initial performance incentives proposed by the research teamEligibility Criteria Incentive Parameter Payment - % of Contract

Asphalt (Traditional)

Profile Index ≤ 8 inches/mile Variation of asphalt binder

content from design content (%) ≤ 0.25

IRI ≤ 1.00 mm/m @ 2 yrs IRI ≤ 1.2 mm/m @ 4 yrs

Rutting ≤ 4 mm @ 2 yrs Rutting ≤ 4 mm @ 4 yrs

2.5 % @ 2 yrs 2.5 % @ 4 yrs

2.5 % @ 2 yrs 2.5 % @ 4 yrs

Asphalt (Superpave)

Dynamic Shear(Original) ≥ 0.89 KPa Dynamic Shear (Residue) ≥ 1.95 KPa

Creep Stiffness ≤ 315 MPa Creep Slope ≥ 0.291 ElasticRecovery ≥ 54 %

IRI ≤ 1.00 mm/m @ 2 yrs IRI ≤ 1.2 mm/m @ 4 yrs

Rutting ≤ 4 mm @ 2 yrs Rutting ≤ 4 mm @ 4 yrs

2.5 % @ 2 yrs 2.5 % @ 4 yrs

2.5 % @ 2 yrs 2.5 % @ 4 yrs



consideration. A discussion ensued on the proposed standards of quality, specifically which indicators were most appropriate for measuring medium-term pavement performance. Since IRI decreases as a result of surface wear as flexible pavement ages, IRI was believed to be more indicative of traffic volume than of medium-term pavement performance. IRI was subsequently eliminated from consideration as an indicator for medium-term pavement performance.

The possibility of using pavement cracking as a performance indicator was then discussed. The con-sensus reached was that control of cracking is often beyond the ability of a contractor to influence, at least for many applications of flexible pavement. Cracking is sometimes caused by the quality of materials used or by laydown procedures, but it sometimes appears in situations where quality control of both materials and construction procedures were excellent. Crack-ing was thus eliminated from consideration for use as an indicator when measuring medium-term pavement performance.

Flushing was the next parameter evaluated. Flush-ing results from excessive binder in the mix, low levels of dust in the aggregate, excessive pavement temperatures relative to performance graded binder specifications or from some combination of these fac-tors. Standards for flushing have not been precisely quantified by the NDOR, but even a small quantity of flushing becomes very evident when driving over a paved surface. The NDOR has no currently pub-lished standards specifying what levels of flushing are acceptable and what levels are considered exces-sive. Flushing in excess of 20% of the pavement surface area was deemed excessive by a majority of the Materials and Research Division personnel. The NDOR is currently considering adding specifications for flushing to their criteria for flexible pavement contracting.

Rutting was the only proposed criteria thought to be measurable by well-defined quantitative standards under a performance based incentive program. Six millimeters was considered excessive for a rutting limit and two years was considered insufficient time to measure performance. Consensus was that a minimum of three years and maximum rutting of four millimeters were the minimum acceptable standards

for any pay incentive. Profilograph Pay Factor (PPF) and Material Pay Factor (MPF) were thought to rep-resent good initial composite estimates of pavement quality. Paving projects with below average PPF and MPF would not normally be eligible for construction incentives, so these factors could be used as eligibil-ity criteria for any performance based incentive. The performance indicators shown in Table 9 were condensed to reflect flexible pavement with a stan-dard of rutting < 4 mm measured at three years. Flushing of less than 20% of the pavement surface area was included in the final recommendation as well. PPF and MPF > 100% were listed as eligibility criteria. The proposed incentive was based upon the NDOR’s current practice of paying for the quantity of asphaltic concrete (in tons or Mg) placed as surface layers only.

ANALYSIS

Researchers then sought to investigate whether recent projects awarded construction incentives showed ac-ceptable or better standards for medium-term perfor-mance. The NDOR was asked to provide data for flexible pavement projects in excess of five miles in length that had been constructed during the past three years, where quality incentives had been paid for smoothness and/or materials and workmanship upon completion of construction. Three years provided post-construction time for pavement performance data to accumulate while length in excess of five miles indicated a significant paving project. Rational for the quality incentive specification rested upon the assumption that a project which failed to earn an incentive for quality of construction would probably not exhibit a high level of medium-term performance.

Table 10 shows cost information for three asphalt paving projects approximately three years old that received construction materials and workmanship incentives from the NDOR for pavement qual-ity. All projects involved pavement placed using Superpave specifications. Three different types of construction incentives were paid for each project. Table 11 shows rutting measured for these three projects over the first three years of their lifespan.



Based upon the meeting with contractor representa-tives, meetings with NDOR’s Materials and Research Division and Construction Division personnel, and an analysis of information in Tables 10 and 11, a revised proposal for a performance-based incentive program for flexible pavement was created.

The proposed system is summarized in Table 12. Re-searchers had initially recommended a 5% incentive based upon total cost of asphalt paving materials for the project. However, the incentive paid for quality at

completion of construction varied for the three proj-ects analyzed from 3.7% to 5.8%. A future incentive less than the initial construction quality incentive would be of marginal interest to most contractors, so a 6% medium-term performance incentive was the final recommended minimum.

Six percent is simply a recommendation. The actual percentage can be adjusted upward or downward by the NDOR until the level of interest (displayed by contrac-tors) is sufficient to satisfy the NDOR’s needs.

Table 10: Projects where construction quality incentives were paidControl Number Smoothness Incen-

tive Pay FactorAdditional Incentive

Pay FactorQuantity Incentive Paid

Smoothness Incentive - Performance Graded (PG) Binder60937 Not Available $5.94 776.92 Mg $4,614.9031345 104.06% $6.46 522.08 Tons $3,372.6460893 100.75% $2.04 248.298 Mg $506.53

Smoothness Incentive – Asphaltic Concrete60937 Not Available $0.84980 17,544.60 Mg $14,909.4031345 104.06% $0.75000 12,733.75 Tons $9,553.3160893 100.74% $0.14800 4,281 Mg $633.59

Superpave Quality Incentive (Air Voids)60937 Not Available $0.90 51,729.62 Mg $46,566.6631345 103.54% $0.66 25,229.95 Tons $16,651.7760893 Not Available $0.68 25,807.04 Mg $17,548.79

Total Construction Quality Incentive Paid60937 $66,090.9631345 $29,577.7260893 $18,688.91

Table 11: Analysis of pavement performance over three years

Hwy Beg Ref Post

End Ref Post

Cont Num

Date

Comp

Avg Rut 2003 (mm)

Avg Rut 2004 (mm)

Avg Rut 2005 (mm)

Avg Rut 2006 (mm)

IRI 2006 Crack Index

2 258.04 270.32 60937 2002 0.75 1.43 1.6 2.4 0.8 1.730 114.31 124.31 60893 2003 0.3 3.39 3.58 4.6 0.9 0

275 31.91 39.31 31345 2002 3.37 4.87 ND 4.5 1 5.5

Table 12: Revised performance incentive recommended for flexible pavementsEligibility Criteria Incentive Standards When Measured Payment

Profilograph Pay Factor > 100%

Materials Pay Factor

> 100%

Rutting < 4 mm

Flushing < 20%

of paved surface

3 years

3 years

6 % of asphalt

materials cost

as determined

by the NDOR



CONCLUSIONS AND RECOMMENDATIONS

Results of this research can be used by the NDOR to provide contractors with incentives to more closely control the quality of materials used in mixes and methods of construction for flexible pavement. Percentages of pay for incentives or time periods associated with a specific incentive can be established at the level of performance needed and adjusted to encourage the desired level of contractor participation.

This performance-based incentive system was developed to align the objectives of paving contractors more closely with the objectives of the NDOR. Under this system, both the NDOR and pavement contractors have an incentive to provide pavement that meets defined specifications upon completion of construction and continues to perform well enough to meet established standards for a period of time afterward. This system will highlight to contractors the need to use quality materials and methods and will also provide a positive financial incentive in later years for contractors who construct quality pavement.

A quality incentive program of this type based upon medium-term pavement performance has the potential to become a nation-wide trend. Many state transportation agencies are experimenting with pavement warranties, best value contracts and performance based contracting procedures in an attempt to procure higher quality pave-ment. A quality incentive program of the type proposed has the potential to provide most of the benefits of these programs at less cost and certainly with fewer legal entanglements.

Similarly, four millimeters of rutting during the first three years of pavement life is merely the researcher’s recom-mendation. Both depth of rutting and/or the time period can be adjusted upward or downward as needs or condi-tions change.

Table 13 illustrates how the proposed performance based incentive would have applied to the three projects ana-lyzed in Tables 10 and 11. Two of the projects would have been ineligible for the proposed incentive as the mea-sured value of rutting exceeded four millimeters at three years into the pavement’s lifespan. The proposed perfor-mance-based incentive was not recommended for application to all projects but only to those projects where the NDOR wishes the resulting pavement to be of superior quality. These situations might include roads where the volume of traffic is sufficiently high to make repair and/or rehabilitation exceedingly dif-ficult or costly.

Table 13: Proposed performance incentive applied to three projectsHighway

Number

Control Number Total Cost of Asphalt

Distance

(Miles)

Average Rutting at 3 Yrs (mm)

Proposed

Performance Incentives

2 60937 $1,150,566 12.28 1.6 $69,03430 60893 $505,205 10 4.6 $0

275 31345 $494,595 7.4 > 4.5 $0



REFERENCES

Carpenter, B. Fekpe,, E. and Gopalakrishna, D. (2003). Performance-Based Contracting for the Highway Construction Industry. Retrieved Febru-ary 19, 2007 from www.ncppp.org/resources/papers.batellereport.pdf.

Federal Highway Administration (2006). Back-ground for Pavement Warranties. Retrieved January 3, 2007 from www.fhwa.dot.gov/Pavement/war-ranty/backgrnd.doc.

National Cooperative Highway Research Program (2001). Synthesis 300. Performance Measures for Research, Development, and Technology Programs. Transportation Research Board, National Research Council, Washington, D.C.

Nebraska Department of Roads (2004). Superpave Asphaltic Concrete, Manual S10-7, Section 1028 of the Standard Specification (Revised 3-22-04), pp. 1-24.

Nebraska Department of Roads (2007). Supple-mental and Standard Specifications for Highway Construction. Retrieved March 1, 2007 from http://www.dor.state.ne.us/ref-man/specbook-2007.pdf.

Richter, C.A. (2004). The Case for Performance Standards [Electronic version]. Public Roads. 67(6) 18-22. Retrieved January 10, 2007 from the Aca-demic Search Premier Database.

Transportation Research Board (2005). Guidelines for the Use of Pavement Warranties. Retrieved March 5, 2007 from http://www.trb.org/TRBNet/ Project Display.asp?ProjectID=288.

Volokh, A. (1996). Recycling and Deregulation Opportunities for Market Development. Resource Recycling. Retrieved February 4, 2007 from http://volokh.com/sasha/resrec.html.

Washington State Department of Transportation (2002). Subparagraph 3.4.1 Warranty under Section 3 Specifications. Retrieved January 20, 2007 from http://training.ce.washington.edu/WSDOT/Mod-ules/08_specifications_qa/08-3_body.htm.

19

A Conceptual Framework: Teaching Reality-Based Construction Project Management via Constrained Virtual

Construction Environment

Dr. Borinara Park, Dr. Ronald Meier

ABSTRACT: In the current virtual construction environment (VCE), users simulate the building process without considering real-life project constraints related to materials, equipment, and administrative require-ments. College students, in particular, tend to focus on only the construction aspect of a project when simulating the building process, instead of realizing the management aspect of the project. To remedy the issue, the authors present a conceptual framework of a constrained virtual construction environment (CVCE), where student users have to be involved in a communication simulation process related to acquiring the right resources and meet-ing administrative/regulatory requirements while performing the graphical building simulation. By imposing the communication simulation outcomes on the unconstrained VCE, only the legitimate activities whose resource and administrative requirements have been met are allowed to be simulated in the CVCE. Therefore, students learn how different project participants impact their own performances, as well as see how project information and document management play a critical role in the success of construction projects. As a result, the proposed system enhances students’ learning as it allows the simulation of real-world construction management processes.

Key Words:Virtual reality, virtual construction environment, con-strained simulation

BACKGROUND

MULTI-DIMENSIONAL CAPABILITIES OF VIRTUAL CONSTRUCTION FOR CONSTRUCTION EDUCATION

Construction-related programs such as architecture, civil engineering, and construction management face a significant challenge of providing students knowl-edge that works in the real world (Tatum, 1987). It is, therefore, not surprising that construction management

programs are allocating more time in their curricula to provide students opportunities that simulate real world experiences (Gibbons, Limoges, Nowotny, Schwartzman, Scott, Trow, 1994). As a way of meet-ing this growing need, a few pedagogical and and technological innovations have been designed, tested, implemented, and accepted successfully into construc-tion management programs. These innovations include internships, multimedia-based learning, service-learn-ing projects, simulation, and games (Park, Chan, and Ingawale-Verma, 2003; Senior, 1998). One common-ality of these innovative teaching methods is that they all strive to address deep-seated troubles of typical construction-related programs such as 1) difficulty in


DR. BORINARA PARK is an Associate Professor and Program Coordinator for Construction Management at Illinois State University. His re-search focuses on various Building Information Modeling (BIM) topics such as evolution of BIM technology, construction process visualizationtechnology, information-constrained virtual construction simulation, and equipment simulators. Dr. Park has a BE in Civil Engineering (1993) and a ME in Geotechnical Engineering (1995) from Korea University, in Seoul, Korea. He earned his Ph.D. in Environmental Design and Plan-ning in 2002 from Virginia Polytechnic Institute and State University, Blacksburg, Virginia.

DR. RON MEIER joined the faculty at Illinois State University as a Professor of Technology Management in August 1994. His previous academic experience includes Professor of Industrial Education & Technology and Manufacturing Program Leader at Iowa State University. In addition to a successful academic career, Ron brings over 20 years of work experience in technology management and research to the role and continues to be active as a trainer, consultant and board member for a number of organizations. He specializes in project, quality and risk man-agement issues – more specifically the integration of tools and techniques to align organizational strategy and operational execution.


20 A Conceptual Framework: Teaching Reality-Based Construction Project Management via Constrained Virtual Construction Environment


providing realistic, context-rich education to stu-dents; and 2) limited resources to bring the real world experience into an education setting (Sawhney and Mund, 1998).

One of the simulation techniques, virtual construc-tion technology, has gained wide recognition as an effective management tool (Schwegler, 2003). This acceptance is mainly due to its capability to allow users to present complex construction information in an easy-to-understand 3D graphical format. Starting from the CAD design technology, 3D virtual con-

struction has evolved as a multi-dimensional simula-tion environment (Park and Wakefield, 2003). A 3D graphical environment has now been transformed into a project database in which project information is generated, stored, and shared as shown in Figure 1. Recently Building Information Modeling (BIM) has been widely accepted by the construction indus-try as a visual project management tool (Goldberg, 2005). 3D models function as a graphical interface to a project database by providing efficient access to construction project data that is easily integrated into the project work plan.

Figure 1: 3D Models as a Hub for Project Database


Dr. Borinara Park, Dr. Ronald Meier 21


The 4D scheduling technique has emerged as a con-struction process visualization tool which allows the presentation of 3D building models in a time-lapsed fashion as shown in Figure 2 (Alianza, 2003; Hay-maker and Fischer, 2001). This advancement allows users to literally construct their buildings without physically building them on job sites. Reported ben-efits of 4D scheduling include: construction schedule verification, alternative schedule comparison, trade coordination, temporary structure design, and active hazard anticipation and safety improvements (Brawn and Sloan, 2003; Ray and Reed, 2003).

Real-time interactivity is another extension in the virtual construction simulation area (Anonymous, 1996; Lipman and Reed, 2000; Phair, 1996). Us-ers interact with graphical models in a simulation system, in which real-time interactivity and control-lability is given to users to allow the simulation of realistic construction operations. Through a number of trial operations using the simulation system, us-

ers can plan better construction processes. Figure 3 shows a crane simulation system where users try crane operations in a virtual site resembling real working conditions so that they can identify the proper size of the crane and its best location in terms of reducing unnecessary operations and wait time (Park, 1998). Also, users can learn safe operating practices under realistic environmental conditions. One of the differences between a real world and a graphical world is that 3D graphical models do not follow the laws of physics. In any visualization ap-plication, where materials and machines are

involved, no inclusion of physics is likely to give unrealistic feedback to simulation users (Park and Wakefield, 2002; Wakefield and O’Brien, 1994). In order to provide physically valid responses, in a simulation system, physics components have been added (Park, Perumpral, and Wakefield, 2001; Wake-field, O’Brien, and Perng, 1996).

FIGURE 2: 4D (3D + Time)

FIGURE 3: Interactive Capability in the Virtual Construction Simulation Environment



Figure 4 shows an excavator simulator, the represen-tation is close to the real machine not only graphi-cally but also physically based on the machine’s hydraulic system, thereby providing a physically valid behavior of the machine to the simulator users.

Figure 4: Physics-Based Simulation System

DRAWBACK OF THE CURRENT VIRTUAL CONSTRUCTION USED IN CONSTRUCTION EDUCATION

Today’s existing 3D virtual construction environ-ments (VCE) provide substantial benefits from several varying educational and pragmatic perspec-tives. However, there is a critical drawback to the current virtual construction technology in terms of its validity as an educational tool to be used in construction project management training (Park and Meier, 2007). In existing VCEs, the building process is not realistically controlled because there are no real-life project constraints. In reality, resources such as materials and equipment should be provided and administrative requirements met in order for specific construction operations to be executed. However, in the reviewed VCEs, student users can complete any virtual construction job without acquiring the appro-priate resources and permissions required to perform the work. The authors have identified four construc-tion pre-requisite conditions that must be addressed by simulation users before building graphically in the VCE.

Construction Pre-requisite Condition #1: Resource Requirements

What are the resource requirements to properly ex-ecute the job? What are the proper machines, materi-als, and people that need to be acquired?

Construction Pre-requisite Condition #2: Administrative/Regulatory Requirements

What are the administrative/ regulatory requirements for this job?

What permits, inspections, submittals, etc must be processed and secured?

Construction Pre-requisite Condition #3: Communication Plan

Who are the stakeholders that need to be commu-nicated with? Has a communications matrix been developed? What information/documents need to be distributed for the above requirements to be satis-fied?

Construction Pre-requisite Condition #4: Project Docu-ment Exchange Process

How will information/documents be disseminated? Who will receive these documents? How often will the stakeholders be informed?

Let us consider a precast concrete wall panel con-struction job for a building project. Student simula-tion users should address these conditions and learn how to handle these before they start laying the precast wall panels graphically. In a real-life situa-tion, if a project manager had not acquired a crane or a proper crew, the precast concrete wall construc-tion could not be done. If each had not acquired the required submittal approval or permits, the precast construction work is not supposed to proceed. The project manager should be able to initiate and com-plete the purchase order process with a precast con-crete supplier, and should be engaged in the submit-tal approval process with an architect. If any of the steps described above are not properly addressed, the precast wall construction job cannot be executed.

This aspect of the project management process is not simulated in existing VCEs. Existing VCEs al-low student users to be exposed to only the visible construction aspect of a project, not exposing them



to the invisible communication aspect of the proj-ect management. Therefore, student users are not given an opportunity to learn that the building is the outcome of many exchanges and communications of project information and data among the stakeholders involved in the project. This data/information must be generated, transmitted, reviewed, and approved, through pre-established communication channels among all project stakeholders.

The key concept of the VCEs is to provide users with the capability of building virtually any facil-ity so that they can try as many different processes, operations, options and sequences as possible. However, this unlimited capability without any realistic restrictions/constraints critically hampers the realism of any simulated construction in an education set-ting. Student users tend to perform work that is not possible in reality by interacting with overly simpli-fied and unconstrained construction processes. It is, therefore, likely that by building their virtual 3D graphical construction projects out of reality, stu-dents do not learn construction management the right way.

PROPOSED VIRTUAL CONSTRUCTION ENVIRONMENT

CONCEPTUAL FRAMEWORK

As a way of addressing this unconstrained virtual construction operation in existing VCEs, the authors propose a Constrained Virtual Construction Environ-ment (CVCE). In this new simulation environment student users perform virtual construction simulation by addressing the aforementioned construction pre-requisite conditions.

Figure 5 summarizes essential components of the proposed Constrained Virtual Construction Envi-ronment (CVCE). The CVCE is comprised of two simulation environments. The first simulation envi-ronment component is a web-based project collabo-ration environment (PCE) such as Autodesk’s Con-structware or Meridian’s Prolog. This commercially available technology enables multiple project partici-pants to generate, exchange, and share construction project information in real-time through a central

3D Virtual Construction Environment

(VCE)

Ex) Purchase

Orders

Ex) Change Orders

Ex) Shop Drawing

Approvals Contractors

Engineers

Construction Managers

ArchitectsOwner

Subcontractors

Ex) RFI

Suppliers

Web-Based Project

Collaboration Environment

(PCE)

Ex) Payment Approval

Ex) Safety Inspection

Ex) Equipment Acquisition Ex) Permit

Acquisition

Figure 5: Essential Components of Constrained Virtual Construction Environment (CVCE)



web server (Leung and Chan, 2003). The second simulation environment component is a conventional virtual construction environment (VCE), which is defined in this paper as a commercially available 3D graphical computer desktop environment, such as Navisworks’ JetStream, or Common Point’s Proj-ect 4D, in which users can interactively assemble 3D graphical building components for construction visualization.

Initially, students are given a construction job which they have to complete by collaborating with each other. The student users log into the PCE by assum-ing roles of general contractors (GC), architects/ engineers, agencies, owners, and suppliers. The GC project manager role-playing student starts by asking questions related to the construction pre-requisite conditions. This student identifies the resource requirements, administrative/ regulatory require-ments, and communication requirements. He/she then initiates the construction management process within the PCE by exchanging appropriate project documents with other students who role-play other

project stakeholders such as architects, vendors, and so on. Some examples of these exchanged project documents are purchase orders, delivery confirma-tion, submittals and approvals, permits, and so on. This communication-oriented simulation takes place first in the PCE before the graphical simulation of the construction operation can be started in the VCE.

If the required project information is not distributed among the stakeholders the lack of proper commu-nications will be flagged as a constraint just as in a real life scenario. For example, if a student project manager has not acquired panels for pre-cast con-crete wall construction through purchase orders and delivery confirmation with a vendor (another role-player) in the PCE or has not secured an approval of the pre-cast panel submittal from the architect (an-other role-player), he/she is not allowed to execute the work in the VCE simply because they did not execute the required project document/ information distribution processes. For this reason any unfin-ished communication process will pose constraint conditions onto the VCE. These constraint condi-

Figure 6: Conceptual Framework of the CVCE




tions provide a basis for the simulation system to accept only the activities that are legitimate to be graphically simulated in the VCE. In order to accept only the legitimate construction operations in the VCE, a new component, called Constraint Management Layer (CML) is introduced as the last part of the proposed simulation environ-ment. The CML layer works between the PCE and the VCE as depicted in Figure 6. The communication requirements that are simulated in the PCE are veri-fied by the CML as to whether or not the required resources and administrative/regulatory requirements have been met. This allows only the approved con-struction activities to be graphically simulated in the VCE. Conceptually, therefore, the whole simulation process starts in the PCE, is then evaluated by the CML, and ends with the graphical simulation in the VCE as presented in Figure 6.

FUNCTIONS AND INTERACTIONS IN THE PROPOSED ENVIRONMENT

This section of the paper describes the specific inner workings in terms of how the three main environ ment components interact with each other during a simulation. In Figure 7, the CVCE depicts three main environment components, the PCE, the CML, and the VCE. Each component’s inner workings and their respective relationships with the other layers are described from the following perspectives: (a) who are the main entities performing the actions required in a specific environment component? (Performing Entities); (b) what prerequisite data is needed for the entities to perform their respective tasks? (Prerequisite Data); (c) what specific actions are taken by the entity? (Actions); and (d) what are the outputs from these actions? (Output).

The simulation process is initiated in the web-based PCE where the performing entities (student role players) act as project participants collaborating with each other. Existing project information such as plans and specifications are used as prerequisite data, from which students identify various requirements such as a bill of materials, purchase orders, permit acquisitions, submittal approvals, and so on. Based on their roles, the students perform specific actions and complete communication requirements for the project by generating and sharing documents with

other players as well as having to respond to the doc-uments created by other role players. The outcome of these collaborative actions is a set of exchanged-documents saved on the PCE web server.

Based on these communication transactions, the contractor role-playing student builds (or assembles) the 3D graphical building models in the VCE. When a student graphically builds the model, they must understand that only those work items whose com-munication requirements have been completed can be executed. If not completed in their entirety those unfinished communication requirements become constraint conditions by which their corresponding virtual construction actions are not allowed to be executed. When virtual building operations are de-tected, it needs to be checked if these operations are do-able in terms of the role-players having met the necessary resources and administrative/regulatory requirements based on the project documents and information exchanged in the PCE.

This verification or checking operation is done in the CML by the supervisor(s), who retrieve and evalu-ate all the exchanged documents from the PCE. The supervisors such as class instructors or graduate assistants perform the following tasks: 1) identify the scope of work performed by the student contrac-tor role player in the VCE; 2) generate resource and administrative/ regulatory requirement conditions related to the work; 3) retrieve the related docu-ments from the PCE; and 4) apply the activity or job requirement conditions by comparing them to the retrieved communication documents so that only the communication-completed work is allowed to display in the VCE. Therefore, only when all the requirement constraining conditions are met will the students’ virtual construction activity or job be executed in the VCE.

APPLICATION: EXAMPLE OF THE PROPOSED SIMULATION ENVIRONMENT

In this section, a precast concrete wall panel con-struction for an office building is used as an example to provide a step-by-step simulation process to dem-onstrate how the proposed simulation learning envi-ronment is used in a typical classroom setting. The following scenario depicts the specific activities that



Figure 7: Functions and Interactions among the CB-VCE Simulation Components



both the instructor and students follow to achieve the successful execution and completion of the precast concrete wall construction simulation. The descrip-tion is also captured graphically in Figure 8 on the following page.

[Step 1] Construction Project VisualizationAs a first step, based on the project drawings and specifications given by the instructor, the students visualize what they build. At this stage, they cre-ate 3D models of the building using a 3D modeling application such as Autodesk Revit, SketchUp, or Autodesk VIZ.

[Step 2] Construction Sequence VisualizationThe students then visualize how they build the construction project. Specific construction steps and alternatives are discussed, which leads to the graphi-cal simulation of the construction in the VCE. At this step, they are required to address the resource requirements involved in the precast wall panel construction. This specific construction requires resourc-es such as concrete pre-cast wall panels, a crane, a crew, and so on.

[Step 3] Identification of Communication Matrix To acquire the resources such as the precast wall panels the students identify the stakeholders involved in the acquisition process. For the panel acquisition, the parties should include a contractor, a precast panel vendor, and an architect.

[Step 4] Communication Simulation InitiationBased on the identified parties in the panel acquisi-tion process, each student takes a role of each party and logs in to the web-based project collaboration environment (PCE) to simulate the communication process. A snap shot of a commercially available PCE product (Autodesk’s Constructware) is shown in Figure 8.

[Step 5] Project Document/ Information ExchangeNext, the contractor role-playing student needs to identify administrative requirements. The submittal submission/approval is an example of such require-ments related to the precast concrete panel acquisi-

tion. Now that he or she has identified the resource and administrative requirements and the commu-nication requirements for the job, the contractor role-player determines the communication routes, how the necessary documents and information are exchanged, and which stakeholders must be involved in each exchange. This results in the communica-tion flowchart as diagramed in Step 5 of Figure 8 (Mincks and Johnston, 1997).

1. The vendor transmits a shop drawing submittal of the precast panels to the contractor.2. The contractor reviews the submittals to approve or deny.3. Once approved, the constructor transmits the sub-mittal to the architect for his/ her approval.4. The architect approves or denies the submittal and notifies the contractor.5. If approved, the constructor grants the fabrication of the precast concrete panels for the vendor.6. The manufacturer confirms when it will be deliv-ered to the site.

Based on the communication flow, the contractor role-playing student exchanges the appropriate docu-ments with the architect and vendor role-playing students in the PCE.

[Step 6] Communication Completion ValidationThese exchanged documents are saved in the web server, which can be accessed by the class supervi-sor who acts as the constraint management layer and oversees the simulation operation. His or her role is to make sure all virtual construction opera-tions performed by the contractor role-player have been documented and approved by the appropriate stakeholder(s). When the student has assembled a series of wall panels in the VCE as shown in Figure 8, the supervisor validates the virtual construction by checking to verify that the contractor role-player has completed all the submittal processes with appro-priate stakeholders. If this student has not finished the submittal process (for example, not getting the architect’s approval, coded “BB” in Step 5 and 6 in Figure 6), the corresponding precast panel wall virtual construction operation is nullified.



