application of tools of quality in engineering education

APPLICATION OF TOOLS OF QUALITY IN ENGINEERING

EDUCATION

A project work report submitted By

SYED ASGHAR RAZA IMAM ABHINAV GUPTA IVTH YEAR B.E. IP IVTH YEAR B.E. IP040908049 040908050

Under the Guidance of

Mr. P. KRISHNANANDA RAO SL. GRADE LECTURER

DEPARTMENT OF MECHANICAL AND MANUFACTURING ENGG.

MIT, MANIPAL

ABSTRACT

This project report provides an overview of how best the quality principles, quality tools could effectively be used in improving the quality of engineering education.

For our analysis we have considered applying the quality tools to analyze the performance of 1st year students of Manipal Institute of Technology, Manipal University.

INTRODUCTION

QUALITY IN EDUCATION

Quality in education can be defined as fitness for purpose,

effectiveness in achieving institutional goals, meeting

customers stated or implied needs, degree to which

education prepares students to be personally effective and

capable within the circumstances of their life and work.

NEED FOR PROJECT

In today’s world of cut-throat competition various factors demand change in the education system. They are:

Global wave of Liberalization, Privatization, and Globalization(LPG), applicable to education( such as multinational universities/institutes)

Increased awareness among stakeholders about their education needs (students, employers, parents etc.)

Initiation of ranking, accreditation and comparison of institutions by regulatory bodies/ others.

Educational institutes should be managed ‘Business-like’ to identify the critical problems in their functioning, so that they become effective and productive in their operations. So we need proven good business quality practices and tools for:

•Designing and examining competitive educational products.

•Analyzing and evaluation of learning process

•Satisfying the varying HR demands

•Assuring the society and employers about quality of pass- outs.

OBJECTIVES

To apply the various tools of quality on the CGPA samples of 1st year

students and to identify the Human, Environmental, Resources,

Processes (HERP) factors likely to affect the performance of 1st year

students.

LITERATURE

TOOLS OF QUALITY

Simple tools can be used by any professional to ease the quality improvement process: Pareto diagrams, cause and effect diagrams, histograms, scatter diagrams, and control charts.

Histogram

In statistics, a histogram is a graphical display of tabulated

frequencies. A histogram is the graphical version of a table

that shows what proportion of cases fall into each of several

or many specified categories. A histogram can be constructed

by segmenting the range of the data into equal sized bins

(also called segments, groups or classes).

Histogram

Pareto Diagram

A Pareto chart is used to graphically summarize and display

the relative importance of the differences between groups of

data. A Pareto chart can be constructed by segmenting the

range of the data into groups (also called segments, bins or

categories).

Example of Pareto chart depiction

Control Charts

In statistical process control, the control chart, also known as the 'Shewhart chart' or 'process-behavior chart' is a tool used to determine whether a manufacturing or business process is in a state of statistical control or not.

If the chart indicates that the process is currently under control then it can be used with confidence to predict the future performance of the process.

A control chart consists of the following:

•Points representing measurements of a quality characteristic in samples taken from the process at different times [the data]

•A centre line, drawn at the process characteristic mean which is calculated from the data

•Upper and lower control limits (sometimes called "natural process limits") that indicate the threshold at which the process output is considered statistically 'unlikely'

Example of Control Chart

Types of control charts

Variables charts • X and R chart (also called

averages and range chart)• X and S chart• moving average–moving range

chart (also called MA–MR chart) • CUSUM (also called cumulative

sum chart) • EWMA (also called exponentially

weighted moving average chart)

Attributes charts • p chart (also called proportion

chart) • np chart • c chart (also called count

chart) • u chart

How to read control charts?

When a process is in controlled state, this means that:1) All points lie within the control limits, and2) The point grouping does not assume a particular form. We would therefore know that an abnormality hasdeveloped if• Some points are outside the control limits, or• The points assume some kind of particular form even

though they are all within the control limits. For example Runs, Trends, Periodicity and Hugging of central line

Cause and Effect Diagram

It is simply a diagram that shows the causes of acertain event. Causes in a typical diagram are normallyarranged into categories, the main ones ofwhich are:The 6 M's • Machine, Method, Materials, Measurement,

Man and Mother Nature (Environment)

Cause and Effect diagram

The Normal Curve

The normal curve for a population is specifically determinedby the mean and the standard deviation of the population.The mean locates the centre of the curve, and the standarddeviation determines the width of the curve. Areas under thecurve are given in terms of z values. The z value representsthe distance from the center measured in standard deviationunits. The formula to calculate z is

Z= (x - m) / s

The x is the measurement of interest, m is the populationmean, and s is the population standard deviation.

Where Σ = Sum of              X = Individual score               M = Mean of all scores               n = Sample size (Number of scores) S,= Standard deviation

The z value leads to area under the curve (from the normal curve table), which is probability, and that probability gives the desired percentage for x.

Normal curve

SAMPLING

Sampling can be defined as a process or technique ofselecting a suitable sample, representative of thepopulation from which it is taken, for the purpose ofdetermining the parameters or characteristics of thewhole population.

Systematic Random Sampling

The various steps to achieve systematic randomsampling are: • Number the units in the population from 1 to N.• Decide on the n (sample size) that you want or

need where (I=N/n=interval size).• Randomly select an integer between 1 to I.• Select every I-th unit.