= [Step 5] Project Document/ Information Exchange

Contractor awards contract

to Vendor

Vendor transmits submittal to Contractor

Contractor reviews submittals

Does submittal meet contract requirements?

No

Yes

Contractor transmits submittal to Architect

Architect reviews submittals

Return submittals to Vendor

Does submittal meet contract requirements?

No

Yes

Approve and return to Contractor

Return to Vendor/Vendor Fabricates &

Deliver

AB

AA

BB

C

[Step 1] Construction Project Visualization

[Step 2] Construction Sequence Visualization

3D VIRTUAL CONSTRUCTION ENVIRONMENT (VCE)

[Step 3] Communication

Matrix Identification

[Step 4] Communication

Simulation Initiation

Contractor

Vendor Architect

WEB-BASED PROJECT COLLABORATION ENVIRONMENT (PCE)

[Step 6] Communication Completion Validation

A

AA

B

BB

C

None

Submittals Submitted?

Submittals Approved?

CONSTRAINT MANAGEMENT LAYER(CML)

CLOSE TO ACTUAL CONSTRUCTION PROCESS(CVCE)

[Step 7] Constrained Simulation Completion

Figure 8: Class Module Example of CVCE Precast Wall Construction Submittal




[Step 7] Constrained Simulation CompletionThe simulation participants repeat the same process to meet other resource and administrative/regulatory requirements. Only when the contractor role-player finishes all the necessary information/ document exchange process per the communication flowchart will the corresponding virtual construction activity be completed.

CONCLUSION

This paper presented a conceptual framework to remedy several problems with existing virtual reality-based construction simulation practices. The current virtual construction environment does not function as a valid educational tool to be used for training the next generation of construction project managers. The building process is not realistically simulated because there are no real-life project con-straints in the current practices.

To make the building simulation a realistic experi-ence, four construction pre-requisite conditions were identified: 1) resource requirements; 2) administra-tive/regulatory requirements; 3) communication plan; and 4) project document exchange process. In a real life setting, these represent an essential aspect of construction project management that needs to be handled through proper communication and infor-mation distribution practices. To realize this com-munication process, a web-based project collabora-tion environment (PCE) was used in the proposed simulation setting. In this environment, users simu-late how the project resource and administrative/ regulatory requirements are handled. Based on this communication simulation, an additional compo-nent, the constraint management layer (CML) was added to provide realistic construction conditions or constraints to the graphical simulation environment (VCE) to allow only the approved virtual construc-tion activities to proceed. This CVCE enhances the educational experience and results in the recognition of the need for better communication because stu-dent role-players immediately see the results of their collaborative outcomes. Specifically, each student gets an opportunity to visually see how different

project participants impact their own performances and how project information and documents play a critical role in the success of construction projects. As a result, the CVCE becomes a tool for better learning and understanding of real construction man-agement processes.

REFERENCES

Alianza M. (2003). Xdcad in facility life cycle. Paper presented at the Automatic Data Collection in Construction (ADCIC ) 2003, Las Vegas, NV.

Anonymous. (1996). Caterpillar at the leading edge of earthmoving technology. Mining Magazine, 26-31.

Brawn R. and Sloan E. (2003). 3d+d - expanding the electronic boundaries of construction management. Paper presented at the Automatic Data Collection in Construction (ADCIC ) 2003, Las Vegas, NV.

Gibbons M., Limoges C., Nowotny H., Schwartz-man S., Scott P. and Trow M. (1994). The new pro-ducttion of knowledge: the dynamics of science and research in contemporary society. London.

Goldberg H. E. (2005). Aec from the ground up: the strengths of bim. Cadalyst, from http://aec.cadalyst.com/aec/article/articleDetail.jsp?id=201190andsk=anddate=andpageID=3Haymaker J. and Fischer M. (2001). Challenges and benefits of 4d modeling on the walt disney concert hall project (no. Cife working paper #64): Center for Integrated Facility Engineering.

Leung N. and Chan S. (2003). Meta-data-based collaboration in construction project management. Paper presented at the Proceedings Of the 4th Joint International Symposium On Information Technol-ogy In Civil Engineering, Nashville, Tennessee.

Lipman R. and Reed K. (2000). Using vrml in construction industry applications. Paper presented at the Web 3D VRML 2000 Conference, Monterey, California.



Mincks W. R. and Johnston H. (1997). Construction jobsite management. Albany, NY: Delmar.

Park B. (1998). Simcrane: graphical crane simulator. Blacksburg, VA: Virginia Polytechnic Institute and State University.

Park B. and Meier R. (2007). Reality-based construc-tion project management: a constraint-based 4d simu-lation environment. Journal of Industrial Technology, 23(1).

Park B., Perumpral J. and Wakefield R. (2001, De-cember 3 ~ 5). Development of an excavator digging model for a virtual reality type excavator simulator. Paper presented at the 6th Asia-Pacific ISTVS Confer-ence and 22nd Annual Meeting of the Japanese Soci-ety for Terramechanics, Bangkok, Thailand.

Park B. and Wakefield R. (2002, November 19 ~ 21). Virtual reality excavator simulation system: calcula-tion methodology and mathematical models. Paper presented at the eSM@RT 2002: The European Con-ference on Information and Communication Tech-nology Advances and Innovation in the Knowledge Society, Salford, United Kingdom.

Park B. and Wakefield R. (2003). Evolution of visu-alization technology in construction: current practices and future implications, and their genuine role. Paper presented at the Third International Conference on Information Systems in Engineering and Construction, Florida.

Park M., Chan S. L. and Ingawale-Verma Y. (2003). Three success factors for simulation based construc-tion education. Journal of Construction Education, 8(2), 101-114.

Phair M. (1996). Construction workers are game for virtual equipment. ENR, 30-33.

Ray J. and Reed D. (2003). Using 4d technology on construction projects - two case studies. Paper present-ed at the Automatic Data Collection in Construction (ADCIC ) 2003, Las Vegas, NV.

Sawhney A. and Mund A. (1998). Simulation based construction management learning system. Paper presented at the Proceedings of the Winter Simulation Conference.

Schwegler B. (2003). Productivity in construction: a challenge for the 21st century. Paper presented at the Automatic Data Collection in Construction (ADCIC ) 2003, Las Vegas, NV.

Senior B. (1998). Infusing practical components into construction education. Journal of Construction Edu-cation, 3(2), 92-10.

Tatum C. B. (1987). Balancing engineering and man-agement in construction education. Journal of Con-struction Engineering and Management, 113(2).Wake-field R. and O’Brien J. (1994). A “virtual reality” type simulation system for construction automation system development. Paper presented at the Proceedings of the 11th International Symposium on Automation and Robotics in Construction, Brighton, UK.

Wakefield R., O’Brien J. and Perng J. (1996). Devel-opment and application of real-time 3d simulators in construction training. Paper presented at the Interna-tional Conference on Construction Training, Hong Kong.

31

Construction Safety: A Lean Construction Research Agenda

Abdelhamid, T.S., Mitropoulos, P., Schafer, D. and Howell, G.A.

ABSTRACT: Occupational accidents are unquestionably wasteful and non-value adding events in any system of pro-duction. Safeguarding construction workers from occupational hazards, whether arising from traumatic, ergonomic, and/or exposure accidents, is part and parcel of the lean construction ideal of waste elimination. Similar to other wasteful and non-value adding events that result in unreliable workflow such as late delivery of material and equipment, design errors, change orders, equipment breakdowns, and environmental effects, occupational accidents also result in the same. When left uncontrolled, these factors create havoc on any construction project resulting in either barely meeting the numbers or suffering devastating losses. A basic maxim in Lean Construction is that achieving reliable workflow is possible only when sources of variation are controlled. Unfortunately, decades long efforts to combat occupational accidents have stalled and improvements have reached a plateau. This paper presents a lean-based research agenda to better the occupational safety and health of the construction workforce while simultaneously reducing workflow variation and enabling lean conversion efforts. To this end, the paper begins first by establishing the nexus to the lean construction ideal of waste elimination. Next, a discussion of traditional and contemporary efforts and practices directed at controlling and eliminating occupational ac-cidents is presented. The paper then introduces and discusses a new paradigm for understanding and eliminating accidents. Guided by this new paradigm, the paper develops and concludes with a construction safety research agenda.The value of the recommended monetary incentive was thus established at 6% of asphalt materials cost. A performance-based incentive has the potential to provide benefits similar to pavement warranties, best value contracts or performance-based contracting procedures with significantly fewer legal entanglements.

Key Words:Lean Construction, Safety Theory, Occupational Safety, Construction Safety, Construction Accidents, Resilience Engineering, Rasmussen Model

INTRODUCTION

Each year, occupational injuries (including illnesses) and fatalities in the construction industry temporarily or permanently disable many and claim the lives of others. This problem is receiving increasing attention in the construction industry, as well as in other indus-

tries, not because of human suffering alone, but also due to many cost-related factors. Such factors include escalating workers’ compensation insurance costs, high direct costs of medical treatment and rehabilita-tion programs, and high indirect costs, such as admin-istrative costs, productivity losses and lower morale. These factors will increase construction costs, and adversely impact a contractor’s competitiveness.The staggering statistics collected and disseminated by occupational safety and health concerned organiza-tions confirm the impact and importance of this prob-lem. The Center for Protecting Workers’ Rights


ABDELHAMID, T.S., Associate Professor, 214 Human Ecology, School of Planning, Design and Construction, Michigan State Uni-versity, East Lansing, MI 48824-1323. [email protected]

MITROPOULOS, P., Assistant Professor, Del E. Webb School of Construction, Arizona State University, PO Box 870204, Tempe, AZ 85287-0204, Tel: 480-965-3378. Email: [email protected] SCHAFER, D., PhD Candidate, 401 Human Ecology, School of Planning, Design and Construction, Michigan State University, East Lansing, MI 48824-1323. [email protected]

HOWELL, G.A., Executive Director, Lean Construction Institute, Box 1003, Ketchum, ID 83340. 208/726-9989.


32 Construction Safety: A Lean Construction Research Agenda


(CPWR 1998) reports that: “Construction safety and health death rate is 15 deaths per 100,000 workers, or more than 1000 killed yearly- is more than three time the rate for manufacturing in the United States. The more than 182,000 serious injuries annually in construction in the United States not only wastes live and hurt productivity, but also substantially increase payroll costs (through workers’ compensation and other costs).” Also, according to the National Safety Council (NSC), 1,040 US construction workers lost their lives at work and another 350,000 received dis-abling injuries during 1996. Construction accounted for approximately 20% and 10% of all US indus-tries’ occupational fatalities and disabling injuries, respectively. This rate is disproportionately high, as the construction workforce represents only 5% of the total workforce (Accidents 1997).

The NSC further reported that costs per fatality and disabling injury in 1996 were $790,000 and $28,000, respectively. These costs include medical expenses, employer costs, administrative expenses, and lost wages. Fatalities and injuries in construction alone totaled approximately $10.6 billion, representing 8.8% of the total costs reported for all industries ($119.4 billion). The cost per construction worker of $1635 ($10.6 billion/6.5 million workers) was higher than the cost per worker of any other industry. For example, the cost per worker in manufacturing was $950 ($17.4 billion/18.3 million workers), and that for a worker in mining was $1170 ($70 mil-lion/600,000 workers).

It is disturbing to report that the NSC figures for 2006 indicate that construction also accounted for 6% of the United States’ workforce but claimed a disproportionate 23 % of all occupational fatalities and 10.5 % of all occupational injuries (Injury Facts 2007). Fatalities and injuries in construction alone totaled $12.32 billion, representing 10% of the total costs reported for all industries ($122.6 billion). The most recent version of the information rich “Con-struction Chart Book” (2008) reports that rates for construction overall work-related fatalities have

decreased by 22% in the period from 1992 to 2005 and nonfatal injuries and illnesses with days away from work (DFW) dropped by 55% in the same period. While the current statistics are considerably better than statistics in the ’70s and ’80s, it is clear that over a ten-year period improvement efforts in construction safety have reached a point of diminish-ing returns.

Based on statistics alone, safety appears to be im-proving somewhat in the construction industry. Further improvement is needed, but improvement has reached a plateau and construction still kills or injures more than eight percent of its workers each year. New approaches and outlooks are needed to break away from this stagnation and improve the safety of construction workers.

Unfortunately, there is no clear path to solve the construction safety problem. Researchers theorize that the problem may be due to entrenched lines of authority and rigid control structures. It appears that there is a “gap” between work as planned at what researchers call the “blunt” end, that comprises the executives, administrators, office engineers, and regulators, in, short those who are distal from the work face; and those who actually do the work, such as laborers and carpenters, who are at the “sharp” end where the work gets completed. This gap forces those at the sharp end to adapt the as-planned work procedures and standards to field conditions to complete the work in a timely, acceptable, and safe manner. In other words, the work as imagined by the blunt end planners oftentimes does not match the reality at the workface. However, accidents sometimes occur because of this adaptability, just as successes stem from the same adaptable behavior. It has been posited that better awareness of the blunt end planners may lead to fewer accidents (Schafer et al 2008). This awareness may lead to better support of sharp end activities such as less stringent produc-tion schedules. Further compounding the safety problem is the fact that decisions occur on different


Abdelhamid, T.S. , Mitropoulos, P. , Schafer, D . and Howell, G.A. , 33


time scales between the sharp and blunt end. Many researchers (Hinzie(2004), and Woods(2006)) sug-gest that safety may be enhanced by “pulling back” on production schedules by the blunt end agents.

Others think that it is human nature to stray into unsafe habits after a period of time has elapsed, no matter how stringent the regulations in place are or how strongly they are enforced (Groeneweg, 1998). Taking an historical overview of industrial safety it is seen that plateaus are not uncommon. From 1937 to 1956 the fatality rate for U.S. workers in all in-dustries decreased from 43 to 23 deaths per 100,000,

respectively (Groeneweg, 1998). This decrease was attributed to the intervention of engineering controls applied to industrial hardware, ergonomics science, and the increased use of person protective equipment (PPE).

Another factor was a decrease in fatigue due to lowering working hours. In the 1960s and 1970s the number industrial accidents stabilized. Efforts were expended on employee behavior in the form of motivational programs and improving the quality of associates. The 1980’s saw a 21% decline in the accident death rate. The current focus is on improv-ing the fatality rate via sociotechnical organizational change while engineering controls and behavior modifications are ongoing. Figure 1 summarizes these trends.

The lack of improvement in the safety record, or

its stagnation, may be a reflection of a fundamen-tal problem in understanding the accident process. Commonly, occupational accidents are seen as the result of a sequential combination of causes, mostly linear. Risks are viewed as arising from unreliable system components, both human and technological. In particular, traditional views attempt to model hu-man performance in systems purely in terms of basic elements or processes and at the level of the individ-ual, ignoring that construction work is accomplished in crews and in a social context. Highly structured analysis, such as fault-tree’s, are used to foretell the future. While these methods are certainly valuable and are useful to describe a certain class of failure,

they necessarily rely on a certain sequence of eventsand probabilities and suppose that accidents can be decomposed into constituent part and recomposed to form a safety model. Traditional methods are by nature reactive and are based on hindsight. Finally, highly structured models of safety tend to ignore the demands imposed upon projects, such as production pressures and tempo changes that are part and parcel of the construction experience (Schafer et al 2009).

A model that recognizes the pressures that push workers towards more risky behaviors is that ad-vanced in Rasmussen’s theory of cognitive systems practices directed at controlling and eliminating occupational engineering (Rasmussen et al. 1994). Howell et al. (2002) proposed a new approach to understand construction accidents based on Rasmus-sen’s model wherein it is emphasized that workers need to receive training to make them more

Figure 1: Plateaus in overall safety reached after interventions (adapted from Schafer et al 2008)



conscious of hazardous work environments and to engage the work with better planning and appropri-ate protection in a very similar way to how fire fight-ers engage hazardous situations. Schafer et al (2008) expanded on the discussions in Howell et al (2002), Saurin et al (2004), and Saurin et al (2007) to en-compass the region where demands placed on the production system force it out of its normal working range with regard to construction safety – novel and dangerous situations are encountered that were not possible to train for or anticipate. This was one of the first forays into resilience engineering and its ap-plication to construction safety.

The main goal of this paper is to introduce these new paradigms for understanding and eliminating acci-dents while simultaneously reducing workflow varia-tion and enabling lean conversion efforts. To this end, the paper begins first by establishing the nexus to the lean construction ideal of waste elimination. Next, the traditional and contemporary efforts and accidents is presented. The paper then develops a construction safety and ergonomics research agenda to enable the implementation of Rasmussen model which holds the potential to better the occupational safety and health of the construction workforce while simultaneously reducing workflow variation and enabling lean conversion efforts. The paper concludes with propositions for ‎implementing Resil-ience Engineering in construction settings and offers pointers to future ‎research.

CONSTRUCTION SAFETY AND ERGONOMICS IN LEAN SYSTEMS

Lean Construction has escaped canonical definition mainly because Lean principles defy easy charac-terization. A frequently referenced definition is that of the Lean Construction Institute (LCI) according to which lean construction is a production man-agement-based philosophy emphasizing the need to simultaneously design a facility and its produc-tion process while minimizing waste and maximiz-ing value to owners throughout the project phases

[including the post-construction phase] by improv-ing performance at the total project level, using a conformance-based vs. a deviation-based perfor-mance control strategy, and improving the reliability of work flow among project participants (Howell 1999 and Abdelhamid et al 2008). Stated differently, Lean Construction forces the explicit consideration of work flow and value management in addition to the traditional construction management focus on transformation management, i.e., transferring ma-terials into building objects (Everett 1992, Koskela 1993, Bertelsen and Koskela 2002, Abdelhamid et al 2008). Adding workflow and value management is integral to the successful delivery of capital projects.Workflow management is primarily concerned with managing the release of work from one production process (not activity) to another, as well as within the activity itself. In Lean Construction, workflow unreliability causes downstream workers to be idle resulting in waste, or results in work waiting on workers. The root causes behind unreliable flow must be identified and removed if work is to proceed reliably.

Unreliable workflow occurs when there is a fluctua-tion in the release of work amongst production pro-cesses (Ballard 1999, Abdelhamid 2003, Moosa and Abdelhamid 2005). This unreliable workflow is a result of variation stemming from single or multiple causes that need to be targeted separately or col-lectively. Under a lean paradigm, variation is con-trolled through the use of material and plan buffers, and/or flexible capacity (Ballard and Howell 1998). Raw and/or processed material is typically used as material buffers. Plan buffers refer mainly to having alternative work for crews in case preceding activi-ties hamper their work. Flexible capacity refers to using resources in multiple ways. This is enabled when workers are cross-trained. Other examples of flexible capacity can be found in Hopp and Spear-man (2000).

In the construction industry, sources of variation include late delivery of material and equipment, de-



sign errors, change orders, equipment breakdowns, tool malfunctions, improper crew utilization, labor strikes, and environmental effects. Other impor-tant indirect sources of variation in construction are traumatic, ergonomics, and exposure accidents that may lead to injuries and fatalities. When realized, injuries and fatalities in turn lead to decreased pro-ductivity and motivation, poor quality work, and job dissatisfaction (Brouha 1967, Janaro 1982, Abdelha-mid and Everett 2002).

HOW DOES CONSTRUCTION SAFETY AND ERGONOMICS RELATE TO LEAN CONSTRUCTION?

This paper proposes that the relation between con-struction safety and ergonomics and lean construc-tion is inherently reciprocal. On the one hand, inter-ventions designed to prevent traumatic, ergonomics, and exposure injuries and fatalities in construction contribute and are consistent with the lean ideal of removing or reducing waste. Considering the human suffering that is brought about by occupational mis-haps and the deleterious effect these mishaps have on the progress and success of production operations leaves no doubt that the lean ideal of eliminating waste should not be restricted to reducing waste by only revisiting material storage and handling proce-dures, equipment and workforce utilization. Simply stated, the reduction in lost workdays and loss of life will result in a more productive and reliable work-flow.

On the other hand, procedures and tools devised based on lean concepts and principles require the ex-plicit consideration of safety and ergonomic issues. This is clearly the case when Lean Design and Lean Assembly, as defined in the Lean Project Delivery System by Ballard (2000), are considered (Abdelha-mid et al 2008).

In the Lean Design phase, wherein the product and process are designed simultaneously, safety issues are considered during constructability reviews. Un-

derscoring the importance of “Design for Safety”, in 1997, the Construction Industry Institute (CII), compiled and disseminated detailed guidelines for designers to help reduce safety issues during con-struction. Examples of such guidelines include avoiding roof edges and skylights as locations for rooftop mechanical equipment, scheduling night work sparingly, and designing slabs on grade and mat foundations with closely spaced reinforcement, which allows a continuous walking surface (Gamba-tese 2000).

Under a lean production paradigm, lean assembly refers to simplifying the process of assembly through industrialization, modularizations, standardization, and continuous flow processes. To aid in this simpli-fication process, the assembly or production opera-tions are placed under scrutiny (e.g., using Kaizen events) and improvements are suggested (e.g., using the 5S process) to reduce waste that manifests itself as overproduction, rework, and long cycle times. It is typical for such reviews to identify opportunities for reducing the number of operations/steps required for production, thus leading to the reduction of waste and increase in quality. Though most likely seren-dipitous, a welcome by-product of these efforts is the improvement of safety and ergonomics related issues in the production process. The mere reduction of operations required for a production process means that there are less chances for traumatic, ergonomics, and exposure injuries to occur. This follows from the same logic that the fewer the number of opera-tions, the higher the quality of the product because there are less chances of making errors.

In the following sections, traditional and contempo-rary efforts and practices directed at controlling and eliminating occupational accidents are discussed. This is preparation for presenting a new paradigm for understanding and eliminating accidents.



UNDERSTANDING ACCIDENTS

Despite that accidents occur in an unending variety of forms, any general definition of the term is bound to be rather abstract, many definitions have been pro-posed and found acceptance, if citation in accident investigation and causation models can be regarded as a measure of acceptance. One of the earliest definitions of an accident belongs to Heinrich: “An accident is an unplanned and uncontrolled event in which the action or reaction of an object, substance, person, or radiation results in personal injury or the probability thereof “ (Heinrich 1980). Another epidemiological-based definition belongs to Such-man where an accident was defined as: “unexpected, unavoidable, and unintentional act resulting from the interaction of host, agent, and environmental fac-tors within situations which involve risk taking and perception of danger” (Suchman 1961).

These definitions, and others, have been criticized for including obvious contradictions demonstrated in the use of words such as “uncontrolled” (Heinrich) and “unexpected, unavoidable, unintentional” (Such-man). The argument against these definitions were that if the accident cannot be controlled, expected, avoided, and assumed unintentional then there is no need for accident causation models or accident pre-vention programs (Abdelhamid and Everett 2000). It was also common practice to define an accident based on the accident causation perspective adopted by the researcher. This resulted in a plethora of acci-dent definitions all claiming to be ‘The’ best defini-tion. While there appears to be no agreement on the definition of an accident, Peyton and Rubio (1991) report that the lowest common denominator among accident definitions is that an accident is: “An unde-sired event that results in physical harm to people or damage to property”.

Many construction workers and managers believe that accidents are inevitable and that risk taking is an accepted part of the job process (Peyton and Rubio 1991, Narang and Abdelhamid 2006). This belief is

attributed to the fact that occupational accidents can be generally regarded as randomly occurring events. However, thorough accident investigations reveal that most accidents could have been anticipated and prevented (Hendrick and Benner 1987). Unfor-tunately, these thorough investigations were only performed when rather serious accidents occur.

Although many accident investigation techniques are available, not all lead to effective prevention plans, and accidents still occur (Hendrick and Benner 1987, Ferry 1988). Many scholars have acknowledged that there are shortcomings in existing investigation techniques and, in general, an effective, comprehen-sive, practical, and systematic approach to accident investigation is lacking (Kjellen and Larsson 1981, Ferry 1988, Thomson 1999). The shortcomings in many investigation techniques are greatly aggravated by the stance most data collecting agencies have adopted, which is the role of providing statistics on the counts and types of occupational accidents and associated financial outcomes or impacts, while neglecting many other important documentable facts surrounding accidents. In fact, the accident inves-tigation forms used by many agencies do not help investigators systematically identify the root causes of the accident, or to make, therefore, effective prevention and improvement plans (Abdelhamid and Everett 2000).

The remarks of Edwards (1981) summarize the situ-ation:

“The haphazard nature of accident investigation and analysis provides none of the factors for a base constructive and positive accident prevention policy, although much time and effort is given to collect-ing information which is not put to constructive use. Accident reporting systems have not been designed as information systems, but have been grown in a relatively unplanned way. We need a method of reporting accidents capable of providing an accurate and effective basis for line management decision making.”



Another comment from Drury (1983) partly covers the situation:

“The poor quality of data has influenced our analy-sis (of accident investigation) from the point of view that some analyses that could have been taken were not and some which were conducted produced con-clusions which were suspicious.”

Despite its necessity as a phase, accident investiga-tion seldom addresses the factors that contributed to the accident causation, i.e., WHY the accident oc-curred. Brown (1995) has argued convincingly that accident investigation techniques should be firmly based on theories of accident causation and human error, which would result in a better understand-ing of the relation between the “antecedent human behavior” and the accident at a level enabling the root causes of the accident to be determined. Con-sequently, prevention efforts could be directed at the root causes of accidents and not at symptoms, lead-ing to more potent prevention efforts.

The following is a review of the most prominent and widely disseminated accident investigation and cau-sation models. For the most part the thinking will be described without attempt at detailed critique. Before proceeding, the reader’s attention is brought to the fact that some models were developed due to a brief surge of interest to a specific theory or con-cept. Such models seldom show up in contemporary literature on occupational accidents. Some models are not of primary interest to this paper, but are only provided such that the reader may appreciate the diversity and complexity of various existing models. This should explain the gaps or discontinuation in the narration.

In spite of being implicit and subtle, a distinction is noticed in the literature reviewed between accident causation models and accident investigation models. The distinction is partly attributed to the fact that an investigation model is expected to have a sequence of steps to be implemented in real accident scenar-

ios, while a causation model is mainly a theoretical approach to explain hypothetical or real accidents scenarios with less emphasis on the implementation part (Kjellen and Larsson 1981, Thompson 1999).

In general, both types of models provide many explanations for the occurrence of injuries and fatalities to industrial workers by stressing identi-fication of the underlying causes of accidents. The overall objective of these models is to provide tools for better industrial accident prevention programs. Accident prevention has been defined by Heinrich et al. (1980) as: “An integrated program, a series of coordinated activities, directed to the control of unsafe personal performance and unsafe mechanical conditions, and based on certain knowledge, atti-tudes, and abilities.” Other terms have emerged that are synonymous with accident prevention such as loss prevention, loss control, total loss control, safety engineering, safety management, incidence loss con-trol, among many others.

It is worth adding that in the process of reviewing the existing literature, it was noticed that disagree-ment on definitions for terms in safety research is rather common and much attention and devotion has been given to the definition of terms such as acci-dent, incident, property damage, loss control, inci-dent control, among many others. This observation points out that many “safety related” terms escape canonical definition, indicating the abstract nature of research in the safety arena. In addition, much of the disagreement on definitions of terms reflects the different views regarding the accident process as provided by many researchers.

ACCIDENT CAUSATION MODELS

The American industrial accident prevention move-ment started in 1892 when the safety department of Joliet works of the Illinois Steel Company was formed. This was followed by formation of the Na-tional Safety Council in 1913 (Zeller 1986). Indus-trial safety or the concept of safety was motivated



by the desire to reduce injuries and to save lives and property. With this objective in mind, a series of theorems were developed in the 1930s to define and explain accidents.

One of these theorems is that proposed by Heinrich in his 10 axioms on industrial safety, which helped many researchers to understand the accident process for the first time. The first and most famous axiom stated that: “The occurrence of an injury invariably results from a completed sequence of factors-the last one of these being the accident itself. The accident in turn is invariably caused or permitted directly by the unsafe act of a person and/or a mechanical or physical hazard.” (Heinrich et al. 1980). Heinrich’s theory became known as the Domino Theory, and until recently was universally accepted as the real description of the accident process.

Heinrich’s views were criticized for oversimplifying the control of human behavior in causing accidents and for some statistics he gave on the contribu-tion of unsafe acts versus unsafe conditions (Zeller 1986). Nevertheless, his work was the foundation for many others. Over the past thirty years the Domino Theory has been updated with an emphasis on management as a primary cause in accidents, and the resulting models were labeled as management models or updated domino models. Other models have evolved separate from the domino theory but were still based on Heinrich’s work. These models are classified into different categories such as behav-ior models, human factors models, systems mod-els, epidemiological models, decision models, etc. (Heinrich et al. 1980).

Management models hold management responsible for causing accidents, and the models introduced try to identify failures in the management system. Examples of these models are the Weaver Updated Dominoes (Weaver 1971), the Updated Domino Sequence (Bird 1974), the Energy Release model (Zabetakis 1975), and the Adams Updated Sequence (Adams 1976). Two other accident causation models

that are management based but not Domino based are: the Multiple Causation (Petersen 1971), and the Stair Step model (Douglas and Crowe 1976).

Human error theories are best captured in behavior models and human factor models. Behavior models picture workers as being the main cause of accidents. This approach studies the tendency of humans to make errors under various situations and environ-mental conditions, with the blame mostly falling on the human unsafe characteristics. As defined by Rig-by (1970), human error is “any one set of human ac-tions that exceed some limit of acceptability.” Many researchers have devoted great time and effort to defining and categorizing human error (e.g. Rook et al 1966, Recht 1970, Norman 1981, Petersen 1984, McClay 1989, DeJoy 1990, Reason 1990, Wagenaar et al 1990, and O’Hare et al 1994).

The foundation of most behavior models is the ac-cident proneness theory (Klumb 1995). This theory assume that there exists permanent characteristics in a person that make him or her more likely to have an accident. The theory was supported by the simple fact that when considering population accident statistics, the majority of people have no accidents, a relatively small percentage have one accident, and a very small percentage have multiple accidents. Therefore, this small group must possess personal characteristics that make them more prone to ac-cidents (International Labor Organization 1983). Other theories in behavior models include the Goals Freedom Alertness Theory (Kerr 1957), the Life Change Unit Theory (Alkov 1972), and the Motiva-tion Reward Satisfaction Model (Peterson 1982). For other behavioral models see Krause et al (1984), Hoyos and Zimolong (1988), Dwyer and Raftery (1991), Friend and Kohn (1992), and Krause and Russell (1994).

The human factors approach holds that human error is the main cause of accidents. However, the blame does not fall on the human unsafe characteristics alone, but also on the design of the workplace and



tasks that do not consider human limitations and may have harmful effects. Therefore, these models study the effect of a particular situation or environ-ment on human performance and the limitations of humans to perform tasks are also addressed. Cooper and Volard (1978) state environment and human characteristics (both physical and psychological) as factors that contribute to accidents and to human er-ror. They have also briefly discussed the concept of overload, which is when an individual is subjected to more than he or she can handle (Peterson 1975). These ideas are common to the field of human fac-tors engineering. Examples of human factor mod-els include, the Ferrel Theory (Peterson 1980), the Peterson model (Peterson 1982), the McClay model (McClay 1989), the DeJoy model (DeJoy 1990), and the HFACS model (Shappell and Wiegmann 2001).

Human factors models have been also adapted to what is collectively known as Human Reliability As-sessment (HRA) techniques. These techniques focus on the identification of error producing conditions (EPC). The first step in human reliability assessment techniques is to define and determine the source of the problem. Once such problems have been identi-fied they are then quantified in terms of their prob-ability of occurrence called human-error probability (HEP) (Kirwan 1994).

HEP is calculated using Equation 1:

HEP = occurtoerrorforiesopportunitofnumberThe

occuredhaserrorantimesofnumberThe

Different techniques exist for determining HEP. To calculate HEP, these techniques consider human fac-tor influences on performance called performance-shaping factors (PSF). PSF is defined as any factor relating to the individual(s), environment or task, which affect performance positively or negatively (Kirwan, 1994). Some of the common techniques to finding HEPs are Human Error Assessment and Reduction technique (HEART), Technique for Hu-man Error Rate Prediction (THERP), Absolute prob-

ability judgment (APJ), Success Likelihood Index Method (SLIM) and many more (Kirwan 1994).

A system model recognizes the strong interaction between individuals, their tools and machines, and their general work environment (Rodgers 1971, Ro-land and Moriarty 1990). Examples of such models are the Firenze Model (Firenze 1971), the Ball mod-el (Ball 1973), and the Perturbations model (Smillie and Ayoub 1976). For more recent models see (Lee, Tillman and Higgins1988). Other examples are also covered in Roland and Moriarty (1990), and Vincoli (1993).

Epidemiological models came about after the safety research community considered an accident to be an epidemic. Epidemiology is the search for casual as-sociation between diseases or other biologic process-es and specific environmental experiences. In 1961, Suchman proposed an epidemiological model that suggests that the accident phenomenon is the “unex-pected, unavoidable, and unintentional act resulting from the interaction of host, agent, and environmen-tal factors within situations which involve risk taking and perception of danger” (Suchman 1961). Exam-ples of other epidemiological models include Had-don (1968), Congress of the United States (1985), and Veazie et al. (1994).

Surry developed a decision model based on the epidemiological model of Suchman (Surry 1974). The model is based on three main steps that a per-son goes through when making a decision. These steps are perception, cognitive processes, and physi-ological ability. Other decision models include the Hale and Hale model (1970), Decision Tables model (Gausch 1972), the Value Table model (Gausch 1974), the Douglas model (Douglas and Crowe 1976), and the Anderson et al. model (1978).

ACCIDENT INVESTIGATION MODELS

Similar to accident causation models, accident inves-tigation models were developed with the same over-

(1)



all objective of finding the root causes of accidents. However, accident investigation has to meet more specific objectives because of the need to apply them to real life accident scenarios. According to Hen-drick and Benner (1987), an accident investigation process is “a process requiring special knowledge and skills applied systematically with self-discipline, and subject to study, examination, understanding, and improvement.” The following are also some of the general objectives of the accident investigation process (Ferry 1988):

• Reduce danger to employees and general public• Prevent loss of company resources• Improve procedures to prevent future accidents• Identify and address management needs• Prevent the loss of working force• Identify hidden and unknown cost impact• Identify managerial errors and procedural error• Educate supervisors, managers, and workers• Research purposes

Many accident investigation models have been developed in an attempt to meet the general objec-tives listed above (Spear 2002). With a myriad of accident investigation available, criteria for compar-ing accident investigation models, was provided by Kjellen and Larsson (1981) including many others (Ferry 1988). Those proposed by Kjellen and Lars-son (1981) are as follows:

• The model should be suitable for practical investigation work.• The concepts and definition in the model should be easy to understand and should be related to concepts and terms that are in general use.• The model should be suitable for use for different types of accidents and system.• The model should be complete, i.e., no important causal factors should be omitted.• Use of the model should result in information that indicates hazards and is suitable for use in the prevention work.

Despite the distinction made in the literature be-tween accident causation models and accident inves-tigation models, some accident causation theories or concepts have been used as the basis for accident investigation models. For example, the Multilinear Events Sequencing Method of accident investigation presented by Benner (Benner 1975), is partly based on the thinking of the Domino Theory by Heinrich discussed in the accident causation section.

In general, accident investigation models can be cat-egorized into many types depending on the type of accident being investigated. For example, if a sys-tem failure is involved a system safety model is used in its original format such as the Fault Tree Analysis, the Failure Mode and Effect Analysis, among many others (Roland and Moriarty 1990).

When management errors are investigated, the Technic of Operational Review (TOR) developed by Weaver (1973) and the Management Oversight and Risk Tree (MORT) method could be used (Johnson 1975). Weaver developed TOR as a diagnostic tool for identifying management failures and omissions. When used in investigating accidents TOR is ex-pected to:

• Review the relevant management and supervisory factors uncovered in the investi gation.• Methodically consider, then accept or reject, more management and supervisory factors through a directed, sequential process.• Automatically consider contributions of supervision, management, and staff to the mishap process without pointing fingers or direct blame assessment.• Allow in-depth probing of difficult-to-detect- management contributions.• Force an in-depth look at portions of the investigation not carried out in enough detail, to resolve acceptance or rejection of a TOR factor.

The other technique, MORT, views management as the primary cause of accidents or “mishaps”. Uti-lizing a standardized checklist based on a fault tree



structure format identifies management failures. In addition to fault tree analysis, MORT also evolved from a technique called change analysis, which was developed by Johnson and Aerojet Nuclear Corpora-tion. The heart of MORT is diagrams and charts.In MORT, Johnson introduced the term mishap rather than the conventional term accident. Accord-ing to Johnson a mishap is an unwanted transfer of energy that produces injury to persons, damage to property, degradation of an ongoing process, and other unwanted losses. This concept, which was also introduced by many others (Haddon 1968), points out that identifying the source of energy is a basic prevention approach. Johnson suggested that mishaps occur due to the lack of energy barriers and/or controls. He attributed the lack of energy barriers and/or controls to specific job oversights and omis-sions, management failure, and assumed risks.A Mini-MORT model was introduced by Fillmore and Cornelison (1986) to simplify the logic and presentation of the full MORT analysis. The Mini-MORT model can be used in oral and written reports that summarize a full MORT analysis. In general, the MORT process, system, or charting technique for accident investigation is considered a sophisticated investigative analysis tool. Due to the complexity of the MORT technique, entire volumes have been ded-icated to introducing and explaining it. The reader is referred to Knox and Eicher (1976) or Ferry (1988) for a more detailed treatment of the technique.Recognizing the lack of a time scale to show the chronological relationships of events in most inves-tigating methods involving charting concepts such as MORT, Benner (1975) introduced the Multilinear Events Sequencing (MES) method. Benner also de-veloped the method to overcome the disadvantage he showed to exist in the use of checklists or extended diagrams. According to Benner these techniques en-courage the search for all causes assuming extensive knowledge of the accident process, while discourag-ing the search for factors not in the checklists or on the diagram.

In MES, it is emphasized that knowing when an ac-cident began and ended indicate a successful acci-

dent investigation, and constitute grounds for effec-tive accident prevention. According to Benner the accident or “mishap” is considered to begin when a stable situation is disrupted and an accident sequence begins. The concept of accident sequence is similar to the basis of Heinrich’s Domino concept. Ben-ner then argued that by common sense, an accident ends with the last injurious or damaging event in the continuing accident sequence.

Benner suggested that avoiding an accident is only possible if the person (or object) involved in the ac-cident sequence adapts to the disruption in the stable situation. If accidents do not occur, Benner suggest-ed that these near misses or incidents be investigated still. Benner believed that MES provides great value to anything that went wrong regardless of the end result.

MES relies on understanding what Benner calls the time line, the event, the actor, and the action. The time line is a scale that parallels the sequence of events to show a time relationship of the events that happen in an accident. The start of a sequence of events is denoted by To, and the end is denoted by Tn. Benner defines an event as something signifi-cant that takes place, usually stemming from one or more actions. An actor is the person or the object which performs an action that causes an event. Fur-ther explanation of the MES method may be found in Ferry (1988).

Kjellen and Larsson (1981) presented an investiga-tion model that classified perturbations, that’s what they chose to call accidents, as deviations of system variables. The model is based on previous work in system models, energy release or exchange concept, the Multilinear Events Sequencing approach, and on human error concepts common to the behavior and human factor models.

Kjellen and Larsson pictured the accident as a se-quence of deviations that occur at different phases. The initiating phase starts the deviations, which con-tinue to the injury phase, and end in the concluding



phase. The model has two main levels. In the first, the accident sequence is specified. In the second, the “Determining Factors” that may have contributed to the accident are identified. Various factors have been suggested that reflect aspects of the man-machine system (Physical/Technical factors; Organizational / Economic factors; and Social/Individual factors). According to Kjellen and Larsson, it is deviations in one of these aspects of the man-machine system that would increase the probability of accidents (Kjellen and Larsson 1981).

Hendrick and Benner (1987) developed the Sequen-tially Timed Events Plotting (STEP). The method is heavily based on the Multilinear Events Sequencing (MES) described above. Similar to MES, STEP uses time lines, events, actors, actions, and build-ing blocks. The first step in STEP is to arrive at the accident scene and document the states of people and objects involved. Reconstruction of how those states came to be are the objectives of the investiga-tion using STEP. Definitions of to, and tend, which are the same as To, and Tn are somewhat more subtle than in MES. For example, tend must sepa-rate between actors and reactors, and changes that occurred after the last harmful event from those that occurred during the accident process. More restric-tions are placed on to. Further explanation of the MES method may be found in Hendrick and Benner (1988).

CONSTRUCTION OCCUPATIONAL SAFETY AND HEALTH

Various parties have made efforts to reduce injuries and fatalities in construction and other work sectors by developing or implementing prevention plans and techniques. Such parties include occupational health and safety agencies, insurance companies, contrac-tors, owners, safety consultant companies and the academic community. In this section these efforts will be briefly reviewed.

OSHA and NIOSH

Soon after its inception in 1970, the Occupational Safety and Health Administration (OSHA) began enforcement and promulgation of the Construction Safety Act. This Act required OSHA to issue and enforce standards for safe work practices and up-dates them regularly (Title 29 Part 1926 of the code of federal regulations), provide training programs for workers, and collaborate on research projects related to safety. However, the Act was narrow in scope and primarily covered employers’ (managers’) administrative actions. The Act failed to recognize that construction is an industry controlled by a busi-ness and management system in which safety plays a central role (MacCollum 1990). Nevertheless, a recent study by Miller (1995) reports that OSHA has made some progress in reducing accident rates in construction. There is no doubt that saving lives of many workers is a credit to OSHA; however, several questions remain unanswered, especially that inves-tigation and prevention efforts are always targeting symptoms rather than the root causes behind the accidents (Culver at al. 1990, Culver at al. 1992, Culver at al. 1993, Culver and Connolly 1994).The National Institute for Occupational Safety and Health (NIOSH) is the other leading government agency on occupational safety and health. NIOSH specializes mainly in performing and collaborating on research projects in safety related issues for many industrial areas. In 1985, NIOSH, through its Divi-sion of Safety Research (DSR), developed a long-term research plan to improve safety in the construc-tion industry by prioritizing research areas within construction where fatality and incidence rates are highest. Typical examples are the studies found in Parsons et al. (1986), Becker at al. (1999), and NISOH (2000). In these studies, most of the identi-fied and investigated trouble spots were symptoms of the more complex and difficult to identify root causes. In spite of that, the NIOSH research efforts have helped reveal many construction sites’ prob-lems, but still more effort is needed to understand the root causes behind accidents.



Currently, through its National Occupational Re-search Agenda (NORA), a partnership program unveiled in 1996 to stimulate innovative research and improved workplace practices, NIOSH organiz-es and conducts research in the most critical issues in workplace safety and health. In 2006, NORA went to a new sector based structure, within which construction has been identified as a major sector of focus. This new structure was conceived to better move research to practice within workplaces.The Center for Construction Research and Training The Center for Construction Research and Train-ing, known widely with its prior acronym CPWR, is a non-profit research organization established by the Building and Construction Trades Departments, AFL-CIO, and its fifteen-affiliate unions. CPWR activities include the collection, evaluation, and dis-semination of research findings related to the safety and health of construction workers. In 1990, NIOSH and CPWR started to collaborate on research that emphasizes an applied nature with the goal of iden-tifying and eliminating safety and health problems in the construction industry using practicable ap-proaches. CPWR remains an active agent in promot-ing the health and safety of construction workers in the United States.

The Construction Safety Council

The Construction Safety Council (CSC) was found-ed in 1989 as a non-for-profit organization dedicated to the advancement of safety and health interests in the field of construction. The CSC states its mission as: “to reduce the tragic and costly accidents, injuries and illnesses that take the lives of six construction men and women in the United States every day. The Construction Safety Council (CSC) is a not-for-profit organization funded by the contributions and grants from both the public and private sector.” (CSC 2009).

INSURANCE COMPANIES, OWNERS, AND CONTRACTORS

Insurance companies are also interested in reduc-ing accident rates by providing safety consulting services, and endorsing the use of safety equipment and personal protective equipment. In addition,

they have adopted safety rating systems such as the Experience Modification Rating (EMR), which “penalize” contractors with bad accident experience ratings by increasing their workers’ compensation in-surance premiums. The experience rating system is designed as an incentive for contractors to improve their safety procedures. However, the EMR system of rating has been criticized by many researchers and contractors, due to the thorough knowledge needed in applying it and the misleading information it sometimes may provide regarding a contractor’s safety record (Everett and Thompson 1995). Other parties, such as contractors and owners, try to reduce accident rates by providing and requiring personal protective clothing and equipment to be used by workers, or/and establishing safety and health train-ing programs that follow OSHA or other agencies’ standards and recommendations.

SAFETY AND HEALTH RESEARCH IN CONSTRUCTION

Efforts to reduce construction accidents have been undertaken by many researchers (Hinze 1978, Hinze 1981, Hinze and Parker 1981, Fullman 1984, Gold-smith 1987, Davies and Tomasin 1990, La Bette 1990, MacCollum 1990, Rietze 1990, Peyton and Rubio 1991, Helnader 1991, Liska et al 1993, Levitt and Samelton 1993, CoVan 1995, Mangum 1995, Hinze 1997, Jonson et al. 1998, Vargas 1998, Mc-Cann 2000, Hinze and Wilson 2000, Abdelhamid and Everett 2000, Pollack and Chowdhury 2001, Molenaar et al. 2002, Mohamed 2002, Fredericks et al. 2002). These efforts have mostly concentrated on managerial prevention plans, safety program imple-mentation, and worker safety training. Although these areas are important and provide valuable tools for construction companies, they have typically been based on symptoms of the accident and not on the root causes. Consequently, effective preven-tion plans or worker training programs cannot be achieved nor would they have the desired objective of “zero accidents” that is repeatedly referred to in many studies.

For example, MacCollum (1990) mentioned that construction contracts are written in such a way that big holes are left for concerned parties to disregard safety. La Bette (1990) suggested that the use of



construction equipment without safeguards is anoth-er cause of accidents on construction sites and that provisions are needed to force the use of safeguards. Peyton and Rubio (1991) have also reported a study performed at Stanford University, which concluded that most workers believe that risk taking is an ac-cepted part of the job process. More recently, design impacts on construction safety received consider-able attention (Hinze and Wiegand 1992, Gambatese 2000, Suraji et al. 2001). Jaselskis et al. (1996) identified the frequency of formal safety meetings with project supervisors and safety budget as factors related to lower incident rates. Liska et al. (1993) identified five safety best practices: pre-project and pre-task planning for safety, safety orientation and training, safety incentives (the effect of this has been debated), alcohol and substance abuse programs, and incident investigations. These best practices are considered essential in reaching zero injuries (Hinze and Wilson 2000).

All the problems identified in the above examples can be considered symptoms that can only be elimi-nated by identifying the underlying root causes. To address this need in the construction industry, con-struction accident causation models were proposed in the literature by a handful of researchers. Mc-Clay’s ‘universal framework’ (1989) identifies three key elements of accidents: hazards, human actions, and functional limitations that are exceeded in the case of an accident. Hinze’s distraction theory (1996) argues that production pressures or other stress factors can distract workers from the hazards and in-crease the probability of accidents. The ‘constraints-response’ model (Suraji et al, 2001) illustrates that any project condition or management decision (distal factors) can cause responses that create inappropri-ate conditions or actions (proximal factors) that may lead to accidents. The root cause analysis model by Abdelhamid and Everett (2000) identified three gen-eral root causes—management deficiencies, training, and workers attitude. Toole (2002) proposed eight root causes: lack of proper training, safety equip-ment not provided, deficient enforcement of safety,

unsafe equipment, method, or condition, poor safety attitude, and isolated deviation from prescribed behavior. These models are based on variants of the ‘industrial’ accident causation and investigation models described above. Despite the contributions of these construction accident causation models in understanding the accident process, we are still far from adequately explaining the underlying causes of construction accidents.

This paper asserts that the shortcomings in accident investigation reflect a fundamental problem in un-derstanding the whole accident process, from which it follows that once the accident process is fully understood, a plausible and effective accident in-vestigation model can be developed. Thus far, most efforts to understand the accident process have failed to recognize the strong dependence of investigation on understanding the accident process, and hence have failed to propose an effective accident investi-gation model.

A new approach to understand construction accidents has been proposed by Howell et al. (2002) based on the work of Rasmussen et al. (1994), and expanded further in Schafer et al (2008). The model suggested recognizes that organizational and individual pres-sures push people to work in hazardous situations. These pressures defeat efforts to enforce safe work rules specifically in a changing work environment such as in construction. Therefore, this approach em-phasizes the need to train workers to be conscious of hazardous work environments and engage the work with better planning and appropriate protection in a very similar way to how fire fighters engage hazard-ous situations (Patel and Abdelhamid 2004, Schafer et al 2008, and Mitropoulos and Cupido 2009).

THE RASMUSSEN MODEL

Despite the contributions of construction accident causation models in understanding the accident pro-cess, the dynamic nature of construction work needs to be explicitly considered and investigated. Howell



et al. (2002) proposed a new approach to understand construction accidents based on Rasmussen’s theory of cognitive systems engineering.

The original model as proposed by Rasmussen is shown in Figure 2. As shown, Rasmussen divided the work environment to three zones. Zone I, which is the region enclosed by the “Boundary of uncondi-tionally safe behavior”, “Organizational Boundary to Economic Failure”, and “Individual Boundary to Unacceptable Workload”, is considered the safe zone (Rasmussen et al. 1994, Rasmussen 1997).

Organizational Boundary

to Economic Failure

Individual Boundary to

Unacceptable Workload

Boundary of unconditionally Safe

Behavior

Workload Gradient

Cost

Gradient

Irreversible loss of Control Boundary

Increasing Risk

Safe Zone

Loss of Control Zone

Hazard Zone

Figure 2: Three Zones of Risk (Howell et al, 2002)

Rasmussen states: “due to economic or workload pressures, workers will shift their work along the workload and/or cost gradients, respectively.” So as long as workers remain within the safe zone, work activities can be safely performed. Current safety regulations and management practices are directed at keeping the workers in the safe zone. Rasmussen suggests that enlarging the safe zone through proper planning of operations will make the work safer.The zone encompassed by the “Boundary of Uncon-ditionally Safe Behavior” and the “Irreversible Loss of Control Boundary” is Zone II or the Hazard zone. Workers working in the Hazard zone are consid-ered to be working at the edge (pushing their luck).

Rasmussen believes that, despite regulatory or supervisory efforts, workers will move to the Hazard zone due to many reasons. He suggested, contrary to current conventional wisdom, that the only effec-tive way to counter these tendencies to work in the Hazard zone would be to make visible the bound-ary beyond which work is no longer safe and teach workers to recognize the boundary.

The third and final zone in Rasmussen’s model is the loss of control zone, in which accidents occur and control is lost leading to injuries and/or fatalities. So he proposed that workers should be educated and trained on how to recover from such situations. This is very similar to instructing drivers on icy roads on how to respond to slips.

Rasmussen’s theory recognizes that organizational and individual pressures will push people to work in hazardous situations. These pressures defeat efforts to enforce safe work rules specifically in a changing work environment such as in construction. There-fore, this approach emphasizes the need to train workers to be conscious of hazardous work environ-ments and engage the work with better planning and appropriate protection in a very similar way to how fire fighters engage hazardous situations.According to Rasmussen the worker him/her self is the best person to judge the boundaries of safe work.

So instead of forcing workers to follow the rules to stay in the safe zone, Rasmussen suggested to train workers to:

1. Identify which zone they are working in2. Identify hazards3. Prevent hazard release4. Recover when hazards are released

While counterintuitive, Rasmussen’s recommenda-tion to train workers to deal with hazards and re-cover from scenarios when control is lost recognizes that workers will frequently and inevitably work in the hazard zone. Management pressures and seeking



less effort are realistic examples pushing workers to the hazard zone.

The Rasmussen model is similar to the current trend in the social sciences of using social norms market-ing. In this approach, informing people what to do versus the traditional approach of telling people what not to do brings about desired behavioral changes. For example, instructing heavy college-aged drink-ers not to drink before driving and citing the grave consequences of such behavior has become an over-rated message that lost its effectiveness. Alterna-tively, social norms marketing follows a tact where people are given instruction on what to do if they would like to drink – To drink moderately and have fun while sparing yourself and others the risks.While Rasmussen still maintains that safety and performance will increase if the safe zone is enlarged with proper planning, he clearly acknowledges the need to tell workers what to do at the edge and when control is lost versus the overrated messages work-ers hear, e.g., don’t dry-cut bricks because you will develop silicosis.

The acceptance and effectiveness of Rasmussen’s approach remains an open question that only future research can answer. Howell et al. (2002) recom-mended that future research efforts consider the fol-lowing three areas:

1. IN THE SAFE ZONE: Establish methods and techniques to enlarge the safe zone.

2. AT THE EDGE: Train workers on the iden tification of safe and unsafe conditions. And once in an unsafe condition, workers should be trained on how to recover from errors.

3. OVER THE EDGE: People will inevitably make mistakes resulting in loss of control. Hence, measures should be in place to limit the effect of this loss.

In the following sections, methods and techniques that could be utilized in the implementation of the Rasmussen model are discussed. This discussion is not intended as directives but rather food for thought for researchers considering future research.

IN THE SAFE ZONE

To enlarge the safe zone, operations planning should consider potential sources of traumatic events that may lead to injuries and/or fatalities. The industrial and construction safety literature abounds with tech-niques for pre-project and pre-task planning. In fact, occupational safety efforts in construction have typi-cally focused on traumatic type injuries and fatalities that occur during work. It is important to note here that OSHA considers traumatic injuries as any physi-cal damage to the body such as cuts, burns, fractures, electric shocks, sprains, crushes, bruises, abrasions, dislocations, etc. In most cases, these traumatic injuries or fatalities are sustained because workers have either performed unplanned work (work they shouldn’t have done) or acted in an unsafe manner.However, workers also get injured during planned work (work they should do exactly as they did). These injuries are the result of performing tasks as planned, and are built into the tools, work methods, and work environment. According to OSHA, these injuries fall into either ‘repetitive use injuries’ and/or occupational illnesses. Examples of ‘repetitive use injuries’ include chronic hand or wrist pain, low-back strain, upper-body disorders, whole-body physical fatigue, etc. Symptoms of these injuries often take weeks or months to develop and include discomfort, numbness, and loss of strength. Repeti-tive use injuries are also known as ‘musculoskeletal disorders’, ‘cumulative trauma disorders (CTDs)’ or ‘overexertion injuries’ (Everett 1999).

As for occupational illnesses, sometimes also re-ferred to as occupational diseases, OSHA considers these as injuries sustained due to exposure to harsh working conditions, to includes toxic poisoning, heart conditions, heat strokes, nerves, ganglia, influ-



enza, pneumonia, mental disorders, skin irritations, dermatitis, etc. Other examples include illnesses and diseases caused by inhalation of metal fumes, lead and benzol poisoning, inhalation of volatile solvents, fumes, and fine particles (on the top of the list of disease-causing fine particles are silica, quartz, and asbestos).

To avoid confusion, the term ergonomic injury and exposure injury will be adopted in this paper to refer to repetitive use injuries and occupational illnesses, respectively.

To properly enlarge the safe zone, sources of poten-tial ergonomic and/or exposure problems that may lead to injuries and/or fatalities should be considered during operations planning. Addressing these prob-lems is the topic of the following two sections.

ERGONOMIC INJURIES

In general, ergonomics injuries are the leading cause of work-related disabilities. Ergonomics injuries account for approximately 31% of worker’s com-pensation claims and cause 60% of lower back pain (Chaffin et al. 1999). The National Safety Council reports that ergonomics injuries in construction rep-resented 22% of all non-fatal construction incidents (Injury Facts 2002).

Prevention of ergonomic injuries is one of the objec-tives of the discipline of ergonomics. In addition, ergonomics is also concerned with the study and development of general principles that govern the interaction between humans and their working envi-ronment (Chaffin et al. 1999). The motto in ergo-nomics is to ‘fit the work to the worker and not the worker to the work’.

The ways ergonomic injuries manifest themselves should not be confused with the conditions that lead to them. These injuries are attributable to biomechanical problems, extreme/unfavorable environmental conditions, and work stressors such as vibration, noise, illumination, and physical work

overload. For example, low-back problems are primarily caused by biomechanical problems (lifting items in awkward positions, carrying heavy items, etc.). Vibration transmitted by the hands and body may also result in low-back strain, chronic hand or wrist pain (e.g. carpal tunnel syndrome, tendonitis, tenosynovitis), degenerative joint disease, decalcifi-cation. Working in noisy environments has adverse effects on the hearing capability such as acoustic trauma, partial or complete hearing loss, tinnitus, etc. Noise has other effects on health including sleepless-ness, nervousness, vasoconstriction, muscle tensing, and fatigue (Proctor and Zandt 1994). In addition, working in noisy environment is directly correlated with deterioration of performance (Colley and Beech 1989).

Proactive and reactive ergonomics programs may be used as primary and secondary prevention of ergonomic injuries, respectively. Preempting the potential for construction ergonomic injuries can be achieved by changing the work methods, including investment in more automated tools and equipment; providing appropriate work-rest cycles; or even adjusting expectations of what workers can reason-ably be expected to accomplish. These and many other examples of administrative and engineering interventions to reduce physical demands and fatigue would provide endless opportunities to improve construction work today and enable lean conversion efforts.

Prior to establishing such programs, ergonomic as-sessment studies are conducted first to determine the type and potency of a risk associated with an activ-ity. Armstrong (1993) identified seven ergonomic risk factors that are associated with overexertion injuries in manufacturing and service jobs. These were: 1) repetition, 2) static exertions, 3) force, 4) localized mechanical stresses, 5) posture stresses, 6) low temperature, and 7) vibration. To reduce expo-sure to these risk factors and the injuries they cause, the quality and design of tasks and tools being used in many construction activities have to be considered with more attention.



Other approaches to assess risks of ergonomics inju-ries include the use of biomechanical, psychophysi-cal, and/or physiological assessments of work. One example of a biomechanical assessment would be to determine how much muscular strength a physical activity requires and relating that to recommended and allowable muscular strength a worker may exert based on concepts of biomechanics. An example of psychophysical assessment would be to ask a subject to carry various objects of various weights, and determine the sensations or perceptions in the psychological domain (mental) via subjective means (survey, direct questions to subjects) of how intense, painful, tiring, etc. was the task, and which parts of the body were more affected. An example of a physiological assessment would be to determine how much energy expenditure a physical activity requires and relating that to recommended and allowable energy expenditure levels a worker may expend for that activity, and/or by observing selected work-rest cycles by the worker and comparing it to work-rest fatigue-free cycles estimated based on concepts of work physiology (Astrand and Rodhal 1986).An approach worth considering also is that of Hen-drick and Kleiner (2001), where they advocate the use of Macro-Ergonomics which they define as “a top-down sociotechnical systems approach to the de-sign of work systems and the application of the over-all work system design to the design of human-job, human-machine, and human-software interfaces.”The following are examples of ergonomic analysis guidelines to consider, in an effort to avoid or reduce incidents of ergonomic injuries (Chaffin et al 1999):

1. Using mechanical handling aids like balanc ers, hoists and conveyors, where possible.2. Optimizing strength by proper positioning of tools, materials3. Keeping materials close at hand (horizon tally) to avoid work with arms outstretched4. Avoiding overhead work.5. Using finger-padded handles to reduce vibra tion and contact stress6. Positioning work to optimize visual capabili ties

7. Evaluating need for anti-fatigue/anti-slip flooring8. Evaluating sit/stand options9. Maintaining a good work environment: consider lighting, temperature, and low noise levels

Attempts to use the risk factors mentioned to as-sess construction work have resulted in a number of ergonomic exposure assessment techniques such as the OWAS method (Kivi and Mattila 1991), MAS method (Schlidge et al 1997), and the PATH method (Buchholz et al 1996). Numerous studies have used variants of these techniques but the findings were similar. The studies reported that most construction activities place workers at increased risk for over-exertion injuries based on one or more of the risk factors mentioned (Schneider and Susi 1994, Schnei-der 1998, Everett and Kelly 1998, Everett 1999). Examples of psychophysical assessment of work are not found in the construction literature. However, biomechanical and physiological studies have been reported (Bernold et al 2001, Abdelhamid 1999, Ab-delhamid and Everett 1999, Abdelhamid and Everett 2002).

INDUSTRIAL HYGIENE

The focus of industrial hygiene is the prevention of exposure injuries (occupational illnesses/diseases) through the use of chemistry, toxicology, and physi-ology (Fraser 1989, Heath 1991). Industrial hygien-ists are concerned with maladies caused by expo-sure to harmful work environments (see examples mentioned earlier). An enormous body of research considering exposure injuries in construction is available in industrial hygiene literature.

AT THE EDGE

Teaching workers to recognize that they have stepped into the hazard zone appears to be achiev-able through intensified and directed training. A tool worth considering in training workers to iden-



tify hazards (unsafe conditions) is Crew Resource Management (CRM). Helmreich et al. (1999) define CRM as:

“The utilization of all available human, informa-tional, and equipment resources toward the effective performance of a safe and efficient flight. CRM is an active process by crewmembers to identify signifi-cant threats to an operation, communicate them to the PIC, and to develop, communicate, and carry out a plan to avoid or mitigate each threat. CRM reflects the application of human factors knowledge to the special case of crews and their interaction.”

CRM primarily evolved as a tool to reduce the number of human-related accidents in the aviation industry. This is not surprising given that the major contributor factor in 80% of commercial aviation accidents is attributed to human error (Federal Avia-tion Administration 1998). CRM has gone through a number of transition phases. The current genera-tion of CRM training is founded on the fact that human error is inevitable. Accepting this reality, it behooves management to institute an organization-wide safety culture.

The implementation of CRM relies on the collection of relevant safety data. Helmreich et al. report five critical sources for such data, namely, form evalu-ations; incident reports; surveys; Flight Operations Quality Assurance (FOQA); and Line Operations Safety Audits (LOSA). Of these, LOSA offers an interesting approach for construction. LOSA is a program where an expert observer monitors crew behavior, actions, and reactions during real-time flights. Superior performance as well as areas of weakness is communicated to the crew. This could take place after the flight lands or during the flight to cease any teaching moments. Adapting this aspect of CRM to construction may reveal many trends related to worker behaviors and response to unsafe conditions. For an in-depth discussion of CRM top-ics and techniques, the reader is referred to Helm-

reich et al. (1999) and Klinect (1999).The focus on worker training to recognize hazards (unsafe conditions) assumes that workers will al-ways recall what constitutes a safe or unsafe situa-tion as well as respond to perceived or actual risks in the same manner. Therefore, a methodology to assess worker ability to recognize hazards is in order. Such a method must account for the individuality of each worker and how variable each one of us is when it comes to sensitivity to unsafe conditions and risk orientation (the tendency of a worker to work in a condition despite knowing its unsafe). Once performed, this assessment could be used to give guidance to workers on how to enhance their abili-ties to identify the boundary beyond which work is no longer safe.

Signal detection theory (SDT) is an assessment technique that was developed for tasks requiring the detection of defective components in an industrial setting. Discussion of signal detection theory and how it could be tailored for assessments of the sensi-tivity and risk orientation of construction workers to unsafe conditions follows.

SIGNAL DETECTION THEORY

In the manufacturing industry, quality inspections are performed on products to reject defective ones. A perfect quality inspection process would be able to identify and reject all the defective products. This is seldom attained even with the use of sophisticated equipment. The inspection problem is also found in other industries or job situations such as detecting a tumor on an X-ray plate by a radiologist, detecting weapons by an airport security guard, and for the purposes of this research determining if the work condition is safe or unsafe.

The number of defective products that escape detec-tion (misses) and non-defective ones that are rejected (false alarms) gives a measure of the effectiveness of an inspection process. These two measures have also become the basis for characterizing the sensitiv-



ity of the operator performing inspection. Research-ers have dubbed the framework leading to such characterization as “Signal Detection Theory” or SDT (Ihara 1993, Swets 1996).

SDT is applicable in situations where two discrete states of the world (signal and noise) cannot be eas-ily discriminated. In such situations, a human opera-tor (or machine) is faced with the task of identifying one of the states. If the state of the world is a signal, e.g., a defective product, the response of the operator (or machine) is either ‘yes’ the product is defective (a HIT) or ‘no’ the product is not defective (a MISS). If the state of the world is noise, e.g., the product is not defective, the response of the operator (or machine) is either ‘yes’ the product is defective (a FALSE ALARM) or ‘no’ the product is not defective (a CORRECT REJECTION). These situations are represented as shown in Table 1. Clearly, the perfect result should not have any false alarms or misses – an ideal situation not possible in real life.

State of the worldSignal Noise

ResponseYes Hit False AlarmNo Miss Correct

RejectionTable 1: The four outcomes of signal detection theory (Wickens, 1992).

In a signal detection task, operators sometimes have response biases and are prone to say ‘yes’ more often than they should, thereby detecting most of the signal but also producing many false alarms. The other response could be conservative by saying ‘no’ and producing few false alarms but missing many of the signals (Wickens, 1992). Depending on the task, an approach with fewer false alarms may be better than not missing any signal (100% inspection) while having many false alarms given the cost of repeated disruption to the work – it is important to note that Lean Construction advocates built-in quality and not 100% inspection at the end of an activity/phase.In SDT, the signal indicator or strength is assumed to have a normal distribution (argued using the central limit theorem). It follows then that the information in Table 1 could be graphically represented as shown in Figure3. Xc, shown in Figure 3, represents the critical level where an observer decides the nature of

a signal. In other words, Xc represents the “mental” cut-off the observer uses to decide whether to say ‘yes’ there is a signal (a hit), or ‘no’ there is noise (correct rejection).

Figure 3: Distribution of detection theory (Wickens 1992)

In Figure 3, the shaded portion on the left of Xc represents the missed signals by the observer. The striped portion on the right of Xc represents the sig-nals the observer incorrectly considered as hits, i.e., these are false alarms. The change in the position of Xc determines the respective proportion of misses to false alarms (Swets 1996). For example, if Xc cuts more into the signal side, then most responses will be ‘no’ resulting in numerous misses and fewer false alarms (as well as fewer hits). If Xc cuts more to the left, most responses will be “yes” resulting in fewer misses but more false alarms.

The mental cut-off, Xc, chosen by an observer is quantified using a parameter termed the response cri-terion or likelihood ratio and is denoted as βcurrent. This parameter has also been termed the judgment or decision criterion of the observer. Mathematically, and as shown in Figure 2, βcurrent is the ratio of the coordinates P (X/S) and P (X/N) for a given level of Xc. P (X/S) and P (X/N) represent the conditional probability of Xc given a signal and the probability of Xc given noise, respectively. βcurrent is calcu-lated using Equation 2.



)/()/(

NXPSXPcurrent

High values of βcurrent indicate a high number of misses, whereas a lower one will generate more false alarms. Because of inter-observer variability with respect to the choice of Xc, evaluating the results of multiple observers requires the normalization of the value of βcurrent or the comparison to an optimal value. The optimal value for β has been taken as the value corresponding to a minimum number of errors, i.e. minimum misses and false alarms. Math-ematically, this value is the ratio of the probability of noise, P(N), and the probability of a signal, P(S). Equation 3 gives this ration.

)()(

SPNPopt

After finding the value of βcurrent and βopt, the pair are compared to determine whether an observer is following a risky or conservative strategy. If βcurrent is greater than the value of βopt, then Xc is positioned more to the right resulting in less false alarms and more misses. According to SDT litera-ture, observers with such a mental-cutoff require more evidence to say ‘yes, a part is defective or a tu-mor exists’, i.e., they’re less likely to say ‘yes’. Un-der SDT, this is considered a conservative strategy for operators to adopt because of the consequences and actions triggered after a false alarm results, such as rejecting a non-defective product or performing unnecessary medical procedures. When βcurrent is less than βopt, then Xc is positioned more to the left resulting in more false alarms and less misses. Based on SDT, this indicates that the observer needs considerably less evidence to say ‘yes, a part is defective or a tumor exists’, i.e., they’re more likely or quick to say ‘yes’. Therefore, this strategy is considered a risky strategy. Consequently, under traditional SDT, the rules for βcurrent and βopt are as follows:

βcurrent > βopt; strategy is conservative (4)

βcurrent < βopt ; strategy is risky (5)

Another important measure of an observer’s perfor-mance in signal detection tasks is sensitivity to the signal and the noise. This is measured by the degree of separation between the means of the two distribu-tions shown in Figure 2 and is denoted as d’. A high value of d’ indicate a high degree of separation and, thus, high observer sensitivity. Data from numerous tasks indicate that d’ ranges in value from 0.5 to 2.0 (Wickens 1992).

The value of d’ is determined by adding the two values z1 and z2 shown in Figure 2. z1 and z2 represent the value of the standard normal variable corresponding to the probability of a false alarm and probability of a miss, respectively. The values are readily available from standard tables. The applica-tion of SDT will be demonstrated using an example of a typical inspection process.

Example

A manufacturer produces DC motors using a process that generates 5% defectives. In response to increas-ing customer complaints, the manufacture instituted a final inspection system that finds 80% of defec-tive motors at the expense of falsely rejecting 1% of good motors. Determine the sensitivity of the opera-tor and the strategy adopted.

Solution

From the information given; the following probabili-ties can be deduced:

P (Noise) = P (product is not defective) = 95%P (Signal) = P (product is defective) = 0.05P (Hit) = 0.80 P (Miss) = 1-P (Hit) = 0.20P (FA) = 0.01 P (CR) = 1- P (FA) = 0.99

(3)

(2)



The following table (matrix) represents the inspec-tion process with its possible outcomes.

State of the worldSignal(Defective Product)

Noise(Good Product)

Response(Is motor defective?)

Yes HIT = 80% FALSE ALARM = 1%

No MISS = 20% CORRECT REJEC TION = 99%

Calculation of the sensitivity, i.e., the value of d’ in-volves the standard normal values z1 and z2. Using the P (FA) and P (Miss), the values of z1 and z2 are:

z1 = 2.326 z2 = 0.842 and ∴d’ = z1 + z2∴d’ = 2.376 + 0.842 = 3.168.

This indicates a high degree of separation between the signal and noise distributions, i.e., the inspector has high sensitivity.

As indicated by Equation 2, calculating βcurrent requires the determination of P (X/S) and P (X/N). However, Figure 2 indicated the following:

P (X/S) = Ordinate corresponding to z2P (X/N) = Ordinate corresponding to z1Using the tables,Ordinate corresponding to z2 = 0.28Ordinate corresponding to z1 = 0.027∴βcurrent = 0.28 / 0.027 = 10.37√

Using Equation 3, βopt is easily calculated:

)()(

SPNPopt

05.095.0

opt = 19

Because βcurrent < βopt, hence the strategy is con-sidered risky. This means that the inspector’s cut-off

level, Xc, is positioned more to the left, i.e., cuts more in the signal distribution.

Similar to a detection task in other industries, in con-struction, workers are expected to identify whether the condition they are working in is safe. In SDT, the state of the world is represented by a signal and noise. From a construction safety standpoint, the state of the world is “Safe condition” and “Unsafe Condition” which are akin to the noise and signal states of SDT.

On the one hand, a worker faced with a “safe” condi-tion and asked whether condition is unsafe has one of two possible responses, namely, ‘Yes’ condition is unsafe (false alarm), or ‘no’ condition is safe (correct rejection). On the other hand, a worker faced with an “unsafe” condition and asked whether condition is unsafe has one of two possible responses, namely, ‘Yes’ condition is unsafe (Hit), or ‘no’ condition is safe (Miss). Table 2 shows the SDT matrix for these scenarios.

The ideal scenario for a given number of safe and unsafe conditions is for a worker to correctly identi-fy them. This may be true of some workers but cer-tainly not of all workers. Some workers will incor-rectly consider a condition as safe while it is unsafe, and vice verse. Signal detection theory allows the determination of the sensitivity of workers

State of the worldUnsafe

Condition(Signal)

Safe Condition

(Noise)

Response(Is condition

unsafe?)

Yes HIT FALSE ALARMNo MISS CORRECT

REJECTION

Table 2: The SDT matrix for detection of unsafe conditions in construction

to unsafe conditions as well as their inclination (bias) to consider a situation as unsafe while it is not.As explained before, assessment of the worker sensitivity to unsafe and safe conditions as well as the inclination to associate a condition with a safe or unsafe, can be found using the SDT parameters d’



and beta (current and optimum). High values of d’ indicate high sensitivity in differentiating between safe and unsafe conditions. Conversely, low value of d’ indicates that a worker needs more training to better differentiate between safe and unsafe condi-tions.

Regardless of the value of d’, the mental cutoff used by a worker to decide the state of a condition is given by the value of βcurrent with respect to βopt. However, considering the implementation of SDT in construction, interpreting the values of βcurrent and βopt requires a modification.

As discussed before, if the value of βcurrent is greater than βopt, then less false alarms and more misses will result and this strategy is considered con-servative. Similarly, under SDT, a value of βcurrent smaller than βopt, is considered a risky strategy because more false alarms and fewer misses result. However, in construction, the cost of a miss could result in a fatality or a serious injury. Therefore, in construction, it is a more conservative strategy to have more false alarms and fewer misses, and a risky strategy to have fewer false alarms and more misses. Hence, the comparison rules for βcurrent and βopt in construction would be:

βcurrent > βopt ; strategy is risky (6)

βcurrent < βopt ; strategy is conservative (7)

Undertaking the assessment of worker sensitivity to unsafe and safe conditions as well as the inclina-tion to associate a condition with a safe or unsafe requires the determination of SDT responses (hit, miss, false alarm, and correct rejection) to a number of safe and unsafe conditions. Assessing worker performance in detecting unsafe and safe condition in real time is both dangerous and infeasible. The alternative is to design a survey (questionnaire) that places the worker in hypothetical safe and unsafe conditions and ask the worker to identify whether the condition is safe or unsafe ( see Patel and Abdel-hamid 2004, Narang and Abdelhamid 2006).

Table 3, shows sample survey questions to ironwork-ers based on OSHA fall protection standards.. For each questions, the worker chooses from one of two

responses: a) ‘Yes’ I would work (or continue work-ing) in such condition or b) ‘No’ I will not work (or continue working) in such condition. This question-naire would help determine how a worker would react or respond if he/she encounters typical safe or unsafe conditions.

Based on the responses, the number of hits, misses, false alarms, and correct rejections are determined and converted to probabilities. This will facilitate the determination of the sensitivity and the risk orientation of the workers towards unsafe conditions. To illustrate how a response will be mapped to a hit, miss, false alarm, or a correct rejection, a sample response is shown in Table 4. For example, consider question 5, which depicts a safe condition. The worker response to this question was “No I would not work (or continue working) in such condition”, i.e., the worker incorrectly considered the condition as unsafe – a “false alarm”. Similarly, in the case of question 9, which depicts an unsafe condition, the worker response was “Yes I would work (continue working) in such condition”, i.e., the worker incor-rectly considered the condition safe – a “miss”. The rest of the questions are analyzed the same way.

Table 4. Example worker responses to the surveyTo illustrate further the type of analysis that would be performed based on survey responses, it is as-sumed that results of a 30-question survey, with 6 unsafe and 24 safe conditions, were as shown in Table 5.

Table 5: Sample survey analysis resultsState of the world

Unsafe Condition(Signal)

Safe Condition

(Noise)

Response(Is condition unsafe?)

Yes HIT = 4 FALSE ALARM = 3

No MISS = 2 CORRECT REJECTION

Note that:P (Noise) = P (safe conditions) = 24/30 = 80%P (Signal) = P (unsafe condition) = 6/30 = 20%.P (Hit) = 4/6 P (Miss) = 1-P (Hit) = 2/6P (FA) = 3/24 P (CR) = 1- P (FA) = 21/24



Table 3: Questionnaire Example

INTERVIEW QUESTIONSResponse

Yes I would work (or continue working) in such condition

No I would not work (or continue working) in such condition

1) Would you work on a 130 ft high Coupler Scaffolds designed by you companies foreman?2) Would you work on the scaffold 8 feet above the lower level without fall protection?3) Would you work on the scaffold 10 feet above the lower level with-out fall protection?4) What will you do if you are asked to work on the 5th floor of the building whose bolting or welding for the 1st floor has just started?5) What would you do if you were asked to work on the 13th floor of the structure with permanent floor installed till the 6th floor of the structure?6) Would you continue working on 3,500 sq.ft decking when u know it is unsecured.7) Would you continue working on 3,000 sq.ft decking when u know it is unsecured.8) What would you do, when you are working on flat roof, which is in renovation, and a 50-inch square opening was created?9) Would you work on a 37 feet high platform surrounded by 3 ½ feet high steel railing with 8-foot diameter vent stack running vertically through the center of platform, with 12 inch annular space between the vent stack and the platform.

Table 4: Example Worker Responses to the Survey

INTERVIEW QUESTIONSResponse

Yes I would work (or continue working) in such condition

No I would not work (or continue working) in such condition

1) Would you work on a 130 ft high Coupler Scaffolds designed by you companies foreman? 2) Would you work on the scaffold 8 feet above the lower level without fall protection? 3) Would you work on the scaffold 10 feet above the lower level with-out fall protection? 4) What will you do if you are asked to work on the 5th floor of the building whose bolting or welding for the 1st floor has just started? 5) What would you do if you were asked to work on the 13th floor of the structure with permanent floor installed till the 6th floor of the structure?

6) Would you continue working on 3,500 sq.ft decking when u know it is unsecured. 7) Would you continue working on 3,000 sq.ft decking when u know it is unsecured. 8) What would you do, when you are working on flat roof, which is in renovation, and a 50-inch square opening was created? 9) Would you work on a 37 feet high platform surrounded by 3 ½ feet high steel railing with 8-foot diameter vent stack running vertically through the center of platform, with 12 inch annular space between the vent stack and the platform.



Calculation of the sensitivity, i.e., the value of d’ in-volves the standard normal values z1 and z2. Using the P (FA) and P (Miss), the values of z1 and z2 are:z1 = 1.21z2 = 0.440and ∵d’ = z1 + z2∴d’ = 1.21 + 0.440 = 1.65.

This indicates a moderate degree of separation between the signal and noise distributions, i.e., the worker has moderate sensitivity. Ordinate corresponding to z2 = 0.362 Ordinate corresponding to z1 = 0.194Using Equation 2:

βcurrent = 0.362 / 0.194 = 1.865

Using Equation 3:

βopt = P (Noise) / P (Signal) = 0.8/0.2 = 4

Clearly βcurrent < βopt which, based on rules 6 and 7, indicates a conservative strategy. Despite that with this strategy the worker will have more false alarms, fewer misses will result.

The above example illustrates how the sensitivity and risk orientation of a worker can be determined. This information sets a benchmark against which the effectiveness of new training can be assessed. Es-sentially, this information would make it possible to determine if a worker’s sensitivity and risk orienta-tion towards safe and unsafe conditions increased, decreased, or remained unchanged. Ultimately, the use of SDT will result in increasing workers abili-ties to judge the boundary beyond which work is no longer safe.

OVER THE EDGE

Rasmussen’s model recognizes that during work operations, workers, knowingly or unknowingly, will inevitably drift to the hazard zone or near the edge. Under such circumstances, mistakes may oc-

cur resulting in loss of control. Therefore, it be-comes a safety imperative to train workers to regain lost control and to also provide them with damage control measures. This is akin to poke-yoke (mis-take-proofing) techniques implemented by Toyota to ensure that if human error occurs then recovery will be instantaneous.

Most safety programs and professionals believe that currently available personal protective equipment such as fall arrest systems, safety nets, gloves, hard hats, etc. are accident prevention tools or in Rasmus-sen’s terms that they can enlarge the safety zone. However, the term prevention has been categorically confused with the term damage control. Damage control can be defined as some form of a suppressing measure to an accident outcome, such as fire sprin-kler systems, air bags in modern cars, fall protection gear, machine safeguards, among many others. This is not to imply that damage control measures should not be incorporated in various systems, tools, ma-chines, etc. However, it should be recognized that these measures do not prevent accidents or reoccur-rence of accidents, rather they decrease the damage caused by an accident. Stated differently, PPEs should not be considered as measures to enlarge the safe zone.

Virtual reality and simulation techniques could also be used to train workers on regaining lost control. The technology is available but has not been adapted to construction needs yet.

Another proactive systems approach to improving safety, termed Resilience Engineering, provides a means to assess the adaptability of organizations in relation to the production demands encountered. Resilience engineering is grounded in sociotechnical systems (STS) theory. STS theory was developed at the Tavistock Institute in London in the late 1940’s to relate social and psychological sciences to the needs of society. Researchers at the Institute studied coal mining production methods in the early 1950’s to compare pre-mechanization methods; specifi-



cally the “shortwall method” approach that relies on multi-skilled teams of autonomous workers to the mechanized approach, the “longwall method” of coal mining, which was Tayloristic in nature, highly structured, and highly mechanized (Trist, 1951). Inherent in resilience is the notion of adaptability to a perturbation. Woods (2007) defines resilience in a broad sense as the ability of a system to “…handle disruptions and variations that fall outside of the base mechanisms / model for being adaptive as defined in that system.” However, as Woods (2007) points out, all systems adapt even though the adap-tation may be slow and difficult to recognize. Of course, adaptability is finite and sometimes trees, humans, and organizations reach a breaking point.Resilience engineering is concerned with how orga-nizations manage unexpected events and how people in these organizations become prepared to cope with surprises (i.e. events that fall outside of planned for events and that are unforeseeable). It views resil-ience as a systems property and moves away from the linear cause and effect thinking that is prevalent when analyzing construction accidents. It looks to organizational factors instead of human errors or machine malfunction as conditional contributors to accidents (Hollnagel et al 2006).

In resilience engineering, safety is not viewed as a system property but “…as something a system or organization does, rather than something an organi-zation has” (Hollnagel et al 2006). In other words, safety is not something placed into a system through rules and standards that will remain in place but rath-er safety is a reflection of how a system performs. This perspective on safety means that it should not be demonstrated by the absence of accidents from a system, but rather by the existence of certain system characteristics.

System resiliency should not be confused with sys-tem reliability, which is often used as a measure of safety. A system that is reliable and has a probability below which failure will occur is not resilient un-less it has the ability to recover from infrequent and

unexpected perturbations and disruptions to expected working conditions (Hollnagel et al 2006). More-over, system resilience cannot be simply integrated in using more procedures, guidelines, personal protective equipment, and barriers. As advocated in Lean Construction, system resilience is achieved through continuous monitoring of system perfor-mance and “how things are done”. Hollnagel (2004) states that resilience is “tantamount to coping with complexity, and to the ability to retain control.”In today’s construction industry firms are generally well-aware of the demands that may be imposed upon them in the course of normal working condi-tion. For instance, construction schedules typically include contingences for inclement weather that may delay production activities (Hinzie, 2008). The contingency is typically derived from data obtained from weather authorities, dictated by the owner, or based on the best guess of the construc-tion manager(s). By including weather as a planned event the project team has (at least in theory) antici-pated a perturbation to the production schedule. If the duration of the weather event occurs within the anticipated timeframe no extraordinary efforts will have to be extended to meet the demand to finish on time. In other words, the organization is adapt-able to the perturbation – in this range (Schafer et al 2008).

Resilience engineering is concerned with the behav-ior and reaction of the organization as it moves from this anticipated working range (i.e. an accounted for disruption to work) to a state outside of the normal working range (e.g. a 100-year rain). Now the firm must pick up the tempo of work and increase capac-ity to meet this new demand. The firm must stretch existing resources to meet the new demands as it exits the normal working zone and ramps up produc-tion to ‘make-up’ for lost time. Research has shown that when this situation occurs firms are apt to sacri-fice safety for production concerns and that individu-als, especially those removed from the workface (i.e. higher level managers) aren’t aware that they are operating outside of the bounds of built-in adaptabil-



ity and are jeopardizing safety (Woods and Wreathall 2007).

Some have described resilience engineering as a paradigm shift in the ‘Kuhnian sense” (Woods, 2006), referring to Thomas Kuhn’s well-known “The Structure of Scientific Revolutions” published in 1962. By proposing a new outlook and vocabulary for system safety it may well be, as Kuhn stated “A revolution occurs when a community changes its lexicon”. However, in the course of scientific (or any type of) discovery there are those who, early on, become embroiled in discussions concerning the use of a phrase such as ‘paradigm shift’ and enmeshed in conducting assessments of the validity of the use of the term instead of seeing the possibilities in the new paradigm.

A resilient system is able to maintain control when faced with disruptions and perturbations in the form of unexpected events. A system is said to be in control if it is able to mitigate or eliminate unwanted endogenous or exogenous variability (Hollnagel and Woods, 2006). This definition harkens back to resilient engineering’s roots in CSE. CSE is invested in examining the internal and external demands placed on a system, especially with respect to time and in the context where there are consequences at stake for those groups and individual involved (Hollnagel and Woods, 2005). In many instances the unexpected event is a consequence of lost sys-tem control, although this is clearly not always the case. Some events are just beyond the control of the system operators. In other instances the unexpected event is the cause of the loss of control. The distinc-tion here is that in the temporal sequencing of the events ‘knowing that control has been lost’ and an ‘accident’. It is more valuable to system operators to know when control is going to be lost ( i.e. when unexpected events are likely to occur) rather than that control has been lost (Woods, 2006). In the lat-ter, the operator is in a reactive mode, in the former the operator is in a proactive mode and can avoid or mitigate danger.

Resilience engineering recognizes that despite even outstanding planning efforts, performance conditions are always underspecified (Hollnagel, 2008; Ras-mussen, 1983). The front-line worker, in both a JCS and non-JCS situation, must always make adjust-ments in the course of operations given the context of changing environmental conditions and the inten-sity of demands. In other words, there will always be performance variability on the part of the JCS due to the need to respond to demands imposed on the system. Resilience engineering aims to dampen the variability that may contribute to adverse events and to amplify the variability that leads to positive outcomes (Hollnagel, 2008).

Hollnagel’s (2004) Efficiency-Thoroughness Trade-Off (ETTO) Principle captures an aspect of this notion. In attempts to optimize performance goals people work to be as thorough as they can be (i.e. follow the rules) given the prevailing conditions. However, there is also pressure to be efficient. People and organizations that are not efficient are, respectively, unemployed and unprofitable (or bank-rupt). Also, those who are not sufficiently thorough possibly endanger safety and may also cease to exist economically. One reason thoroughness is some-times shunted is that in the quest to optimize people skip seemingly unnecessary steps in work tasks. In construction this is manifest in the phrase “We have always done it this way” when discussing field op-erations with, for example, a subcontractor, when in fact, field conditions may necessitate that operations be revised from what has always been done. The shortcut, (e.g. always doing it this way) is the norm in work rather than the exception given that work en-vironments are relatively stable places and that acci-dents are rare events (Hollnagel, 2004). People may take certain aspects of their work for granted and skip seemingly inefficient steps. This can sometimes lead to an accident. An example might be to neglect to “tie-off” a ladder to save time. Excuses can range from “We never have ladders slip” to “I was only go-ing on the roof for a minute.” In the event of an ac-



cident incurred costs can wipe out efficiency gains. Closely allied with the ETTO Principle is the out-look that the cause of failure is the temporary in-ability to effectively cope with complexity under demanding conditions. A situation that resilience engineering addresses that is closely allied to com-plexity is the all too common production and effi-ciency tensions inherent in construction. Research has shown that workers implicitly choose produc-tion over safety concerns when a trade-off is avail-able and therefore act in a riskier manner than they normally would (Woods, 2004). An example in construction would involve knowing when to relax production pressures by, for example, reducing over-time hours, adding additional crews, subcontracting extra work in critical periods, or simply knowing when to slow down production so that safety is not endangered.

Resilience engineering has multi-faceted uses. One is to provide indicators that allow firms to recog-nize when they are moving to an area outside of its normal working capacity and into the area where production demands impinge upon safety so that an intervention can be made to stay out of dangerous working situations. Here the firm is better prepared and not surprised by perturbations to the system, re-sources that add capacity to the firm are located and at the ready if needed. Another goal of Resilience Engineering is to help the organization to become, as expressed by Woods and Wreathall (2008), well-cal-ibrated. A well-calibrated firm knows, adaptability-wise, when it is in the normal working zone, when it is changing, and knows its limits, thus allowing it to invest in extra capacity or other means of adaptation when extraordinary events are encountered. In summary, resilience engineering is concerned with system control of performance variability rather than constraining it with the goal of maintain stabil-ity and control. Hollnagel (2008) defines the essen-tial attributes of a resilient system as the ability to cope with the actual, the critical, the potential, and the factual. Examples of how a construction firm might demonstrate resilience are presented below with each explanation when appropriate. Schafer at

al (2008) provides propositions for ‎implementing Resilience Engineering in construction settings and offers pointers to future ‎research.

CONCLUSION

This paper argued that the relation between construc-tion safety and ergonomics and Lean Construction is a reciprocal one. Notwithstanding the reciprocal relation between Lean Construction and safety and ergonomics, there is still a need to promote occupa-tional safety and ergonomics concepts under Lean Construction.Traditional and contemporary research efforts and approaches in construction occupational safety and health were reviewed. Many accident causation models are linear, such as Heinrich’s popular Dom-ino Model and Reason’s Swiss Cheese Model (Hol-lnagel, 2004). As a result, numerous risk assessment techniques, such as fault-tree analysis, have been developed to be sequential in nature and focus on the failure of a single component of the system, the physical structure of devices such as fault-trees and event trees bolster the idea that accident causation is linear. Additionally, regulatory standards are often just bolstered incrementally to cover the latest crack in the regulation as exposed by the latest accident. In existing safety approaches and schemes safety is often measured by the absence of accidents and di-sasters. Safety is often managed by error tabulation and probabilities, such as in setting goals to reduce falls by 33%.

Despite the progress achieved by using these ap-proaches, improvements in construction safety and health appear to have reached a point of diminishing returns. To change this situation, the paper intro-duced a new accident causation paradigm based on a model developed by Jens Rasmussen (1994), and its extension to Resilience Engineering as proposed in Schafer et al (2008 and 2009).

As mentioned before, Rasmussen’s model departs from conventional thinking that either workers



or management are at fault for accidents and that through telling each side what to do and how to perform, safety will be improved. The model states that organizational and individual pressures push people to work in hazardous situations and that these pressures cannot be ignored nor magically relieved through worker training and enforcement of safety regulations. The model indicates that workers will inevitably choose to work or find themselves forced to work in the hazard zone. Therefore, management must adopt a social norms marketing approach and train workers on how to work with hazards and how to regain control once it’s lost.Implementation of the Rasmussen-Resilience model will require a concerted effort on the part of both academia and industry. Future research should be guided by the strategy recommended in Howell et al (2002), which, at the expense of being repetitious, is reproduced here:

1. IN THE SAFE ZONE: Establish methods and techniques to enlarge the safe zone.2. AT THE EDGE: Train workers on the iden tification of safe and unsafe conditions. And once in an unsafe condition, workers should be trained on how to recover from errors.3. OVER THE EDGE: People will inevitably make mistakes resulting in loss of control. Hence, measures should be in place to limit the effect of this loss (safety net).

A brief discussion on Resilience Engineering was provided given that it offers a new perspective on system safety. We see Resilience Engineering as an excellent framework to deal with providing graceful transitions between the normal zone to the extraor-dinary zone, and a possible zone of extreme restruc-turing (the zone of loss of control in the Rasmussen model). The idea is to identify sources of resilience that workers and organizations possess that can counteract the challenges and demands placed on the systems both externally and internally and to continuously monitor the systems model of how it creates safety and identify where safety is or may be

endangered (Schafer et al 2009). In general, Resil-ience Engineering has evolved as a way to overcome the limitations of existing accident analysis and risk assessment that is used to manage safety. Resilience engineering is a proposal to the research community to cease from relying exclusively on hindsight in ex-plaining what has happened to exploring the sources of resilience that prevent and mitigate accidents. The hindsight view can cause accident investigators to “…think that a sequence of events inevitably led to an outcome…underestimate[ing] the uncertainty people faced at the time…” and to “see a sequence of events as linear, leading nicely and uninterrupt-edly to the outcome we now know about…” as well as to “…oversimplify causality” (Dekker, 2006). As food for thought, not directives, this paper dis-cussed tools and techniques that could be used to implement the above recommendations. Other re-searchers may choose to further develop them or find other methods from other disciplines or industries. Regardless of the approach, we believe that the Ras-mussen model and Resilience Engineering should be the main source of inspiration for new tools and ideas aimed at improving construction safety and health.

REFERENCES

Abdelhamid, T. S. (1999). Measuring and Evaluat-ing Physiological Demands of Construction Work. Ph.D. Dissertation, University of Michigan, Ann Arbor, Michigan.

Abdelhamid, T. S. (2003). “Six-Sigma in Lean Con-struction Systems: Opportunities and Challenges”. Proceedings of the 11th Annual Conference for Lean Construction, 22-24 July 2003, Blacksburg, Virginia, 65-83.

Abdelhamid, T. S., and Everett (2000). “Identifying root causes of construction accidents”. Journal of Construction Engineering and Management. ASCE, 52-59.



Abdelhamid, T. S., and Everett, J. G. (1999). Physi-ological demands of concrete slab placing and finish-ing work. Journal of Construction Engineering and Management, ASCE, 125(1), 47-52.

Abdelhamid, T. S., El-Gafy, M.A., and Salem, O. (2008). “Lean Construction: Fundamentals And Principles.” The American Professional Constructor Journal, American Institute of Constructors – Fall 2008.

Accident facts. (1997, 2002, 2007). National Safety Council, Itasca, Ill.Adams, E. (1976). “Accident Causation and the Management System.” Professional Safety, October, 26-29.

Alkov, D.(1972). “The Life Change Unit and Acci-dent Behavior.” Lifeline, September-October 1972.Andersson R., Johansson, B., and Linden, K. (1978). “Development of a model for research on occupa-tional accidents.” Journal of Occupational Acci-dents, Vol. 1, 341-352.

Armstrong, T.A. (1993). Analysis and design of jobs for control of upper limb musculoskelatal disorders. University of Michigan, Ann Arbor, MIAstrand, P., and Rodahl, K. (1986). Text book of work physiology, physiological bases of exercise. McGraw-Hill, New York.

Ball, L. in R. Stein (1973). Recording and Cataloging Hazards Information. NASA, Huntsville, Ala.

Ballard, G. (1999). Improving Work Flow Reli-ability. Proceedings of the 7th Conference of the International Group for Lean Construction, Berkeley, California, USA, 26-28 July 1999.

Ballard, G. (2000). Lean Project Delivery System. White Paper #8, Lean Construction Institute (www.leanconstruction.org).

Ballard, G. and Howell, G. (1998). What Kind of Production Is Construction? Proceedings of the 6th

Conference of the International Group for Lean Con-struction, Guaruja, Brazil, August 13-15.

Becker, P., Akladios, M, Fullen, M. (1999). “Fall prevention in construction by organizational inter-vention”. Professional Safety. 917-924

Benner, L. (1975). “Accident Investigation. Mul-tilinear Events Sequencing Methods.” Journal of Occupational Accidents, Vol. 7 (2), 67-73.

Benner, L. (1979). “Accident Theories and Their Implications for Research.” Quarterly Journal, American Association for Automotive Medicine, Vol. 1, No. 1.

Benner, L. (1979). Four Accident Investigation Games, Lufred Industries, Inc., Oakton, VA.Bernold L. E., Lorence, S. J., and Davis, L. M. (2001). “Technological intervention to eliminate back injury risks for nailing.” Journal of Construc-tion Engineering and Management, ASCE, 127(3), 245-250.

Bertelsen, S. and Koskela, L. (2002). “Managing the three aspects of production in construction”. Pro-ceedings of the 10th International Group for Lean Construction Conference, Gramado, Brazil, August 5-8.

Bird, F. (1974). “Management Guide to Loss Con-trol.” Institute Press, Atlanta, 1974.

Bird, F. and O’Shell, H. (1972). “Principles of Loss Control.” International Safety Academy, Houston.

Brouha, L. (1967). Physiology in industry. Pergam-on Press, New York.

Brown, I. D. (1995). “Accident reporting and analy-sis.” In Wilson, J. R., and Corlett, E. N. (1995). Evaluation of Human Work. Taylor and Francis, London, England.



Buchholz, B., Paquet, V., Punnet, L, Lee, D., and Moir, S. (1996). “PATH: A work sampling-based approach to ergonomics job analysis for construction and other non-repetitive work.” Applied Ergonom-ics, 27(3): 177-187.

Chaffin, D. B., Andersson, G. B., Martin, B. J., and Andersson, G. (1999). Occupational biomechanics. Wiley, New York.

Colley, A. M., and Beech, J. R. (1989). “Acquisition and Performance “ of Cognitive Skills.” Wiley, New York.

Congress of the United States (1985). “Preventing Illness and Injury in the Workplace.” Office of Tech-nology Assessment, Washington, D.C.

Cooper, K. and Volard, S., V. (1978). The influence of the individual on industrial accidents. Accident Prevention, Toronto.

CoVan, J. (1995). “Safety Engineering.” Wiley, New York.

CPWR (1998). Research Project Guidelines. CPWR, Washington, D.C.

CSC (2009). Construction Safety Council; http://www.buildsafe.org/about.htm visited 2/15/2009.Culver, C., and Connolly, C. (1994), “Prevent Fatal Falls in Construction”. Safety and Health , NSC, Itasca, IL

Culver, C., Florczak, G., Castell, R. Jr., Pelton, G., and Connolly, C. (November 1990). Analysis of Construction Fatalities- The OSHA Data 1985-1988. Occupational Safety and Health Administration (OSHA) Report, USA.

Culver, C., Hardesty, C., and Anderson, F. (1993). “Ironworker Fatalities in construction”. Occupation-al hazard, 47 pp.

Culver, C., Marshall, M., and Connolly, C. (April 1992). Construction Accidents: The Workers’ Compensation Data Base 1985-1988. Occupational Safety and Health Administration (OSHA) Report, U.S.A.

Davies, V.J. and Tomasin, K. (1990). Construction Safety Handbook. Thomas Telford, London, 1990.

DeJoy, M. (1990). “ Towards comprehensive human factors model of workplace accident causation.” Professional Safety, 35(5), 11-16.

Dekkar, S. (2006). The Field Guide to Understand-ing Human Error. Ashgate, Aldershot, UK.

Douglas, H., and Crowe, J. (1976). Effective Loss Prevention. Industrial Accident Prevention Associa-tion, Toronto.

Drury, C. G. (1983). “Human Factors in Consumer Product Investigation.” Human Factors, vol. 25, No. 3, 329-342.

Dwyer, T. And Raftery, AE. . (1991). “Industrial Ac-cidents Are Produced By Social-Relations Of Work - A Sociological Theory Of Industrial Accidents.” Applied Ergonomics, vol. 22, No. 3, 167-178

Edwards, M. (1981). “The Design of an Accident Investigation Procedure.” Applied Ergonomics, vol. 12, No. 2, 111-116.

Everett, J. G. (1992). Let’s pass on Mass Produc-tion. Construction Business Review (CBR), 2(2), 13-14.

Everett, J. G. (1999). “Overexertion Injuries in Con-struction”. Journal of Construction Engineering and Management, ASCE, 125(2), 109-114-.

Everett, J. G. and Thompson, W. S. (1995). “Experi-ence Modification Rating For Workers’ Compensa-tion Insurance.” Journal of Construction Engineer-ing and Management, Vol. 121, No. 1, 66-79.



Everett, J. G., and Kelly, D. L. (1998). “Drywall Joint Finishing: Productivity and Ergonomics”. Journal of Construction Engineering and Manage-ment, ASCE, 124(5), 347-353.

Fedearl Aviation Administration (1998). Crew Re-source Management Training. Advisory Circle No: 120-51C (available online at http://www.crm-devel.org)

Ferry, T. S. (1988). Modern Accident Investigation and Analysis. Wiley, New York

Fillmore, D. L. , and Cornelison, J. D. (1986). Mini-MORT Presentation Guide. Wiley, New York.

Firenze, R. (1971). “Hazard Control.” National Safety News, August.

Fraser, T. M. (1989). The Worker At Work. Taylor and Francis, New York.

Fredericks, T., Abduayyeh, O., Palmquist, M., and

Torres, H. N. (2002). “Mechanical contracting safety issues”. Journal Of Construction Engineering And Management. 128(2), 186-193

Friend, M. A., Kohn, J. P. (1992). “A Behavioral Approach To Accident Prevention.” Occupational Hazards, 54(10), 112-115.

Fullman, J. B. (1984). Construction safety, security, and loss prevention. Wiley, New York.

Gambatese, J. A. (2000). ‘‘Safety in a designer’s hands.’’ Civil Engineering, ASCE, June, 56–59.

Gausch, J. (1972). “Safety and Decision-Making Tables.” Journal of the ASSE, November.

Gausch, J. (1974). “Value Engineering and Decision Making.” Journal of the ASSE, May. .

Goldsmith, D. (1987). Safety Management in Con-struction and Industry. McGraw-Hill, New York.

Groeneweg, J. (1998). Controlling the Controllable. DSWO Press, The Netherlands, 1998.

Haddon, W. (1968). “The Changing Approach to Epidemiology, Prevention and Amelioration of Trauma.” American Journal of Public Health, Vol. 58 (8), 1413-1438.

Hale, A. R., and Hale, M. (1970). “Accidents in perspective.” Occupational Psychology, Vol. 44, 115-121.

Heath, ED. (1991). “ Identifying Those Worker Populations That Are At Higher Levels Of Risk.” American Industrial Hygiene Association Journal, Vol. 52, No. 4, A211-A212

Heinrich, H. W., Peterson, D., and Roos, N. (1980). Industrial Accident Prevention. McGraw-Hill, New York.

Helander, M. G. (1991). “Safety hazards and mo-tivation for safe work in the construction industry”. International Journal of Industrial Ergonomics, vol. 8, no. 4, 205-223.

Helmreich, R.L., Klinect, J.R., and Wilhelm, J.A. (1999). Models of threat, error, and CRM in flight operations. In Proceedings of the Tenth International Symposium on Aviation Psychology (pp. 677-682). Columbus, OH: The Ohio State University.

Hendrick, H. W. and Kleiner, B. M. (2001). Mac-roergonomics: An Introduction to Work System Design. Human Factors and Ergonomic Society. 310/394-1811

Hendrick, K., and Benner, L. (1987). Investigating Accidents With STEP. Dekker, New York.



Hinze, J. 1978. “Turnover, new workers, and safety.” Journal of the Construction Division, 104(4), pp. 409-417.

Hinze, J. 1981 “Human Aspects of Construction Safety”, Journal of the Construction Division, 107(1), pp. 61-72.

Hinze, J. (1996). The distraction theory of accident causation.” Proc.Int. Conf. On Implementation of Safety and Health on Constr. Sites, CIB Working Commission W99: Safety and Health on Construc-tion Sites, L. M. Alvez Diaz and R. J. Coble, eds, Balkema, Rotterdam, The Netherlands, pp. 357-384.

Hinze, J. and Wiegand, F. (1992). “The Role of De-signers in Construction Worker Safety.” Journal of Construction Engineering and Management, ASCE, 118(4), 677-684.

Hinze, J. W. (1997). Construction Safety. Prentice Hall, Upper Saddle River, New Jersey.

Hinze, J. W. (1997). Construction Safety. Prentice Hall, Upper Saddle River, New Jersey.

Hinze, J., and Gordon, F. 1979. “Supervisor-worker relationship affects injury rate.” Journal of the Con-struction Division, 105(3), pp. 253-261.

Hinze, J., and Parker, H. W. 1978. “Safety, produc-tivity and job pressures.” Journal of the Construction Division, ASCE, 104(1), pp. 27-35.

Hinze, J., and Wilson, G. (2000). “ Moving toward a zero injury objective”. Journal of construction engi-neering and management, 126(5), 399-403.

Hinzie, J.W. (2008). Construction Planning and Scheduling, 3rd Ed. Pearson, Columbus, Ohio.

Hollnagel, E. (2004). Barriers and Accident Preven-tion. Ashgate, Aldershot, UK.

Hollnagel, E., Woods, D, and Leveson, N. (2006). Resilience engineering: concepts and precepts. Al-dershot, England: Ashgate.

Hopp, W. J., and Spearman, M. L. (2000). Factory Physics. McGraw-Hill, New York.

Howell, G and Ballard, G (1994). Lean Production Theory: Moving Beyond ‘Can-Do. Proceedings of the 8th Conference of the International Group for Lean Construction, Santiago, Chile, September.

Howell, G. (1999). What Is Lean Construction. Proceedings of the 7th Conference of the Interna-tional Group for Lean Construction, Berkeley, Cali-fornia, USA, 26-28 July 1999.

Howell, G. A., Ballard, G., Abdelhamid, T. S., and Mitropoulos, P. (2002). “Working near the edge: a new approach to construction safety.” Proceedings of the 10th International Group for Lean Construc-tion Conference, Gramado, Brazil, August 5-8, 2002.

Hoyos, C. G., and Zimolong, B. (1988). Occupa-tional Safety and Accident Prevention, Behavioral Strategies and Methods. ElSevier, Amsterdam, Netherlands.

Ihara, S. (1993). “Information theory for continuous systems” World Scientific, River Edge, N.J.Injury facts (2002). National Safety Council, Itasca, Ill.

International Labor Organization (1983). Accident Prevention, A Workers’ Education Manual. Geneva, Switzerland.

Janaro, R. E. (1982). The development and imple-mentation of optimal multi-rest break scheduling models. Ph.D. Dissertation, Florida State University, Florida.




Jaselskis, E. J., Anderson, S. D., and Jeffrey S. R (1996). “Strategies for Achieving Excellence in Construction Safety Performance.” Journal of Construction Engineering and Management, ASCE, 122(1), pp. 61-70.

Johnson, H., Singh, A., and Young, F. (1998) “Fall protection analysis for worker on residential roofs”. Journal of construction engineering and manage-ment. ASCE, September/October ,418-428.Johnson, W. G. (1975). “MORT: The Management Oversight and Risk Tree.” Journal of Safety Re-search, vol. 7, No. 1, 4-15.

Kerr, W. (1957). “Complementary Theories of Safety Psychology.” Journal of Social Psychology, 45, 3-9.

Kirwan, B. (1994). A guide to practical human reli-ability assessment. McGraw-Hill, New York. Kivi, P. , and Mattila, M. (1991). “Analysis and improve-ment of work postures in the building industry: ap-plication of the OWAS method.” Applied Ergonom-ics, 22(1): 43-48.

Kjellen. U., and Larsson, T. R. (1981). “Investigat-ing accidents and reducing risks –– A dynamic ap-proach.” Journal of Occupational Accidents, Volume 3, Issue 2, May 1981, Pages 129-140.

Klinect, J.R., Wilhelm, J.A., and Helmreich, R.L. (1999). Threat and error management: Data from line operations safety audits. In Proceedings of the Tenth International Symposium on Aviation Psychol-ogy (pp. 683-688). Columbus, OH: The Ohio State University.

Klumb, P. L. (1995). “Cognitive failures and per-formance differences: validation studies of a Ger-man version of the cognitive failures questionnaire.” Ergonomics, vol. 38, no. 7, 1456-1467.

Knox, N. W., and R. W. Eicher (1976). MORT User’s Manual, ERDA, 76-45-4, SSDC-4, NTIS, Washington, D. C.

Koskela, L. (1993). “Lean production in construc-tion”. Proceedings of the 10th ISARC, Houston, Texas, May 24-26, 47-54.Krause, T. R., Hidley, J. H., and Lareau, W. (1984). “Behavioral science ap-plied to accident prevention.” Professional Safety, July, 21-27.

Krause, T. R., Russell, L. R. (1994). “The Behavior-Based Approach to Proactive Accident Investiga-tion.” Professional Safety, March, 22-28.La Bette, C. R. (1990). Innovative procedures in construction safety. Professional Safety, April, 35-39.

Lee, K. W., Tillman, FA., Higgins, JJ. (1988). “A Literature Survey Of The Human Reliability Compo-nent In A Man-Machine System. IEEE transactions on reliability, Vol. 37, No. 1, 24-34.Levitt, R. E. And Samelson, N. M. (1993). Con-struction safety management. McGraw-Hill, New York.

Liska, R. W., Goodle, D., and Sen, R. (1993). “Zero accident techniques,” Source document 86, Con-struction Industry Institute, Austin, Texas. MacCollum, D. V. (1990). “Time for change in construction safety.” Professional Safety, February, 17-20.

Mangum, G. (1995). “Documents: Construction Safety Put at Risk.” New Solutions, Fall p.84,86-90McCann, M. (2000). “ Cause of Roofer Death”. The center to Protect Workers’ Rights. Silver Spring, MD.

McClay, R. E. (1989). “Toward a more universal model of loss incident causation-Part I.” Profes-sional Safety, January, 15-20.

McClay, R. E. (1989). “Toward a more universal model of loss incident causation-Part II.” Profes-sional Safety, February, 34-39.



Miller, G. (1995). “Safety in the Workplace.” Congressional Record, Daily ed., p.H6381-6389, 27 June.

Mitropoulos, T., and Cupido, G. (2009). “The role of production and teamwork practices in construc-tion safety: A cognitive model and an empirical case study.” Journal of Safety Research, 40 (4), 265-275.

Mohamed S. (2002). “Safety Climate in Construc-tion Site Environments, Journal of Construction Engineering and Management, ASCE, 128(5), pp. 375-384.

Moosa, T., and Abdelhamid, T. (2005). “Production Planning Process in Residential Construction Using Lean and Six Sigma”. Construction Congress VII, American Society of Civil Engineers , April 5-7, 2005, San Diego, CA.

Narang, P. and Abdelhamid, T., S. (2006). “Quanti-fying Workers’ Hazard Identification Ability Using Fuzzy Signal Detection Theory”. Proceedings of the 14th Annual Conference of the International Group for Lean Construction, 25-27 July 2006, Santiago, Chile.

NIOSH ( 2000). A summary of Surveillance Find-ings and Investigative Case Report. Technical Report November 2000. U.S. Department Of Health and Human Services

Norman, D. A. (1981). “Categorization of action slips.” Psychological Review, 88, 1-14.

O’Hare, D., Wiggins, M., Batt, R., and Morrison, D. (1994). “Cognitive failure analysis for aircraft accident investigation.” Ergonomics, vol. 37, no 11, 1855-1869.

Parsons, T. J., Timothy, J. P., and Collins, J. W. (1986). “Safety analysis of high risk injury cat-egories within the roofing industry.” Professional Safety, June, 13-17.

Patel, B., and Abdelhamid, T. S. (2004). “Assess-ment of Structural Steel Workers’ Safety and Health Competencies Using Signal Detection Theory”. Proceedings of The Human Factors and Ergonomics Society’s 48th Annual Meeting, September 20-24, 2004, New Orleans, Louisiana, USA.

Petersen, D. (1971). Techniques of Safety Manage-ment. McGraw-Hill, New York.

Petersen, D. (1975). Safety Management-A Human Approach. Aloray, Englewood, N.J.

Petersen, D. (1980). Analyzing Safety Performance. Garland, New York.

Petersen, D. (1982). Human Error-Reduction and Safety Management. STPM Press, New York, N.Y.

Petersen, D. (1984). “The safety profession: Con-cepts and programs.” Professional Safety, Decem-ber, 26-29.

Peyton, R. X. and Rubio, T. C. (1991). Construc-tion Safety Practices and Principles. Van Nostrand Reinhold, New York.

Pollack, E., and Chowdhury, R. (2001). “Trends in Work-Related Death and Injury Rates among U.S Construction Workers, 1992-98”. The center to Pro-tect Workers’ Rights, Silver Spring, MD

Proctor, R. W., and Zandt, T. V. (1994). Human Factors in Simple and Complex Systems. Allyn and Bacon, Boston.

Rasmussen, J. (1997). “Risk Management in a Dynamic Society: A Modeling Problem.” Safety Sci-ence, Vol 27, No 2/3, pp.183-213.

Rasmussen, J., Pejtersen, A.M., Goodstein, L.P. (1994). Cognitive System Engineering. John Wiley and Sons, Inc. New York.



Reason, J. T. (1990). Human Error. Cambridge University Press, New York

Recht, J. (1970). System Safety Analysis: Error Rated and Cost. National Safety Council, Chicago.

Rietze, R. B. (1990). “Proactive construction man-agement.” Professional Safety, January, 14-16.

Rigby, L. V. (1970). “The nature of human error.” In Annual Technical Conference Transactions of the American Society for Quality Control, 475-566.

Rodgers, W. (1971). Introduction to System Safety Engineering. Wiley, New York.

Roland, H. E. and Moriarty, B. (1990). System Safety Engineering and Management. Wiley, New York.

Rook, L., Altman, J., and Swain, A. (1966). Human Error Quantification. Sandia Corporation, Albuqer-que, New Mexico.

Saurin, T.A.; Formoso, C.T.; Cambraia, F (2004). “A human error perspective of a safety planning and control model.” Proceedings of the 12th annual conference of the International Group for Lean Con-struction, Copenhagen, Denmark.

Saurin, T.A.; Formoso, C.T.; Cambraia, F (2007). “ An analysis of construction safety best practices from a cognitive systems engineering perspective.” Safety Science, doi:10.1016/j.ssci.2007.07.007.

Schafer, D.‎ ‎, Abdelhamid, T.S.‎ ‎, Mitropoulos, P.‎ ‎, and Howell, G.A (2008).‎ ‎ “‎Resilience Engineering: A New Paradigm For Safety In ‎Lean Construction Sys-tems.” Proceedings of the 16th Annual Conference of the International Group for Lean Construction, 16-18 July 2008, Manchester, United Kingdom.

Schafer, D., Abdelhamid, T.S., and Mitropoulos, P. (2009). “Resilience Engineering: A New Approach

For Safety Management.” Proceedings of the 2009 Construction Research Congress. .

Schlidge, B., Wakula, J., and Rohmert, W. (1997). “Ergonomic analysis of load on the knees at the tile-setter’s work.” Proceedings of the 13th Triennial Congress of the International Ergonomics Associa-tion, Tampere, Finland, June 29 – July 4, 1997.

Schneider, S. P. (1998). “Ergonomics in the Con-struction Industry”. In Karwoski, W., and Marras, W. S. (1998). The Occupational Ergonomics Hand-book. CRC Press, New York, New York.

Schneider, S. P., and Susi, P. (1994). “Ergonomics and Construction: A review of potential hazards in new construction.” Journal of the American Indus-trial Hygiene Association, AIHA, 55(7), 635-649

Shappell. S. A., Wiegmann, D. A. (2001). “Human factors analysis and classification system (HFACS)”. Proceedings of The Human Factors and Ergonomics Society’s 45th Annual Meeting. Minneapolis, Min-nesota.

Smillie, R. J., and Ayoub, M. A. (1976). “Accident causation theories: A simulation approach.” Journal of Occupational Accidents, Vol. 1, 47-68.

Spear, J. (2002) “Incident investigation: A problem solving process” Professional Safety. 25-30

Suchman, E. (1961). “A Conceptual Analysis of the Accident Phenomenon. Behavioral Approaches to Accident Research.” Association for the Aid of Crippled Children, New York.

Suraji, A., Duff, A. R., and Peckitt, S. J. (2001). “Development of Casual Model of Construction Ac-cident Causation”. Journal of Construction Engi-neering and Management, ASCE, 127(4), 337-344.

Surry, J. (1974). Industrial Accident Research. Uni-versity of Toronto, Toronto, 1974.



Swets, J. A. (1996). Signal detection theory and ROC analysis in psychology and diagnostics: col-lected papers. Lawrence Erlbaum Associates, Mah-wah, N.J.

Thomson, B. J. (1999). “International co-operation in hazardous materials accident prevention.” Journal of Loss Prevention in the Process Industries, Volume 12, Issue 3, May 1999, Pages 217-225

Toole, T. M. (2002). “Construction Site Safety Roles”. Journal Of Construction Engineering And Management, ASCE, 128(3), 203-210.

Trist, E., and K. Bamforth. (1951). “Some Social and Psychological Consequences of the Longwall Meth-od of Coal-Getting”. Human Relations. 4 (1): 3-38.

Vargas, C. A. (1998). Investigating Construction Falls using Fault Tree analysis and developing a pro-totype tool to reduce falls using expert system and computer assisted instruction method. Ph.D. Disser-tation, Civil Engineering, Ohio State University.

Veazie, MA., Landen, DD., Bender, TR., Amandus, HE. (1994). “Epidemiological Research On The Etiology Of Injuries At Work.” Annual Review Of Public Health, Vol. 15, 203-221

Vincoli, J. W. (1993). Basic guide to System Safety. Van Nostrand Reinhold, New York.

Wagenaar, W. A., Hudson, P.T., and Reason J. T. (1990). “ Cognitive failure and accidents.” Applied Cognitive Psychology, Vol. 4, 273-294.

Weaver, D. A. (1973). “TOR Analysis: A Diagnostic Training Tool.” ASSE Journal, June, 24-29.

Weaver, D. A.(1971). “Symptoms of Operational Error.” Professional Safety, October, 24-34.

Wickens, C. (1992). Engineering Psychology and Human Performance. HarperCollins Publisher Inc.,24pp.

Womack, J., and Jones, D. (1996). Lean Thinking: Banish Waste and Create Wealth in Your Corpora-tion. Simon and Schuster, New York, NY.

Womack, J., Jones, D., and Roos, D. (1990). The Machine That Changed the World,. MacMillan Publishing, New York, NY. Woods, D (2006). “Es-sential Characteristics of Resilience “ in Hollnagel, E., Woods, D., and Leveson, N. (2006). Resilience engineering: concepts and precepts. Aldershot, Eng-land, Ashgate.

Woods, D and Wreathall, J ( 2007). “Stress-Strain Plots as a Model of an Organization’s Resilience.” In press, accessed 4/3/2008 at http://csel.eng.ohio-state.edu/productions/ibm/Woods_Wreathall-stress-strain%20w-refs.pdf

Woods, D. D. (2007). Proactive safety management: Engineering resilience makes the difference between a disruption and a disaster. Industrial Engineer -Nor-cross-. 39, 44-48.

Woods, D. D., and Hollnagel, E. (2006). Joint cogni-tive systems: patterns in cognitive systems engineer-ing. Boca Raton, CRC/Taylor and Francis.

Woods,D. and Wreathall, J.( 2007). “Stress-Strain Plots as a Model of an Organization’s Resilience” in Hollnagel, E., Nemeth, C. P., and Dekker, S. (2008).

Woods,D. and Wreathall, J.( 2007). “Stress-Strain Plots as a Model of an Organization’s Resilience” in Hollnagel, E., Nemeth, C. P., and Dekker, S. (2008).

Woods,David and Wreathall, John ( 2007). “Stress-Strain Plots as a Model of an Organization’s Resil-ience.” In press, accessed 4/3/2008 at http://csel.eng.ohio-state.edu/productions/ibm/Woods_Wreathall-stress-strain%20w-refs.pdf

Zabetakis, M. (1975). Safety Manual No. 4, Acci-dent Prevention, MSHA, Washington.

Zeller, D. B. (1986). “Heinrich revisited.” Profes-sional Safety, October, 40-42.

68

Integrated Collaborative Experiences - Construction and Architecture Programs in the Same College

Richard C. Ryan AIC, CPC, LEED AP and Kenneth F. Robson AIC, CPC

ABSTRACT: The role of the Construction Manager is being greatly expanded from the traditional role of the General Contractor. The authors believe that there is great opportunity for construction and architecture academic programs to use joint integrated collaborative educational experiences to benefit construction and design students. Programs housed in the same College provide the most optimal envi-ronment and natural place to develop and implement these types of experiences. The authors found no comprehensive reporting of what types of integrated collaborative experiences are being developed and used today based on literature review and conversations with Associated Schools of Construction (ASC) members. It also appears that there are several interpretations of the meaning of collaboration between construction and architecture students. The purpose of this article is to document and categorize what fifteen U.S. Construction programs housed in the same College as an architecture program are doing or have done to provide integrated collaborative experiences for their students. It is hoped that this infor-mation can be used as the basis for discussion about the need and incorporation of these types of expe-riences into construction curriculums.

Key Words:Academic, Architecture, Collaboration, College, Construction Science, Integrated

INTRODUCTION

The role of the construction manager is being ex-panded greatly from the traditional role of the general contractor. Alternative project delivery driven pre-construction interaction with the architect and other project leadership during the design phase is often required of the contractor today.

The introduction of the budget and schedule to a three dimension building model is a catalyst for owner man-dated design phase collaboration between the architect and contractor. The definition, intent and guidelines for project collaboration are evolving for the owner,

architect and contractor. Cultivating preconstruction team leaders requires broader views and skill sets from educators and graduates. To address these challenges, the authors believe that there is great opportunity for academic construction and architecture programs to use integrated collaborative educational experiences to better prepare students for these roles in the future. In 1996 University of Oklahoma College of Architecture Professors Robson, Caldwell and Reynolds spoke of the benefit of integrating construction and architecture students to perform a collaborative exercise. They per-ceived that the “initiative enriched learning beyond the confines of a single course or discipline….more reflec-tive of a real-life situation than the typical University classroom experience.” (Robson, Caldwell and Reyn-olds, 1996) Ten years later Holley and Dagg discuss their efforts to permanently implement a collaborative experience in Auburn’s Building Science program.


RICHARD RYAN is a professor at Oklahoma University.

KENNETH ROBSON is a professor at Oklahoma University.


69 Integrated Collaborative Experiences - Construction and Architecture Programs in the Same College


“Conversely many academic institutions continue to neglect the opportunity to collaborate between the design and construction aspects of building by ignor-ing shared responsibilities and the shifts in industry This is significant because pedagogical models for teaching within the two curriculums have historical-ly been discipline specific contributing to the contin-ued lack of interaction between design and building education.” (Holley and Dagg, 2006)

Using the authors’ knowledge of Associated Schools of Construction programs and literature review, it seems that since 1996, efforts to implement collab-orative based experiences into academic construction curriculums, with a few exceptions, are limited and primarily developed and implemented by individual faculty, not programs. The authors found limited reporting about integrated construction and architec-ture student collaborative experiences being devel-oped and used. It also appears that there are several types or levels of experiences considered integrated and collaborative.

The purpose of this article is to document and cat-egorize what the fifteen United States Construction programs housed in the same College as an archi-tecture program are doing or have done to provide integrated collaborative experiences for their stu-dents. It is hoped that these results can be used as the basis for discussion about the need and incorporation of these types of experiences in the construction curriculum and to better determine best practices for integrated collaborative experiences.

THE STUDY

STUDY INTENT

The basic intent of this research was to document in-tegrated collaborative efforts so that other university construction programs might use this information to build on these efforts. The authors felt that because of convenience and proximity, construction and architecture programs housed in the same college are in the most optimal environment and natural place to develop and implement collaborative experiences.

This premise was not intended to exclude or suggest that programs not in the same college are not imple-menting these types of experiences.

To the authors’ knowledge there was no resource available that classified perceived collaborative ef-forts into types of experiences ranging from students being taught in the same room to students partici-pating in activities designed specifically to promote interaction and collaboration as part of a structured required class format. Work has already started to explore this issue by the recently formed A + C Alli-ance comprised of Colleges housing both construc-tion and architecture programs. However there has been no leading effort by construction educators to determine and compile a resource of actual imple-mented collaborative experiences.

STUDY OBJECTIVES

As stated in the cover letter included as part of the initial correspondence to the targeted program, this study was not intended to highlight positively or negatively programs incorporating or not incorporat-ing these types of experiences. Also requesting this information from the construction side of the col-laboration was not intended to exclude input from architecture colleagues.

Specific study objectives include:

• Document targeted construction programs’ reported integrated Construction and Archiecture undergraduate student experiences included in their programs.

• Report the breakdown by type of experience as a percentage (%) of total experiences.

• Report the breakdown by percentage of collaborative or non-collaborative experiences as a percentage (%) of total experiences.

• Report the breakdown by percentage of fac ulty collaboration (both construction and


Richard C. Ryan AIC, CPC, LEED AP and Kenneth F. Robson AIC, CPC. 70


architecture instructors) experiences as a percentage (%) of total experiences.

• Report the average support rating for each affiliated group for all reporting programs (Question 2).

• Report noted common challenges for all reporting programs (Question 3).

• Report noteworthy future experiences to be instituted by reporting programs (Question 4).

STUDY GROUP

Table 1 lists construction programs included in this study. These programs were targeted because their College is included in the A + C Alliance group.

Table 1: A+C Alliance Member Universities

No. University 1 Auburn University2 California Polytechnic State University3 Clemson University4 University of Florida5 Georgia Institute of Technology6 Mississippi State University7 University of Oklahoma8 Prairie View A&M University9 Pratt Institute10 University of Southern California11 Southern Polytechnic State University12 Texas A&M University13 Virginia Tech14 University of Washington15 Washington State University

REQUESTED PROGRAM INFORMATION

The following information was requested from the respondent completing the survey.

• University Name.• College Name.• Construction Department Name.• Architecture Department Name.• Name (Person completing this survey.).• Title (Person completing this survey.).• Date of submission.• Number of Construction students in program.• Number of full time Construction faculty. • Number of part time Construction faculty.• Number of Architecture students in program. • Number of full time Architecture faculty. • Number of part time Architecture faculty.

IMPORTANT DEFINITIONS USED

The respondents were asked to consider these defini-tions when completing the survey questions.

An Experience is considered one of the following:

• A required course – listed as part of the required Construction curriculum.

• Part of a required course – is included as part of a required course.

• An elective – a course credited in the curricu lum hours, but not required.

• A seminar – a short term organized meeting/ session for delivering, exploring or discuss ing information.

• An extracurricular activity – an activity re lated to the program, but not required.

• Other activities.



Undergraduate means the experience is offered primarily for undergraduate Construction and Archi-tecture students (it might include graduate students as participants).

• Integrated means the participating group is combined Construction and Architec- ture students.

• Collaborative means Construction and Archi tecture students are required to interact using discipline based expertise and perspectives to explore a topic, solve a problem or complete a deliverable or an activity. Examples might be a design/build project or design-estimating exercise.

• Non-collaborative means the experience is integrated, but discipline-based expertise or interaction are not required. Examples might be history of built environment, software or structures courses.

• Support means agreement with the invest ment of time and program resources to provide integrated collaborative experiences.

• Faculty collaboration means Construction and Architecture instructors share mutual responsibility to prepare and deliver the experience.

SURVEY QUESTIONS

Appendix A at the end of the article includes the exact survey text listed by question with abbreviated table examples included for the respondents’ hand-written or word-processed input.

DATA COLLECTION

A cover letter explaining the study and purpose, the survey and an informed consent form to be com-

pleted and returned with the survey was initially emailed to the targeted construction program de-partment heads or chairs. Cover letters encouraged respondents to get input from others in their program and the related architecture program if necessary. The authors communicated to potential respondents that survey results would be reported as a whole and not by program. Submission of a copy of the com-pleted survey and signed consent signed form were requested by a specific date. Two follow-up emails were sent as reminders. These emails were followed up by telephone calls as required to answer questions or encourage submission.

About half of the returned surveys were returned by email as an edited Word document and the rest by hard-copy in the mail or in person. In several in-stances completion of the survey was delegated to a responsible party in the Construction department.

RESULTS AND ANALYSIS

The largest construction program responding to the survey listed 775 construction, 550 architecture stu-dents and 27 full-time construction faculty members in the college. The smallest construction program responding to the survey listed 75 construction, 614 architecture students and 1 full-time construction faculty member in the college. The 6th and 7th larg-est programs by number of students listed the most integrated collaborative experiences.

Twelve of the fifteen targeted programs responded for an 80% response rate. Based on positive experi-ences with some respondents during the data collec-tion part of this study it is worth noting that pro-grams that were exploring these types of experiences were typically supportive and anxious to complete the survey and explain what they were doing.

The authors acknowledge the potential for personal bias or subjectivity due to the construction perspec-tive of the survey respondents and authors. The authors’ assessment of the type (category) of re-



ported experience was based on perceived collabora-tion of the students and was applied consistently to all responses. Only undergraduate experiences were included. Construction minors and joint degrees are considered “Other” experiences. They are considered integrated due to architecture students being in the construction class, but typically are not collabora-tive because content, delivery and activities are the same for all class participants and discipline based skills are not necessarily used or incorporated. Team interaction in this type of class is typically not multi-discipline based. Unless offered as a course specifi-cally, international study programs and participation in student competitions were considered extracur-ricular activities.

REPORTED EXPERIENCES

Fifty integrated (combined construction and archi-tecture students) experiences were reported and categorized from the twelve responding programs. Thirty experiences (60%) were considered collabora-tive (students are required to interact using discipline based expertise and perspectives) including student competitions. This percentage included participation in design/build student competitions. Ten of these collaborative experiences (33%) were required, elec-tive or part of a course and utilized construction and architecture faculty collaboration.

Three required courses from three different programs utilizing faculty collaboration were reported (30% of student and faculty collaborative experiences). The following discussion includes the course name and a brief description for each.

1. Integrated Project Services: Overview of proj-ect delivery methods including integrated services, entity organizational structures, process variations, selection and procurement methodologies to achieve maximum project quality and value. The class is in-tended to be 50% construction and 50% architecture students.

2. Design and Construction Administration: De-signed to explore concepts and tools that can be

used to promote better understanding and collabora-tion between parties in the design and construction process. Presentation and interaction with prominent, active industry members will be used to reinforce this objective and supporting class content. Content includes introduction to design and construction administration procedures, necessary collaborative communication and documentation, contract admin-istration, project permitting, field documentation and reporting. This course was offered 2006 and 2007.

3. Interdisciplinary Capstone: A senior capstone for students preparing to enter the design-build sector of the construction industry; integration of the design and construction processes into a single, cohesive project delivery system, starting with project incep-tion, and carrying through construction, operation and maintenance of various types of construction projects.

Four elective courses from four different programs utilizing faculty collaboration were reported (40% of student and faculty collaborative experiences). The following discussion includes the course name and a brief description for each.

1. Undergraduate Study Abroad: 2006, 2008 and 2009 efforts include students working together to de-velop retrofit solutions, budgets and business model for the local community.

2. Building Information Modeling: An immersive eight day course designed to develop team building skills; convey an operational understanding of the sustainable design strategies utilizing the U.S. Green Building Council LEED criteria; software training on the development of concept modeling utilizing AutoDesk Revit; and training on the interoperability of Revit models with DesignEst estimation interface and Primavera scheduling support software.

3. Capstone Project: integrated teams work on a project to better prepare for interdisciplinary experi-ences.

4. Design/Build Studio: Students work in a team to develop conceptual designs, cost estimates, construc-



tion schedules and project management plans.

Three “part of course” experiences utilizing faculty collaboration was reported (30% of student and fac-ulty collaborative experiences). These experiences were all project based.

92% of reporting programs listed at least one inte-grated collaborative experience.

SUPPORT RATINGS

Table 2 shows responses by affiliated group to Sur-vey Question 2. The average rating is the strength of support for undergraduate integrated collaborative experiences using a scale of 0 = no support to 5 = strongly support. The number of responses varies due to the survey respondent not knowing how to rate the support of some of the specific groups. Two respondents did not complete this section of the survey.

Table 2: Survey Question 2 Average Support Ratings

Group Number of Responses

Average Rating

Construction program students

10 3.60

Architecture program students

10 3.25

Construction pro-gram faculty

10 4.10

Architecture pro-gram faculty

10 2.90

College adminis-tration

10 4.30

University admin-istration

5 4.40

Construction industry

9 4.22

Architecture industry

7 3.00

The average support rating from both student groups

is considered moderate (3.6 and 3.25). Two pro-grams reported that architecture students supported these types of experiences more than Construction students. In both of these cases it was perceived that there was very little College support (1).

Construction faculty strongly support (4.1) these experiences. However one construction program’s faculty was perceived as much less supportive (2.0). The average perception of architecture faculty sup-port was much lower (2.9). One program listed no support (0.0) from architecture students or faculty in their college. Architecture faculty support (2.9) is less than architecture students (3.25).

Average College (4.3) and University (4.4) support appears to be strong. One program reported little support from any students or faculty, but optimum support (5.0) from the College, University and both related industries.

It is no surprise that construction industry support was perceived as strong (4.2), paralleling construc-tion faculty support (4.1). Similarly architecture industry support was perceived as moderate (3.0), paralleling architecture faculty support (2.9).

COMMON CHALLENGES

Most respondents’ comments to Question 3 (per-ceived challenges to implementing undergraduate integrated collaborative experiences) listed logisti-cal issues as challenges. These included lack of adequate class room space, available willing faculty and elective hours in the curriculum. Meeting differ-ent accreditation requirements and faculty teaching efficiency and credit when courses are team taught were included too. Several respondents listed lack of or minimal support by both groups of students and architecture faculty as major challenges also.

NOTABLE FUTURE EFFORTS

Two respondents listed new Design/Build courses to



be implemented in 2009. One program listed a new integrated BIM course proposed for Fall 2009.

OBSERVATIONS

The authors recognize the limited study group size, but consider the results representative of the targeted A+C Alliance programs. Respondents’ subjectivity in completing the survey is also acknowledged. All respondents were considered suitable to complete the survey and were knowledgeable about their program. Based on this input it appears there are a limited number of required or elective courses incorporating student and faculty collaboration. It appears that cur-rent notable efforts are driven by individual faculty developing and implementing these efforts, not as part of a program’s curriculum policy or approach. Perhaps the limited number of offerings is influenced by the perceived moderate support of construction and architecture students and architecture faculty. The program reporting the most collaborative ex-periences ranked seventh in number of construction students.

It seems the limited number of collaborative course offerings and faculty and student support do not match the perceived College and University sup-port. Though construction students appear to support collaboration based experiences more, the average support ratings from both groups of students is con-sidered moderate (3.6 and 3.25).

Strong construction industry support and weaker/moderate architecture industry support is somewhat predictable as the results parallel faculty percep-tions (4.1 to 4.22 and 2.90 to 3.00). This similarity is possibly attributable to the close industry ties both disciplines typically maintain thru advisory groups, job fairs and classroom interaction.

Architecture faculty perceived support (2.90) is less than their students (3.25). This is worth noting due to the usual juxtaposition of faculty leading or influenc-ing student perceptions.

Common logistical challenges require extra attention to execution details and greater planning by respec-tive construction and architecture programs. The re-sistance by either discipline to include collaborative experiences into curriculums is a looming challenge and will require a culture shift for additional or im-proved implementation of experiences. The limited number and types of new experiences being planned for the future reported in this survey may be reflec-tive of this challenge.

CONCLUSION

The authors realize that this Fall 2008 survey only provides a snapshot of what is happening. It will change. The results also only represent the A+C Al-liance Construction programs. Documenting what other ASC programs not housed in the same college are doing is the next step to establishing a compre-hensive baseline for comparison and documentation of best practices. Also gathering similar input from the respective architecture programs would be valu-able for comparison and validation of these reported results. The moderate perception of students’ support suggests more exploration is needed o determine the need, the desired value and resultant best practices.

Support ratings suggest architecture students, faculty and the industry are less supportive of collaborative efforts than the other listed groups. To some de-gree this is understandable and predictable as it has been the topic of conversations for many years. The authors hope that all construction and architecture programs can use this information to review their strategy for including integrated collaborative expe-riences to help their students better understand the possible needs of the disciplines in the future.

REFERENCES

Holley, Paul and Dagg, Christian. (2006). Develop-ment of Expanded Multidisciplinary Collaborative Exercises across Construction and Design Curricula.



International Journal of Construction Education and Research. Taylor and Francis, Group, LLC. Volume 2, Issue 1. p. 29-41.

Robson, Kenneth, Caldwell, Mack and Reynolds, Jerlene. (1996). Enhancing Communication in the Design and Construction Industry through Multi-Disciplinary Education. Journal of Construction Education. Associated Schools of Construction. Summer 1996, Vol. 1, No. 1, p. 50-58.

APPENDIX A

Question 1: Please complete the table below. Consider definitions on page 2 and the following explanation of column headings.

• Column 2 - Experience: Briefly list integrated experiences included in your program. Include the course title, activity title or other designation.

• Column 3: Designate the Type of experience – color highlight or circle/mark the selection.

• Column 4: Designate whether the experience was Collaborative or Non-collaborative – color highlight or circle/mark the selection.

• Column 5: Designate whether the experience involved Faculty Collaboration – color highlight or circle/ mark the selection.

No Experience Type Collaborative - Cor

Non-collaborative - NC

Faculty Collabora-tion

Yes or No

1 Required course Part of course

Elective Seminar

ExtracurricularOther

C NC Yes No

For each experience listed in the table (Experience column) on page 3 please briefly describe in the spaces be-low the objective(s) and benefits or drawbacks for each.

1.

Question 2: Use this scale to answer Question 2. 0 = no support; 1 = strongly against; 2 = against; 3 = neu-tral; 4 = support; 5 = strongly support and DK = don’t know

76

The Impacts of Undergraduate Construction Internships on Recruitment, Training, and Retention of Entry-Level Employees of the Construction Industry

David Bilbo, Jose L. Fernández-Solís, Nathan Bohne, M.Sc., and Mohamad Waseem

ABSTRACT: The purpose of this study is to identify and analyze the impact of undergraduate con-struction internships on recruitment, training, and retention of entry-level employees in the construc-tion industry. This study analyzes short term and long term benefits, if any, to both the intern and the company. To obtain this information, a survey was conducted of construction companies that recruit and hire entry-level employees from the Construction Science Department at Texas A&M University. A total of 253 registered Construction Science Internship Providers were asked to participate in the study. The survey gathered data identifying current trends of internships in construction education, analyzed the performance of interns versus non-interns, and assessed the overall effectiveness of the undergraduate internship program. Descriptive and inferential statistics were used to analyze the results of the surveys. This study allowed the sampled Internship Providers the opportunity to provide feedback regarding recruitment, training, and retention of undergraduate interns, as well as the opportunity to suggest im-provements to the undergraduate internship program in the Construction Science Department at Texas A&M University.

Key Words:Construction Internship, Construction Science

BACKGROUND

An internship is a mutual work agreement between an intern and an employer (before the student becomes a full time employee), for a specified period of time, in exchange for pay, college credit hours, or both. To ac-

quire the skills, knowledge, and methods necessary to become a successful construction manager, the student intern must be willing to learn and to invest his or her time, patience, and abilities. While the skills needed to become a construction manager are many, the fun-damentals begin in the classroom. A college degree is generally accepted as essential in today’s competitive work environment. However, simply having a college degree is not sufficient. The well-qualified employee


DR. DAVID BILBO is the holder of the Clark Endowed Professorship in Construction Science. He joined the faculty of Construc-tion Science in 1977. Bilbo is a Faculty Fellow of the Hazard Reduction and Recovery Center at Texas A&M University. He has done extensive research on construction education graduates, including a 20 year longitudinal study, and earlier studies on the supply and demand for construction graduates. Bilbo continues to research on the issues of diversity and gender issues faced by the construction industry. Dr. Bilbo formerly served as the graduate program coordinator and associate department head for the Construction Science Program at Texas A&M University.

DR. JOSE L. FERNANDEZ-SOLIS is an Assistant Professor in the Construction Science Department at Texas A&M University. Dr. Solis is a member of Royal Institution of Chartered Surveyors in the United Kingdom. His research interests include green build-ing, Corps of Cadets Company A-2 Academic Mentor, CRS Fellow and Sustainable Urbanism Fellow.

MR. NATHAN BOHNE holds a Masters of Construction Management from Texas A&M University.

MR. MOHAMAD WASEEM is a current student in Masters of Construction Management program at Texas A&M University.


77 The Impacts of Undergraduate Construction Internships on Recruitment, Training, and Retention of Entry-Level Employees of the Construction Industry

also needs experience in the workplace in order to clearly understand the job expectations. Ideally an internship will provide this opportunity. This study proposes to identify and analyze the impact of un-dergraduate construction internships on recruitment, training, and retention of entry-level employees in the construction industry.

Companies rely on interns more and more as labor shortages have resulted in an increasing backlog of work (Fedorko, 2006). As the construction industry faces these labor shortages, the recruitment, train-ing, and retention of entry-level employees remains an important task. Entry-level employees represent a small but important subgroup of the total employees within the construction industry. Interns are consid-ered to be temporary employees, yet they share the same responsibilities as their professional counter-parts, are usually paid less than full time profession-als, and seldom receive full-time benefits such as health insurance and retirement.

Coco (2000) reported that companies, when hiring interns, experience increased job placement, de-creased turn-over rates, and increased job satisfac-tion and starting salaries after the internship ends. The economic health of the construction industry has a direct impact on consumer confidence and spending. According to a report published by the Bureau of Labor Statistics, the construction industry was valued at $647.9 billion dollars (United States Department of Labor Statistics, 2006). From 2001-2006, the Gross Domestic Product (GDP) of the con-struction industry increased an impressive 37.99%,. The raw forces behind construction are demograph-ics (population increase) and increase in affluence (the ability to afford more, new or upgraded space) (Fernández-Solís 2008). Since increased affluence translated as consumer spending fuels new con-struction and increases employment opportunities within the industry, a strong economy has resulted in construction employment rising from 5,813,000 employees in 1997 to an all-time high of 7,689,000 employees in 2006. Future growth and develop-

ment will require greater recruitment and training of entry-level personnel into the workforce to offset growth and those workers leaving the industry. The Department of Labor Statistics predicts that the con-struction industry will experience a 12% increase in employment from 2004-2014.

The short and long term future outlook for the in-dustry remains positive. Globally, all construction in place in 2000 will double by 2030. In the USA all construction in place in 2000 is projected to double in 2050 with certain regions (North East) increasing slower than others (South Central) (Fernández-Solís 2007). However, inflation (such as commodities), rising energy prices, and increased foreign competi-tion have led to cost increases for contractors. Since economic factors are uncontrollable, a company’s ability to recruit and retain competent, productive employees is essential to survival. “Clearly, every business is only as good as the people it brings into the organization” (Carrison, Walsh, 1999). As em-ployers demand increasingly more from potential employees, locating and attracting new employees that are dependable, ethical, and loyal remains a top priority. A 2003 study conducted by the National Association of Colleges and Employers (NACE) suggested that, before employers interview new col-lege graduates, they prefer to look within their own internship or coop programs.

Increased demand for labor and professionals has re-sulted in more universities offering programs in con-struction education. In 2007, the Associated Schools of Construction (ASC) reported a total of 116 ac-credited construction programs within the United States. Universities that provide training in construc-tion education remain a significant resource for the recruitment of interns and entry-level employees. A study conducted by Beard (1998) rated recruitment of future employees as the top benefit of internships. Training the leaders of tomorrow requires a substan-tial commitment from students, faculty, and employ-ers. Internships offer students practical experience in a controlled environment as they acquire relevant


Dr. David Bilbo, Dr. Jose L. Fernández-Solís, Nathan Bohne, M.Sc., Mohamad Waseem 78

work experience. Experiential learning involves more than simply learning job skills: socialization and acculturation into an organization are important developmental challenges that interns encounter (Tovey, 2001).

Presently, no set industry or academic standards exist for developing or implementing a coop/intern-ship program. According to Marshall (1999), major internship goals include: a) provide the opportunity to integrate and apply the knowledge, skills and at-titudes developed in the college or university cur-riculum; b) provide students an opportunity within an enterprise, while meeting the organization’s performance standards; c) enable students to refine planning, communication, and technical abilities in a real world environment while establishing resume worthy experience; and d) demonstrate professional-ism and accountability in meeting all commitments required of the intern to make consistent contribu-tions to his or her employer.The implementation of internship programs has in-creased widely among universities across the nation. Much literature exists hailing the positive effects of internships. However, few studies have been con-ducted to determine the long-term impact of hiring interns and to assess their retention as entry-level employees.

PROBLEM STATEMENT

The purpose of this study is to identify and analyze the impact of undergraduate construction internships on recruitment, training, and retention of entry-level employees in the construction industry.

RESEARCH OBJECTIVES

1. To determine the current trends of internships in construction education.2. To analyze the impact of undergraduate internships on the recruitment, training and retention of entry-level employees.3. To assess the effectiveness of undergraduate

internships within the Construction Science Department at Texas A&M University.

RESEARCH METHODOLOGY

POPULATION AND DATA COLLECTION

The methodology utilized to collect data for this study was an online survey conducted during the spring and summer of 2007. The Texas A&M Uni-versity Institutional Review Board approved this study. The sample population consisted of 253 Internship Providers of the Texas A&M University Construction Science Department, each of whom was contacted by phone. Participation in the study was voluntary and no monetary benefits were dis-tributed for participating in the study. Participants were emailed the on-line survey as an attachment, and each participant was allowed to take the survey only one time to prevent multiple submissions from the same participant. The responses were received in a secure database established at Surveymonkey.com. The survey consisted of 29 multiple-choice questions, allowing only one response per question. Descriptive statistics were used to analyze the data collected from the online survey responses.

DEMOGRAPHICS

Of the 253 Internship Providers contacted by phone, 111 responses were collected, for a response rate of 44%. Respondents to the survey represented compa-nies of varying sizes and revenues. The companies ranged in size from one to over 500 employees, with annual revenues ranging from one million to over $750 million. Participants represented all sectors of the construction industry: residential, commercial, heavy, highway, industrial, and specialty contractors. Respondents reported recruiting interns and entry-level employees from construction programs that offered internships. Responses were collected from human resource personnel, superintendents, con-struction managers, company executives, and owners of companies.



DATA ANALYSIS AND RESULTS

Currents Trends: Internship ProvidersThe first objective of the survey was to identify companies that recruited interns and entry-level employees. Company respondents were identified by their size and recruitment methods. Company size was assessed by the number of employees at the company and average annual revenue. Questions 2 and 3 were used to analyze this objective. The data revealed that the largest group of Internship Pro-viders (34.2%) was smaller companies consisting of 1-49 employees. It can be deduced that smaller construction companies find value in hiring interns as a source of qualified labor. The top recruiters of interns and entry-level employees reported average annual revenue of $26-$150 million.

ACADEMIC CREDITS AND DURATION FOR INTERNSHIP

Question 6 asked participants whether students should receive academic credit for doing an intern-ship. Overwhelmingly, 99.1% of the respondents reported that students should receive credit for doing an internship. Question 7 asked participants how many academic credit hours interns should receive for completing an internship. In order to clarify the basis for one academic credit hour, responders were informed that 1 credit hour is equivalent to 15 con-tact hours in a classroom setting. Overall, 41.3% of respondents reported that interns should receive 3-4 credit hours for completing an internship, as seen in

Figure 1.

Question 18 assessed the average duration of intern-ships at these companies. Of the 79 responders, the average duration of an internship at their company was between three and four months, with 17.9% reporting the duration to be between five and six months. The results indicate 3-4 months is an appro-priate duration for an internship, as seen in Figure 2.

SALARY TRENDS

Question 13 was a yes/no response question that asked participants if the Texas A&M University Construction Science Department should limit the salary of interns. Over half (57.3%) of the respon-dents answered “Yes,” the university should limit the salary of interns. Question 14 asked participants what they thought was an “appropriate” hourly sal-ary for interns: of the respondents, 47.3% reported $10-$13 per hour was an appropriate hourly salary for interns, while 40.9% reported an appropriate sal-ary of $14-$17 per hour. See Figure 3 for complete results.

ESTABLISHED TRAINING PROGRAMS

The following questions assess the status of employ-ee training and the importance of continuing edu-cation. Questions 27 and 28 were yes/no response questions that evaluated whether the company had an established training program set up for interns and full-time employees. As seen in Figure 4, a ma-

Figure 1. Credit Hours for an Internship



jority (50.9%) of responders reported that their com-pany had established training programs for interns. As seen in Figure 5, the majority of respondents (74.6%) reported that their company had established training programs for full time employees. Question 26 asked participants about the importance of con-tinuing education and training. The survey indicated that continuing education and training are valued by companies. Question 29 assessed the time frame in which training was provided. The majority of re-spondents reported their company provided monthly training for full time employees. 3.2 Effectiveness of Undergraduate Internship Pro-grams

Question 1 assessed the recruitment patterns of these companies at schools that offer construction pro-grams. Recruitment patterns were limited to under-graduate construction programs with internships and without internships. Respondents reported that they recruit from both construction programs that offered internships (44.0%) and from programs that did not offer internships (3.7%), and a majority from both (with or without an internship program: 52.3%) as seen in Figure 6. This indicates a preference for recruiting students with internship experiences.

HOW IMPORTANT IS AN INTERNSHIP?

Question 4 measures the importance of internships toward an undergraduate’s education: all participants responded to the question. Participants were asked

to measure “importance” by selecting one choice from the following categories: “definitely impor-tant,” “probably important,” “undecided,” “probably unimportant,” “definitely unimportant.” The results indicate that 67.6% of the Internship Providers be-lieved participation in an internship was “definitely important” to an undergraduates’ construction educa-tion, as seen in Figure 7.

RETENTION RATES FOR INTERNS

Question 25 was a yes/no response that asked par-ticipants whether job retention is higher for those that had internship experience versus those that had no internship experience. The majority (65.7% of responders) indicated that retention is higher for entry-level employees hired with past intern experi-ence; 34.3% indicated that retention was not higher for entry-level employees hired with past internship experience. The survey indicates that companies that hire interns have higher retention rates (of those interns) than companies that hire candidates with no internship experience. The results are displayed in Figure 8.

Question 23 asked participants what percentages of interns come to work for their company as full-time employees upon graduation. The majority of respon-dents reported that 26-50% of interns come to work for the company as full time employees upon gradu-ating. The results are displayed in Figure 8. Question 24 asked participants what percentage of new gradu-

Figure 2. Internship-Durations

Figure 2. Internship-Durations



ates hired were still with the company after five years; the results are displayed in Figure 10. Respon-dents were undecided on the retention rates of new graduates after 5 years of full time employment.

IMPACT OF UNDERGRADUATE INTERNSHIPS ON RECRUITMENT POLICIES

The following three questions investigated whether the internship experience has any effect on the start-ing salary and other recruitment policies for entry level employees. Question 21 was a yes/no response question that asked participants if internship experi-ence was a factor in determining the starting salary of new graduates. Respondents (68.5%) reported internship experience is a factor in determining the starting salary of new graduates (See Figure 11).Question 22 assessed the impact of the internship experience in determining the starting salary for entry-level graduates. Participants were asked to ap-proximate this impact as a percentage increase above the average starting salary. The results were rather interesting, with two categories tied: 32 respondents reported that internship experience has no impact in determining the starting salary for entry-level gradu-ates and 32 respondents reported internship experi-ence results in an increase of 3-5% in the starting salary for entry-level graduates. Overwhelmingly, 76 out of 108 respondents reported internship experi-ence leads to an increase in starting salary ranging from 1-7%. The data for this question can be seen in Figure 12.

Respondents were asked if interns are better pre-pared for entry-level employment over their non-intern counterparts. Participants selected one answer from the following rating factors: “strongly agree,” “agree,” “undecided,” “disagree,” “strongly dis-agree.” All 111 respondents answered the question. Overwhelmingly, 70.3% of respondents “strongly agreed” that students who participated in internships were better prepared for entry-level employment than were non-interns. The results are displayed in Figure 13.

SUMMARY AND CONCLUSIONS

Internships remain an important tool in the recruit-ment and training of entry-level employees in the construction industry. This study sought to provide additional data and insight into the trends, recruit-ment, compensation, retention, and training meth-ods applied and conducted by current construction Internship Providers at Texas A&M University. Participation in internships is becoming the norm as more companies shift their recruiting focus toward hiring interns. The study reveals internships to be an important and effective method of recruiting and re-taining qualified entry-level employees. The major-ity of companies have established training programs in place for interns and full-time employees. Overall, respondents reported Texas A&M Construction Sci-ence graduates with an internship experience to be productive and prepared for the work environment.

Figure 3. Hourly salary of interns



The results of this study indicate that students participating in an internship are better prepared for entry-lev-el employment than students who have not participated in an internship. In addition, internship experience results in a higher starting salary over non-interns. Clearly, successful interns have a greater opportunity to become satisfied employees. Evaluating the success and job retention rates of interns after five years of full-time employment would be a significant follow-up study.

Figure 4. Established training program for interns?

Figure 5. Established training program for full time employees?

Figure 6. Internships-Does it matter?



Figure 7. Internships as a part of construction education

Figure 8. Effect of internship on retention of new employees

Figure 9. Percentage of interns joining the company back upon graduation



Figure 10. Retention after 5 years of employment

Figure 11. Internships and starting salaries of graduates

Figure12. Internship-Impact on starting salaries



REFERENCES

Beard, D. F. (1998). The status of internships/cooper-ative education experiences in accounting education, Journal of Accounting Education, 16, 507-516.

Carrison, D., Walsh, R. (1999). Business leadership the Marine Corps way. New York, NY: MJF Books.

Coco, M. (2000). Internships: A try before you buy arrangement. S.A.M. AdvancedManagement Journal (Society for Advancement of Management, 65, 41-43.Fedorko, J. (2006). The intern files: how to get, keep, and make the most of your internship. New York, NY: Simon and Schuster.

Fernández-Solís, J. L. 2008, “Is Building Construc-tion Approaching the Threshold of Becoming Unsus-tainable? A System Theoretic Exploration Towards a Post-Forrester Model for Taming Unsustainable Exponentialoids,” VDM and Co. Saarbruecken, Ger-many. Accessible at: http://archone.tamu.edu/faculty/jsolis

Fernández-Solís, J. L. 2007, “The Exponentialoid of Resource Consumption,” Proceedings, CIB World Building Congress 2007, Cape Town, South Africa, Proceedings CIB-457. Accessible at: http://archone.tamu.edu/faculty/jsolis

Marshall, J. A. (1999). Professional internships as a requirement for graduation. Journal of Industrial Technology, 15, 2-8.

Tovey, J. (2001). Building connections between industry and university: Implementing an internship program at a regional university. Technical Commu-nication Quarterly, 10, 225-239.

United States Department of Labor Statistics. (n.d.). Industry at a glance: NAICS 23: Construction. Re-trieved 22 June 2007, from http://www.bls.gov/oco/cg/cgs003.htm

Bohne, N. (2007). The impacts of undergraduate construction internships on recruitment, training, and retention of entry-level employees of the construc-tion industry. Unpublished Professional Paper. Texas A&M University

Figure 13. Internship and its impacts on performance

86

Construction Education at Texas A&M University: A Comparative Longitudinal Study of Graduates

David Bilbo, Ph.D., Jose L. Fernandez-Solis, Ph.D, Kristen M. Ramsey-Souder, M.Sc.

ABSTRACT: This research study reports the findings of a series of three surveys, over a period of 17 years, sent to graduates of the Department of Construction Science (COSC) at Texas A&M University in an attempt to assess the needs of the industry for construction graduates. A comparison analysis was made of the descriptive data regarding salary, employment information, employer demographics, cur-riculum ratings, and professional development. The data and findings can be used by institutions to respond to changes in trends of the construction industry.

Key Words:Construction education, construction graduates, con-struction industry, longitudinal study, outcome assess-ment

INTRODUCTION

The Construction education programs’ mission state-ments typically include striving to prepare students for successful careers in the construction industry. To achieve this goal, construction education programs must provide students with the skill-sets and compe-tencies needed to excel after graduation. These skills and competencies can be identified through feedback acquired from surveys of former students as well as from industry entities, such as the Construction In-dustry Advisory Board which most programs have in place. This study focuses on a series of surveys of graduates of the Construction Science (COSC) pro-

gram. Implementing this feedback into the curriculum allows construction education programs to remain responsive to the changing demands of the industry as well as continue preparing students for future success (Newitt, 1987).

The Department of Construction Science at Texas A&M University periodically conducts follow-up studies of department graduates. These surveys al-low the Department to assess the strengths and weak-nesses of the curriculum being taught, as well as obtain descriptive information such as salary, employment information, employer demographics, and professional development after graduation. Each singular study provides the graduates’ responses at a single point in time, which offers only a limited number of conclu-sions to be drawn from the results.

Over the last several decades, construction education


DR. DAVID BILBO is the holder of the Clark Endowed Professorship in Construction Science. He joined the faculty of Construc-tion Science in 1977. Bilbo is a Faculty Fellow of the Hazard Reduction and Recovery Center at Texas A&M University. He has done extensive research on construction education graduates, including a 20 year longitudinal study, and earlier studies on the supply and demand for construction graduates. Bilbo continues to research on the issues of diversity and gender issues faced by the construction industry. Dr. Bilbo formerly served as the graduate program coordinator and associate department head for the Construction Science Program at Texas A&M University.

DR. JOSE L. FERNANDEZ-SOLIS is an Assistant Professor in the Construction Science Department at Texas A&M University. Dr. Solis is a member of Royal Institution of Chartered Surveyors in the United Kingdom. His research interests include green build-ing, Corps of Cadets Company A-2 Academic Mentor, CRS Fellow and Sustainable Urbanism Fellow.

MS. KRISTEN M. RAMSEY-SOUDER holds a Masters of Construction Management from Texas A&M University.


87 Construction Education at Texas A&M University: A Comparative Longitudinal Study of Graduates


has evolved, and some even say transformed, in response to the needs of the construction industry (Burt and Hatipkarasulu 2007). Schools of construc-tion’s department directors have found it necessary to conduct self-studies periodically to keep abreast of industry needs and changes, confirm the aca-demic curriculum, in order to provide a rationale on how a curriculum should change. By replicating the follow-up studies of graduates and surveying at regular intervals, a longitudinal study can be devel-oped to follow graduates’ responses over time. A longitudinal study allows for a comprehensive set of conclusions to be drawn from the data gathered, leading to better informed decisions regarding pro-gram development.

Problem Statement

The purpose of this study is to analyze data collected from Construction Science graduates of Texas A&M University from 1987 through 2004.

Research Objectives

• Identify trends in the average salaries of Con struction Science graduates surveyed in 1987, 1998, and 2004.• Identify trends in job experience for those graduates surveyed in 1987, 1998 and 2004.• Identify trends in curriculum ratings from those graduates surveyed in 1987, 1998, and 2004.• Identify trends in professional development for those graduates surveyed in 1987, 1998, and 2004.

Delimitation

• This study is limited to data collected from surveys of Texas A&M Construction Science graduates conducted in 1987, 1998, and 2004.

METHODOLOGY

The Surveys and Data Organization

The survey recipients in all cases came from the department set of graduated students. The alumni database is regularly updated by the department through correspondence and more recently through emails. Although some alumni contact is lost over time, those that could be reached constitute the set used in the surveys in this timeline.

The surveys used for this study were refined from the original survey conducted in the fall semester of 1982. In 1982, the department initiated a survey of prior graduates as part of a movement toward ac-countability and outcomes assessment. This was designed to coincide with the American Council for Construction Education (ACCE) re-accreditation process. Data gathered from the first survey is inconsistent and incomplete, therefore, not incor-porated into this study. Since the original survey in 1982, the study was refined and replicated in 1987, 1998, and 2004.

Table 1 shows the total number of surveys mailed in each of these studies along with response rates.

Study Year Number of Surveys Mailed

Number of Usable

Responses Received

Response Rate (%)

1987 1500 493 32.91998 3,582 1,087 30.32004 3,965 935 23.6

TOTAL 9,047 2,515 27.8

The events in the economy and in the construction industry during the intervening times of the survey is not in the scope of this research.

The data collected through the surveys conducted in 1987, 1998, and 2004 were compiled into a compre-hensive database for graduates from the Department

Figures

1.3 1.8

28.0

1.7 1.7 1.8

10.95.6

1.32.9

30.7

48.6

1.0

24.1

12.6

53.7

10.7

52.1

0

10

20

30

40

50

60

None Some - NoAdditionalDegree

SecondBachelor's

Degree

Masters Degree PhD AGC'sSupervisory

TrainingProgram

Additional Education Level

Perc

ent 1987

19982004

Figure 1. Education since graduation averages


Dr. David Bilbo, Dr. José L. Fernández-Solís, Kristen M. Ramsey-Souder, M.Sc. 88

of Construction Science at Texas A&M University (COSC). The data was then sorted into groups based on the number of years since graduation (Gradu-ate Groups). The groupings of students into each Graduate Group were then chosen to coincide with the number of years comprising the groups from the 1987 survey, as described by Bilbo (1992). In other words, the groupings were designed and described by Bilbo 1992 and the same groupings were used in subsequent surveys.

The Graduate Groups are defined in Table 2.

Graduate Groups

Number of years

since graduation

1987 1998 2004

Group 1 0-8 years 1980-1987 1991-1998 1997-2004Group 2 9-18 years 1970-1979 1981-1990 1987-1996Group 3 19-28

yearsPrior to 1969

1971-1980 1977-1986

Group 4 29 years + N.A Prior to 1970

Prior to 1976

Salary datum (Figure 12) were collected and ana-lyzed, both as reported and adjusted for inflation to 2004 dollars (present day value). The Consumer Price Index (CPI) was utilized for the adjustment. Although there are many methods of translating dol-lars for inflation, CPI is a commonly used method, and is often considered the best method for changing income to “real” or “inflation-free” dollars ( U.S. Department of Labor, 2004).

The translated dollar amounts standardize the report-ed salaries so that conclusions can be drawn from the data. The formula utilized to adjust the reported salaries may be found in the Average Salaries portion of the Analysis section of this paper.

DATA ANALYSIS

Professional Development

In Figure 1, the data shows that approximately half of the graduates surveyed pursued no additional edu-cation after obtaining their bachelor’s degree. Ap-proximately 28% of graduates have received some education since graduation but earned no additional degree. Many graduates completed continuing education for career development or to meet require-ments for membership in some professional associa-tions. Of those graduates who did earn an additional degree, a master’s degree was the most common level of continued education, with response levels around 11% for each of the three surveys. Figure 2 shows the data for those who have earned a master’s degree, distributed across the Graduate Groups for each of the surveys. The data shows a positive trend between the number of years since graduation and the percentage of graduates who have obtained a master’s degree. The construction in-dustry has a demand for experienced graduates with management knowledge (Chan et al. 2002), and as some construction graduates progress through their

Figure 1. Education since graduation averages

14.6

6.0

11.4

20.4

24.1

6.0

8.8

12.211.1

6.9

N.A.

23.3

0.0

5.0

10.0

15.0

20.0

25.0

30.0

Group 1 (0-8 yrs.) Group 2 (9-18 yrs.) Group 3 (19-28 yrs.) Group 4 (29+ yrs.)

Graduate Group

Perc

ent 1987

19982004

Figure 2. Education since graduation by education level – Master’s Degree

9.7

53.6

6.59.0

1.9

8.412.9

6.3

1.96.8

N.A.N.A.

8.2

41.9

2.65.8

0.0

10.0

20.0

30.0

40.0

50.0

60.0

AIC AC CPC AIA PE CSI Realtor Other

Associations & Certifications

Perc

ent

19982004

Figure 3. Professional association and certification averages JUNE 2010 — Volume 33, Number 3

The American Institute of Constructors | PO Box 26334 | Alexandria, VA 22314 | Tel: 703.683.4999 | Fax: 571.527.3105 | www.professionalconstructor.org


careers, after earning a Master of Science in Con-struction Management degree and secure a Master in Business Administration, for instance (Chan et al. 2002).

The data shows the Ph.D. as the educational level with the lowest percentage of responses. The low numbers may be attributed to a lack of Ph.D. pro-grams available for construction graduates who want to continue on the same course. Burt and Hatipkara-sulu 2007 speculate that in times of full employment by the industry, graduates of bachelor’s and master’s programs are eager to enter the workforce and pro-duce, and since their career tracks incremental ad-vances in the industry, there is no perceived need to return to academia and re-tool or change careers (for example: from industry to academia) (Bilbo et. al 2000). Burt and Hatipkarasulu (2007) note that the industry as a whole has indicated that there is little additional value in master’s and Ph.D. programs as attested by the current research findings (See Fig-ure 1 between 1 to 1.7% PhD are reported in three surveys, it is inferred that if the industry needed

PhD, there would be more). Most building construc-tion Ph.D.’s come from civil engineering. Accord-ing to Williamson (1999) few building construction (or construction science) programs throughout the nation provide a Ph.D. aligned with a College of Ar-chitecture, rather than civil engineering aligned with a College of Engineering. The nature of the industry (Fernandez-Solis 2008) plays a part in this as well, considering that there are not many positions within the industry that require a Ph.D. or research experi-ence Burt and Hatipkarasulu (2007).

Figure 3 shows the data indicating the graduates’ professional associations and certifications. This data were not sought in the 1987 survey, and so is not included in the current analysis. The highest per-centage (an average of 48%) of responses was in the category of “Other,” probably due to the large num-ber of associations in which graduates can take part, many of which were not listed as choices for these surveys. Among those associations most often listed as “Other (41.9% in 2004 and 53.6% in 1998)” were (as a whole) the Texas Society of Architects (TSA),

Figure 2. Education since graduation by education level – Master’s Degree

Figure 3. Professional association and certification averages

2.6

7.6 7.9

3.9

0.0

11.6

20.3

16.7

0.0

5.0

10.0

15.0

20.0

25.0


Graduate Group

Perc

ent

19982004

Figure 4. Professional associations – Realtor. JUNE 2010 — Volume 33, Number 3The American Institute of Constructors | PO Box 26334 | Alexandria, VA 22314 | Tel: 703.683.4999 | Fax: 571.527.3105 | www.professionalconstructor.org


the National Council of Architectural Registration Board (NCARB), and the American Society of Civil Engineers (ASCE). Also listed under the category of “Other” were many state licenses and certifications. The survey was not intended to capture the level of detail of all the different associations but specified those that are most common among building con-struction professionals.

Figure 4 shows the percentage of responses for the Realtor category, distributed across the Graduate Groups for the 1998 and 2004 surveys only. The percentage of graduates possessing a realtor’s license at the time of the 1998 survey is noticeably lower than that for the 2004 survey for Graduate Groups 2, 3, and 4. The larger percentages in 2004 indicate the desirability of having a realtor’s license during the housing boom during the early 2000’s (Mankiw and Weil, 1989; Weller 2006).

Both the Associate Constructor (AC) and Certified Professional Constructor (CPC) certifications have been in existence since 1996; however neither AC nor CPC were listed as responses in the 1998 survey, so this data is unavailable for comparison with the 1987 survey. Beginning in the 2004-2005 academic year, COSC students in their last semester of stud-ies were required to sit for the Associate Construc-

tor (AC) exam, leading to professional certification. The data shows that 40% of those in Graduate Group 1 (those that have graduated up to 8 years ago from the date of this research) hold the AC certification, in contrast with zero responses from Graduate Group 2 – from 9 to 18 years, Group 3, - from 19 to 23 years and Group 4 - 29 years and more, see Table 2). The percentage of those taking the AC exam are no longer predicted to rise in the future, as the AC exam was dropped as a requirement for graduating seniors in 2008. To sit for the CPC exam, AC certifi-cation along with seven years of construction project management experience is required. Although the percentage of responses from the 2004 survey are low (23.6% in comparison with previous surveys), the numbers were expected to rise as more gradu-ates became AC certified and gained the necessary project management experience required to take the CPC examination. This has not been shown to be the case, as the numbers taking the CPC remain far below previous years as the incentive to take the exam was removed from graduation requirements. There is little, if any, indication that this will change in the foreseeable future.

Curriculum Content

Graduates were asked to rate the value of particu-

Figure 4. Professional associations – Realtor.



lar course clusters, see Table 3, with regard to the graduates’ past and present occupational responsibil-ities. For example Structure has a cluster of courses in statics, dynamics, steel and reinforced concrete coursework. The courses were rated according to their value, given the choices of “No Value,” “Little Value,” “Some Value,” “Valuable,” and “Extremely Valuable.” Those ratings were then converted to numerical values of 1 through 5, respectively. The data shows that all course clusters rate in the 3.5 to 4.5 range, from “Some Value” to “Valuable.” Be-cause all course clusters fell into this range, no at-tempt has been made to determine statistical signifi-cance. Once the numerical average was figured for each course cluster, the clusters were ranked for each survey year (number one being the highest rank), as shown in Table 3 with the exception of the Materials and Methods course cluster, due to a survey instru-ment error in 2004 when it was left out and thus this cluster is not considered on the table.

For informational reference, if Materials and Meth-ods were reported in the 1987 and 1998 survey years, it would rank first and third, respectively, among the other course clusters.. This is a decrease in perceived value over the survey time span. It is important to note that this decrease in ranking corresponds to an increase in the rankings of the Professional and Managerial and Estimating and Scheduling course clusters from second and third, respectively, in 1987, to first and second, respective-ly, in 1998. These changes appear to be due to an increased number of graduates employed in project management positions, therefore increasing the

perceived value of the Professional and Managerial and Estimating and Scheduling course clusters over Materials and Methods.Of the course clusters analyzed across all three sur-vey years, the Professional and Managerial and Esti-mating and Scheduling clusters rank first or second, respectively, in all three survey years. The Science cluster ranks last across all three surveys. Increasing in perceived value between 1987 and 1998 are the course clusters of Math and English and Humani-ties. Decreasing in perceived value between 1987 and 1998 are the course clusters of Legal Aspects, Structures, and Mechanical and Electrical. With the exception of the Mechanical and Electrical and Structures course clusters, all other course clusters were ranked the same in 2004 as in 1998. The Me-chanical and Electrical cluster increased in perceived value, while the Structures cluster decreased in per-ceived value between the 1998 and 2004 surveys.

These rankings infer that the majority of graduates from the Department are employed in manage-ment positions, and thus most often utilize the skills gleaned from the top ranked course clusters in their occupational responsibilities. The increasing per-ceived value of the English and Humanities course cluster indicates an increasing demand from the con-struction industry for graduates to possess well de-veloped oral and written communication skills used in Request for Qualifications (RFQ) and Request for Proposals (RFP) proposals, contract negotiations, and executive transactions.

Graduates were also asked to indicate “Curriculum Table 3. Course Cluster Rankings*

Ranking 1987 1998 20041 Professional and Managerial Professional and Managerial Professional and Managerial2 Estimating and Scheduling Estimating and Scheduling Estimating and Scheduling3 Legal Aspects Math Math4 Structures Legal Aspects Legal Aspects5 Math English and Humanities English and Humanities6 Mechanical and Electrical Structures Mechanical and Electrical7 English and Humanities Mechanical and Electrical Structures8 Science Science Science

* Rankings of the Materials andMethods course cluster are not included due to survey instrument error in 2004



Areas for Increased Emphasis.” Topping the list are the curriculum areas of: (a) Estimating and Sched-uling; (b) Professional and Managerial; (c) Project Management; and (d) Legal Aspects. Three of these curriculum areas are also among those rated “Valu-able” in the course rating section of the survey. The fact that the top ranking course clusters top the list of curriculum areas needing increased emphasis serves to highlight the Estimating and Scheduling, Professional Managerial, and Legal Aspects course clusters as the most crucial of the curriculum areas which need increased emphasis. Other curriculum areas mentioned in all three survey years as needing increased emphasis are: (a) Finance/Accounting; (b) Mechanical/Electrical; (c) Technical Writing; (d) Computer Applications; (e) Materials and Methods; (f) and Speech Communications. By continuing to focus on the evolution and improvement of the courses pinpointed by these surveys as being most valuable and most in need of increased emphasis, construction education can continue to meet the demands of the construction industry.

Gender and Ethnicity

Increased emphasis has been placed on diversity within both the construction industry and construc-tion education programs. In an effort to track the progress made because of such an emphasis, gender and ethnicity profiles were gathered for the compa-nies which employ those graduates surveyed in both 1998 and 2004. Figure 5 shows the gender profile averages for the two surveys.

Figure 5. Gender make up of employing companies.

The number of males employed decreased and the number of females employed increased by the same 3% between the 1998 and 2004 surveys (since the sum is always 100%). This information appears consistent with a predominantly male workforce that is gradually becoming more diverse and closely mir-rors the gender ratios of the Department of Construc-tion Science’s student population.

Figure 6 shows the ethnic makeup of the companies employing the surveyed graduates.

Figure 6. Ethnic make up of employing companies.

The number of White/Anglo employees has de-creased by 14% while the numbers of employees who are African American, Hispanic, and other ethnicities have increased by six percent, six percent, and two percent respectively. This data also dis-plays a diversifying construction industry. The 14% decrease in the number of White/Anglo employees shows that there have been greater steps in ethnic diversification than for gender diversification, with a decrease of only 3% in the number of male employ-ees. With the number of construction degree pro-grams available at major universities and a relative large number of scholarships from the construction industry available, there is opportunity and entice-ment for minorities, both gender and ethnic, to take part in construction degree programs but perhaps not to persist in the workforce, as in other professions.

Market Sector

Sector data was collected from all three surveys and is summarized in Figure 7. The Commercial

84

16

81

19

0

20

40

60

80

100

Males Females

Perc

ent

19982004

Figure 5. Gender make up of employing companies.

68

4

199

54

10

25

11

0

20

40

60

80

White/A

nglo

African

America

n

Hispan

ic

Intern

ational

Perc

ent

19982004

Figure 6. Ethnic make up of employing companies.



sector (36.8%; 50.6%; and 50.3%) continues to be the leading employing sector for graduates of the Construction Science program. The second highest percentage of graduates indicated Residential (9.2%; 11.8% and 21.7%) as the sector in which they cur-rently are employed, followed by Architect/Engineer (7.9%; 9.2% and 6.8%) and Industrial (10.1%; 8.3% and 5.0%). There has been growth in the number of graduates employed in the Commercial and Resi-dential sectors of the construction industry. Most notable is the approximately 10% change in the Residential sector between the 1998 (11.8%) and 2004 (21.7%) surveys, which may be attributed to the housing boom in the early 2000s.

Figure 7. Sector averages

Shifts toward Project Management

The data in Figure 8 shows that, on average, most graduates of the COSC are employed as project managers or project engineers (PM/PE) (an average of 27%). There are also large numbers of graduates (average 21%) who are either owners or manage-ment officers of their firm. This shows that, on average, graduates are successfully progressing into management positions. Most notable are the ris-ing percentages of PM/PE over the years. Figure 9 shows the PM/PE data distributed across the Gradu-ate Groups. This data shows the rising percentage over the years for all Graduate Groups, with the exception of Graduate Group 4. The data shown for

Graduate Group 1 displays a positive response by the Department of Construction Science to the indus-try’s demand to produce an increased numbers of graduates prepared for project management positions by increasing admissions while rising the average GPA, with changes of approximately 12% between the 1987 and 1998 surveys, and approximately 10% between the 1998 and 2004 surveys. The lower numbers of Project Manager/Project Engineers for Graduate Groups 2, 3, and 4 are most likely due to promotions to Management Officer or Owner posi-tions. This conclusion is derived from the trend of PM/PE decreasing from those in the 1987 while high in 2004. Apparently the alumni move from entry po-sition to those of estimators, schedulers, cost control

purchasing and owner in time, as expected in trends in other profes-sions.

Figure 10 shows the Estimator data distributed across the Gradu-ate Groups. The data for Graduate Group 1 shows that a lower per-centage of

graduates are beginning their careers in estimat-ing, with an approximately 9% drop since the 1987 survey, reflecting the trend that higher percentages of graduates are starting their careers in Project Man-agement/Project Engineer positions.

There are very low percentages of graduates in the Scheduler, Cost Control, and Purchasing positions across all Graduate Groups, indicating that these positions are not often taken by graduates of the Department. Although the percentages for these three positions are low for all Graduate Groups, the data show that many more graduates(see figure 8 for graphic symbolisms) in Graduate Group 1 at the time of the 1987 survey were taking positions in scheduling, cost control, and purchasing, than those

8.3

21.7

6.36.07.9

N.A.9.3

10.1

36.8

5.8 3.3

11.8

50.6

9.26.7 5.84.3 4.02.5

6.83.1

6.55.0

50.3

0102030

405060

Commerc

ial

Industr

ial

Reside

ntial

Heavy

/High

way

Constr

uctio

n Man

agemen

t

Archite

ct/Eng

ineer

Military

Govern

ment

Sector

Percen

t 198719982004



Figure 8. Employment position averages

in Graduate Group 1 at the time of the 1998 and 2004 surveys.

This data, along with the data for the categories of Estimator and PM/PE, support the fact that graduates are now starting their careers in management related positions.

Job Experience

Job experience information was gathered through questions of how many employers the surveyed graduates have had since graduation and how long the surveyed graduates have been working with their current firm. Table 4 shows the average number of employers that each Graduate Group has had sincegraduation, along with the average number of years that each Graduate Group has worked with their current firm. The data shows, as expected, that the average number of employers is lowest for those graduates in the first Graduate Group and the num-ber of employers increases for the subsequent Grad-uate Groups. The average number of employers is consistent for each graduate group across the survey years (1987- 2.57%; 1998 – 3.01%; 2004 – 3.38%).

Table 4 also shows that the relative average of num-ber of years with the current firm across the survey years is approximately the same (1987 – 7.74 yrs; 1998 – 6.01 yrs; 2004 – 7.06 yrs). Economic condi-tions should have a strong influence on these numbers but this survey was not designedto capture the

data. However the industry goes through down-times approximately every ten years which may be an equalizer in the numbers that are shown in this part of the survey. Figure 11 shows the average number of years work experience the graduates have had with their current firm at the time of the survey. As expected, this data also shows that the average number of years work experience with the graduates’ current firms is lowest for the first Graduate Group and higher for subsequent Graduate Groups. The Department’s graduates are enjoying steady employ-ment with their firms.

Table 4: Work Experience SummaryAvg. # of Employers Avg. # of Yrs. With

Current FirmGradu-

ate Group

1987 1998 2004 1987 1998 2004

Group 1 (0-8 yrs.)

1.74 1.89 1.75 2.33 2.01 3.16

Group 2 (9-18 yrs.)

2.83 3.11 3.30 6.33 5.62 5.98

Group 3

(19-28 yrs.)

4.16 4.01 4.34 13.92 9.99 8.66

Group 4 (29+ yrs.)

N.A 4.41 4.62 N.A 15.22 14.55

ALL 2.57 3.01 3.38 7.75 6.01 7.06

4.5

11.4

19.5

13.4

3.6

27.6

4.80.9

23.2

18.6

4.8

29.4

6.0

1.1 0.9 0.43.5

4.04.1

21.4

12.3

21.3

0.30.805

101520253035

Ow

ner

Man

agem

ent

Offi

cer

Supe

rinte

nden

t

Proj

ect

Man

ager

/Pr

ojec

tEn

gine

er

Estim

ator

Sche

dule

r

Cos

t Con

trol

Purc

hasi

ng

Position

Perc

ent 1987

19982004

Figure 8. Employment position averages



For Graduate Group 3, the average number of years experience with the graduates’ current firms decreased between the 1987 and 2004 surveys. This indicates greater job mobility for those graduates at the peak of their careers (see Figure 11 and Figure 12 where apparently the greater earning potential is achieved between 19 and 28 years of experience), which may be due to the widening supply-demand gap in the construction industry, and thus, the high demand for experienced workers (see the increased salary rate between the 2004 and 1994 alumni across all groups in Figure 12). This demand will cause companies within the industry to make attractive offers to experienced workers, enticing them to change employers, as evi-denced by the high average adjusted salaries reported by Graduate Group 3 for the year 2004 with respect to those reported in 1987. Average Salaries

The salary figures reported by the respondents were averaged for each of the Graduate Groups. From the reported salaries alone, it appears that average salaries have been rising through the years for all Graduate Groups. However, in order to accurately assess the changes in salaries reported over the years, adjust-ments were made to the salary averages and salary ranges reported for the Graduate Groups. The salary adjustments were made utilizing annual average CPIs for each survey year, found in a database of the Con-sumer Price Index available from the U.S. Department

of Labor’s Bureau of Labor Statistics (2007). The an-nual average CPIs for the years 1987, 1998, and 2004 are 113.6, 163.0, and 188.9 respectively. To adjust the reported salaries from the 1987 and 1998 surveys to 2004 dollars the following formula was used:

Sadj = Sy (1 + (CPI2004 - CPIy) / CPIy)

where, y = year of survey to be adjusted (either 1987 or 1998)

Sadj = salary adjusted to 2004 dollars Sy = unadjusted salary for year y CPI2004 = annual average CPI for 2004 CPIy = annual average CPI for year y

Table 5 shows the reported average salaries for all Graduate Groups, for the surveys conducted in 1987, 1998, and 2004, along with the salaries adjusted for inflation in 2004 dollars. Figure 12 shows a trend of increasing salaries for each Graduate Group. One pos-sible reason for the increasing salaries is the increasing gap between the supply of construction graduates and the industry’s demand for those graduates (per in-crease in start-up salary per Figure 12). The increased difference in salary levels between the surveys con-ducted in 1998 and 2004 for Graduate Groups 3 and 4, which is higher than those for Graduate Groups 1 and 2, indicates that graduates with more years of experi-ence are in greater demand than their less experienced

Table 5: Average Salaries as Reported and Adjusted for Inflation (2004 dollars) by Graduate

Group

Group 1 (0-8 yrs.) Group 2 (9-18 yrs.) Group 3 (19-28 yrs.) Group 4 (29+ yrs.) Survey Year

Reported Adjusted

Reported Adjusted

Reported Adjusted

Reported Adjusted

1987

$33,156 $55,134

$55,857 $92,882

$80,222 $133,397 N.A. N.A.

1998 $51,391 $59,557

$88,390 $102,435

$114,027 $132,145

$91,104 $105,580

2004

$62,177 $62,177

$115,734 $115,734

$169,885 $169,885

$163,982 $163,982

28.3 25.2

10.3

40.6

27.5

13.3 12.2

50.4

33.7

15.210.9

N.A.0.0

10.020.030.040.050.060.0


Graduate Group

Perc

ent 1987

19982004

Figure 9. Employment positions – Project Manager/Project Engineer

16.6

12.3

5.26.03.9 4.1

7.8

3.9

6.3

N.A.

4.67.3

0.02.04.06.08.0

10.012.014.016.018.0


Graduate Group

Per

cent 1987

19982004

Figure 10. Employment positions – Estimator

2.33

6.33

13.92

2.01

5.62

9.99

3.16

5.98

8.66

N.A.

15.22 14.55

0.002.004.006.008.00

10.0012.0014.0016.00


Graduate Group

Year

s 198719982004

Figure 11. Average number of years work experience with current firm



Figure 9. Employment positions – Project Manager/Project Engineer

Figure 10. Employment positions – Estimator

Figure 11. Average number of years work experience with current firm

55,134

92,882

133,397

59,557

102,435

132,145

105,580

62,177

115,734

169,885163,982

N.A.0

30,000

60,000

90,000

120,000

150,000

180,000


Graduate Group

Sala

ry (2

004

$)

198719982004

Figure12. Average salaries by Graduate Group adjusted for inflation to 2004 dollars.




counterparts as the supply-demand gap widens. Those in Graduate Group 3 are enjoying greater job mobility due to the increasing supply-demand gap, and thus, are being offered greater salaries to change companies, or greater salaries to entice them to stay with their current employer. Another possible ex-planation for the increasing salaries over the years is the construction industry’s recovery after the reces-sion of the late 1980’s, leading to a rise in construc-tion spending and the recovery of theeconomy as a whole (Weller 2006). Salaries reported by Group 4 are slightly lower than those in Group 3, indicating that graduates in Group 4 on average have reached the peak of their careers and some in that group are nearing or have reached retirement age, see Figure 12).

CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE RESEARCH

As a longitudinal study, consisting of three indi-vidual studies for the Department of Construction Science at Texas A&M University, the data gathered through the surveys may be used together as an additional departmental self assessment tool. The objective of this longitudinal study was to explore the trends in average salaries, job experience, curriculum ratings, and professional development of graduates from the Department of Construction Science over time. It is useful for both the Department and the construction industry as a whole by giving educators and employ-

ers an accurate overview of the graduates’ develop-ment as professionals, as well as their views on the effectiveness of the curriculum to which they were exposed while at the university.Approximately half of the graduates surveyed received no additional education after graduating. Approximately 28% of the graduates received some additional education after graduation without earn-ing an additional degree. Of those who did earn an additional degree (average of 12%), a master’s degree was the most common (average of 11%). The construction industry has a demand for experi-enced graduates with management knowledge, and as construction graduates are progressing through their careers, they are taking steps to gain the knowl-edge necessary to take on management or ownership positions by acquiring additional degrees. Because of the large number of professional associations and certifications available to graduates, all of the categories listed in this section of the survey instru-ment received low percentages of responses with the exception of the “Other” category which was the most common category chosen (approximately 50%). The number of graduates holding the AC and CPC certifications is not expected to rise because the requirement for seniors to take the AC examination was dropped in 2008 unless the industry provides additional incentives that were deleted by academia.When asked to rate the value of several coursework categories, regardless of how long the graduates had been out of school and what year the surveys were taken, all coursework categories rated between ap-

Figure12. Average salaries by Graduate Group adjusted for inflation to 2004 dollars.



proximately 3.5 and 4.5, corresponding to “Some Value” to “Valuable” categories. When asked what courses needed increased emphasis, the following course areas were most often mentioned: Estimating and Scheduling, Professional and Managerial, Proj-ect Management, and Legal Aspects. Three of the curriculum areas identified as needing increased em-phasis were also among those rated “Valuable” in the course rating section of the survey. This information serves to highlight the most crucial of the curriculum areas needing increased emphasis as Estimating and Scheduling, Professional and Managerial, and Legal Aspects.

The number of males employed has decreased and the number of females employed has increased by 3% between the 1998 and 2004 surveys. This is consistent with a predominantly male workforce that is gradually becoming more diverse. These figures closely mirror the gender ratios (82% to 18% on the average) of the Department of Construction Sci-ence’s student population. In terms of ethnicity, the 14% decrease in the number of White/Anglo em-ployees shows that there have been greater steps in ethnic diversification than for gender diversification. The survey assumes that the number of male survey respondents in a set is approximately the same from year to year.

The Commercial sector employs the highest num-ber of graduates from the Department, followed by Residential, Architect/Engineer, and Industrial, re-spectively. There has been growth in the number of graduates employed in the Commercial and Residen-tial sectors. Most notable is the approximately 10% change in the Residential sector between the 1998 and 2004 surveys, which may be attributed to the housing boom in the early 2000s as well as suburban sprawl.

Most graduates from the Department are employed as PM/PE. There are also large numbers of gradu-ates who are either owners or management offi-cers of their current firm. The data indicates a shift

towards project Management. In particular, the increasing percentages of PM/PE in Graduate Group 1 over the years displays a positive response of the Department of Construction Science to the industry’s demand to produce an increased number of gradu-ates prepared to enter project management positions immediately upon graduation, with changes of ap-proximately 12% between the 1987 and 1998 sur-veys, and approximately 10% between the 1998 and 2004 surveys.

Graduates of the Department of Construction Sci-ence are enjoying steady employment and job mobility across all Graduate Groups for all surveys. For Graduate Group 3, the number of years experi-ence with the graduates’ current firms decreased between the 1987 and 2004 surveys. This indicates greater job mobility for graduates at the peaks of their careers, stemming from a higher demand for experienced employees as the supply-demand gap for workers widens. The average salaries adjusted for inflation indicate that graduates’ salaries have risen over the years. One reason is the widening gap between the number of graduates being supplied to the industry from construction education programs and the industry’s demand for those graduates as measured by placement and increased initial salaries. For Graduate Groups 3 and 4, the jumps in salaries between the 1998 and 2004 surveys are larger than those for Graduate Groups 1 and 2, indicating an even greater demand for experienced workers.Future studies will help expand our knowledge base and continuously improve construction education and the construction industry. In order to remain re-sponsive to changes in the construction industry and to provide undergraduate construction students with the best possible academic program, it is important that the Department of Construction Science con-tinue to periodically assess the needs of the regional industry. It should continue to use the assessments in order to make informed decisions about program development and to provide an up-to-date curricu-lum that meets the current and future demands of the construction industry.



REFERENCES

Bilbo, D. (1992). Construction education at Texas A&M University three decades on the job. American Professional Constructor, 16(2), 2-8.

Bilbo, D., Collins, C., Waseem, M., and Burt, R. (2007). A study of the supply and demand for con-struction education graduates. Proceedings of the 43rd Annual Conference of the Associated Schools of Construction. Study retrieved April 25, 2007, from http://www.asceditor.usm.edu/archives/2007/CERT82002007.pdf

Bilbo, D., Fetters, T., Burt, R., Avant, J. (2000). A study of the supply and demand for construction education graduates. Journal of Construction Educa-tion, 5(1), 78-89.

Burt, R,, Hatipkarasulu, Y. (2007). Evolution of Construction Education in the United States: A Case Study. Proceedings of the Construction and Build-ing Research Conference of the Royal Institution of Chartered Surveyors (RICS), September 6-7 2007, Atlanta, GA.

Chan, E. H. W., Chan, M. W., Scott, D., and Chan, A. T. S. (2002). Educating the 21st century construc-tion professionals. Journal of Professional Issues in Engineering Education and Practice. 128(1), 44-51.

Chinowsky, P.S., and Vanegas, J.A. (1996). Com-bining practice and theory in construction education curricula. Proceedings of the 1996 ASEE Annual Conference, Session 1221.

Department of Construction Science (2006). His-tory of COSC. Retrieved December 26, 2007, from Texas A&M University, Department of Construction Science Web site: http://archone.tamu.edu/cosc/about.html

Fernández-Solís, J. L. 2008, “Is Building Construc-tion Approaching the Threshold of Becoming Unsus-

tainable? A System Theoretic Exploration Towards a Post-Forrester Model for Taming Unsustainable Exponentialoids,” VDM and Co. Saarbruecken, Germany.

Fernández-Solís, J. L. 2007, “The Exponentialoid of Resource Consumption,” Proceedings, CIB World Building Congress 2007, Cape Town, South Africa, Proceedings CIB-457.

Mankiw, G. N., and Weil, D. N., 1989. “The Baby Boom, the Baby Bust and the Housing Market,” Re-gional Science and Urban Economics. 19, 235-258.

Newitt, J. (1987, April). Validation of construction management curriculum. Paper presented at the 23rd Annual Conference of the Associated Schools of Construction. Abstract retrieved November 5, 2005, from http://www.asceditor. usm.edu/ar-chives/1987/Newitt87.htm

U.S. Department of Labor (2004). Bureau of Labor Statistics frequently asked questions. Retrieved De-cember 8, 2006, from http://www.bls.gov/cpi/cpifaq.htm#Question_1

U.S. Department of Labor (2007). Consumer price index. Retrieved April 26, 2007, from ftp://ftp.bls.gov/pub/special.requests/cpi/cpiai.txt

Weller, C. E., 2006. The End of the Great American Housing Boom – what it means to you, me and the U.S. Economy. Center for American Progress, Wash-ington, D.C.

Williamson, K., 1999. A Road Map to an effective Graduate Construction Education Program. Journal of Construction Education 4(3) 260-277. Tables

Reviewer/Publication Interest SurveyThe American Professional Constructor is a refereed journal published two times a year by the American Institute of Constructors (AIC). Each author’s manuscript submission is given a blind review by three AIC members. to evalu-ate the content and style, and appropriateness as either a general interest or scholarly publication. Based upon the decision of the reviewers, each article is accepted or rejected for publication. Acceptance can be predicated upon incorporation of reviewer comments.

Approximately 10 articles are annually published. At present, 60% of the articles submitted are rejected by the refer-ees. To maintain the high standard of published articles in the journal. AIC requires 50 to 60 reviewers annually. Members are periodically polled to express their willingness to serve as a reviewer or referee. For each member that is willing to provide this valuable service, it is necessary for them to identify their area(s) of expertise or interest. If you have served as a reviewer and wish to continue doing so, or if you have not served as a reviewer and would like to do so, please take five (5) minutes to complete the survey below. If you would like to publish a manuscript in the Journal of American Institute of Constructors, the similar topic areas given consideration is provided here also. Please submit the reviewer’s interest or submit your manuscript to:

Don Jensen, JD, Ph.D. Western Carolina University The Kimmel School of Construction Management & Technology Construction Management Department 211 Belk Building, Cullowhee, NC 28723 Fax: (828) 227-2201 Email: [email protected]

Please place a mark beside each keyword that is a topic area indicating your expertise or interest. Thank you, in advance, for serving as a reviewer for The American Professional Constructor.

Name: ______________________________________________________ Member No.:_____________________E-Mail: ______________________________________________________ Phone No.:_____________________Address: ___________________________________________________________________________________________________________________________________________________________________________________

_____________________________________________________________________________________________

Topic Areas

[ ] Computer Applications

[ ] Construction Safety

[ ] Estimating

[ ] Financial Management

[ ] Personnel/Human Resource Management

[ ] Contract Law and Legal Applications

[ ] Materials and Methods

[ ] Project Management

[ ] Steel Construction

[ ] Concrete Construction

[ ] Design-Build Construction

[ ] Mechanical Construction

[ ] Contract Documents

[ ] Strategic Planning

[ ] Planning and Scheduling

[ ] Site Management

[ ] Marketing and Sales

[ ] Community Planning

[ ] Labor Relations

[ ] Quality Management

[ ] Productivity

[ ] Cost Control

[ ] Undergraduate Education

[ ] Graduate Education

[ ] Wood Construction

[ ] Masonry Construction

[ ] Heavy/Highway Construction

[ ] Electrical Construction

[ ] Residential Construction

[ ] International Construction

[ ] Architecture

[ ] Real Estate and Factors Affecting Contractors

[ ] Housing and Related Issues

[ ] Procurement

[ ] Bonding

[ ] Bidding

[ ] Ethics

[ ] Commercial Construction

[ ] Industrial Construction

[ ] Utilities Construction

[ ] Institutional Construction

Other

____________________

____________________

____________________

Journal of the American Institute of Constructors ???

101

The American Professional Constructor is a refereed journal published two times a year by the American Institute of Con-structors (AIC). Each author’s manuscript submission is given a blind review evaluation by three AIC members regarding manuscript content and style, and appropriateness as either a general interest or scholarly publication. Based upon the decision of the reviewers, each is accepted or rejected for publication. Acceptance can be predicated upon incorporation of reviewer comments.

Approximately 10 articles are annually published. At present, 60% of the articles submitted are rejected by the referees. To maintain a high standard of quality published articles, AIC requires 50 to 60 reviewers annually. Members are periodi-cally polled to express their willingness to serve as a reviewer or referee. For each member that is willing to provide this valuable service, it is necessary for them to identify their area(s) of expertise or interest. If you have served as a reviewer and wish to continue doing so or, if you have not served as a reviewer and would like to do so, please take five (5) minutes to complete the included survey. If you would like to publish a manuscript in the Journal of American Institute of Constructors, the similar topic areas given consideration is provided below also. Please submit the reviewer’s interest survey or submit your manuscript to:

Constructor CertificationThe Professional Certification for the Practicing Constructor

Distinguish yourself in the Industry

Join the over 12,000 professionals who have sought the Certified Professional Constructor (CPC) and Associate Constructor (AC) designations.

Certification Examination Schedule

Mark your calendar for the following Constructor Qualification Examination (CQE) Nationwide Testing Dates

November 7th, 2009

Applications

Candidate handbooks and applications are available upon request from the AIC and Constructor Certification Commission National Offices, or may be downloaded from the Commission website at

www.ConstructorCertification.org. Paper and Interactive CD-ROM study guides are available for all test levels.

Standard Fees

CQE Level I Examination and Application Fee (Includes CD-ROM Study Guide) .....................................................$155.00CQE Level II Examination and Application Fee (For current ACs, includes CD-ROM Study Guide) .........................$370.00CQE Level II Examination and Application Fee (with Level 1 Exemption, includes CD-ROM Study Guide ) ............$480.00

AIC Constructor Certification CommissionPO Box 26334, Alexandria Virginia, 22314

Tel 703.683.5053 | Fax: 703.683.0018

AlABAMAAuburn University

AlASKA University of Alaska-Anchorage

ArIZONA Arizona State UniversityNorthern Arizona University

ArKANSASJohn Brown UniversityUniversity of ArkansasUniversity of Arkansas

CAlIFOrNIA California State UniversityCalifornia State University, Chico

COlOrADOColorado State University

CONNECTICuTCentral Connecticut State University

IDAHOBoise State University

IllINOISIllinois State UniversitySouthern Illinois University

INDIANAIndiana State University

Purdue University

IOWA Iowa State UniversityUniversity of Northern Iowa

KANSASKansas State UniversityPittsburg State University

KENTuCKyEastern Kentucky University

IllINOISAGC Builder’s Association

MAINEUniversity of Maine

MASSACHuSETTS AGC of Massachusetts

MICHIgANEastern Michigan UniversityFerris State UniversityMichigan State University

MINNESOTAMinnesota State University - MoorheadMinnesota State University - Mankato

MISSISSIPPIUniversity of Southern Mississippi

MISSOurI

Central Missouri State University

MONTANAMontana State University

NEBrASKAUniversity of Nebraska

NEVADAUniversity of Nevada

NEW JErSEyNew Jersey Institute of Technology

NEW MExICOUniversity of New Mexico

NEW yOrKPratt InstituteSUNY College of Env ScienceAlfred State College

NOrTH CArOlINAEast Carolina UniversityWestern Carolina University

NOrTH DAKOTANorth Dakota State University

OHIOBowling Green State UniversityUniversity of Akron

OKlAHOMA

Oklahoma State UniversityUniversity of Oklahoma

PENNSylVANIATemple University

rHODE ISlANDRoger Williams University

SOuTH CArOlINAClemson University

SOuTH DAKOTASouth Dakota State University

TExASNorth Lake CollegeTexas A&M UniversityUniversity of Houston

uTAHBrigham Young UniversityWeber State University

WASHINgTONCentral Washington UniversityUniversity of Washington

WISCONSINMilwaukee School of EngineeringUniversity of Wisconsin-Stout

Nationwide testing locations

AICJournal2009.indd 6 6/16/2009 2:53:12 PM

March 27, 2010 – November 6, 2010

www.professionalconstructor.org

Journal of the American Institute of Constructors ???

102

November 6, 2010 - April 2, 2011

www.professionalconstructor.org.

700 N. Fairfax St., Suite 570, Alexandria VA, 22314

$405.00$535.00

AmericAn institute of constructorsConstructor Code of Ethics

The Construction Profession is based upon a system of technical competence, management excellence and fair dealing in undertaking complex works to serve the public safety, efficiency, and economy. The members of the American Institute of Constructor are committed to the following standards of professional conduct:

I. A Constructor shall have full regard to the public interest in fulfilling his or her responsibilities to the employer or client.

II. A Constructor shall not engage in any deceptive practice, or in any practice which creates an unfair advantage for the Constructor or another.

III. A Constructor shall not maliciously or recklessly injure or attempt to injure, whether directly or indirectly, the professional reputation of others.

IV. A Constructor shall ensure that when providing a service which includes advice, such advice shall be fair and unbiased.

V. A Constructor shall not divulge to any person, firm, or company, information of a confidential nature acquired during the course of professional activities.

VI. A Constructor shall carry out responsibilities in accordance with current professional practice, so far as it lies within his or her power.

VII. A Constructor shall keep informed of new thought and development in the construc-tion process appropriate to the type and level of his or her responsibilities and shall support research and the educational processes associated with the construction profession.

103

Publications Department

American Institute of Constructors PO Box 26334 Alexandria, Virginia 22314

ADDRESS SER VICE REQUESTED

700 N. Fairfax St., Suite 570

american professional constructor journal - june 2010

Documents

construction industryissues

construction profession

related fields of construction

dbia sundt construction

individual constructor

professional relationships

aic erie affiliates

e r ic