PROJECT METHODOLOGY

In trying to use industrial standards in education, it becomes inevitable that education is modeled as a manufacturing process and students are viewed as products. Hence for our evaluation we have considered students as the ‘raw material’ to whom all the efforts of the systems are directed.

Based on the literature review done it was decided to apply quality tools in two phase’s i.e. first problem identifying tools such as histogram and control chart are applied on the CGPA samples collected. Systematic random sampling was to be carried out on the data collected. Second the tools for generating idea like the cause and effect diagram and the Pareto diagram will be used to identify the causes affecting the performance of first year students.

The project is being carried out in three phases.

Phase 1: Collection of Data and Sampling

Phase 2: Application of problem identifying tools.

Phase 3: Application of quality tools for identifying causes.

IMPLEMENTATION

PHASE 1: COLLECTION OF DATA AND SAMPLINGResults of 1st year students of batch 2010-2011 werecollected. Total batch capacity was found out to be 1270. Out of 1270 students 300 students were selected throughsystematic random sampling for our analysis andapplication of the previously mentioned tools. Thereforesample size n=300.

Total number of samples N=1270. interval size, I = N/n

I = 1270/300 = 4.23

Hence ‘2’ was selected as our random integer. Starting from the 2nd sample every 4th student was selected till we got 300 students. Following is the list of CGPA samples selected randomly for our analysis and application of the quality tools. The students were grouped into sub groups of size 10.

10 7.62 9.65 8.12 5.85 8.62 3.77 9.15 4.85 8.23

8.62 9.31 8.65 8.85 7.65 9.12 7.88 8.15 9.31 7.46

9.00 9.15 8.08 2.92 6.54 5.46 6.69 8.08 8.38 9.19

8.04 8.23 8.35 7.62 8.58 3.92 8.88 7.62 5.81 9.04

8.62 5.92 6.04 5.5 3.54 6.23 2.46 1.42 9 6.23

6.96 9.27 8.5 7.31 9.54 0.77 6.35 8.12 8.35 6.73

6.42 8.69 9.04 6.54 2.5 8.58 6.85 8.54 6.81 7.42

7.73 6.88 6.12 7.77 5.96 6.96 8.81 7.77 6.65 7.23

5.46 7.92 7.65 1.62 5.42 5.27 4.23 3.38 0.54 8.54

7.62 7.62 6.85 6.54 6 3.46 6.38 5.42 5.12 1.35

7.65 8.38 8.42 7.12 9.04 3.54 5.38 6.69 6.31 2.5

8.58 4.31 4.65 2.65 7.62 7.15 3.96 8.08 9.31 4.88

4.15 4.77 6.35 8.08 3.58 7 6.12 7.62 3.31 6.50

1.42 3.15 5.31 2.08 1.81 7.12 7.46 7.00 7.23 3.58

3.19 4.73 5.46 6.58 9.50 0.65 5.31 8.54 6.23 9.04

6.54 7.15 3.38 8.08 4.85 7.81 7.58 7.19 3.50 3.04

7.50 6.00 9.31 8.54 9.50 7.81 7.50 8.00 3.50 8.42

6.12 9.15 8.35 7.54 8.46 6.65 7.96 7.58 4.62 8.46

4.96 5.73 3.73 6.92 8.04 9.04 8.04 6.27 2.00 8.35

8.19 7.15 9.15 9.00 7.50 8.73 8.69 2.85 1.31 8.58

4.62 6.58 4.88 7.35 8.31 8.31 8.69 7.73 8.65 4.31

6.27 9.19 7.04 2.62 8.69 8.15 5.19 8.19 9.96 7.50

6.92 7.85 8.23 4.96 3.77 2.65 8.27 1.88 1.38 9.12

7.58 7.85 7.65 7.19 3.58 1.85 4.54 7.12 6.27 6.62

2.54 3.69 3.00 6.31 8.35 7.19 7.50 8.15 8.65 6.50

8.12 5.46 7.81 5.88 6.96 7.19 8.04 6.12 0.00 6.73

1.69 3.73 8.62 9.08 5.88 6.65 8.65 7.92 6.96 7.65

7.27 3.73 0.23 4.77 3.58 1.46 8.31 8.00 7.96 6.77

8.92 7.85 7.23 6.81 7.00 6.50 3.19 3.46 7.58 9.23

8.00 7.73 7.00 7.85 8.88 8.12 6.27 8.73 8.12 8.35

PHASE 2: APPLICATION OF PROBLEM IDENTIFYING TOOLS

• Histogram The CGPA samples were grouped into sub-groups of size 10.

Total number of sub-groups = 30. The data will then be used to construct the data in the following way:

1) Count the data, N 2) Record the largest values in each group as X and the

smallest value as S. Next, the largest X and the smallest S on the whole.

3) The range (R) of all the data is: R = X-S. This range can be divided into classes and the data belonging to each class can be enumerated. The number of classes (the number of histogram bars) can be determined on the basis of table.

Number of Data (N) Number of classes (K)

Under 50 5—7

50—100 6—10

100—250 7—12

Over 250 10--20

4) The class interval, h, which will be used as the horizontal graduation unit for the histogram is determined by dividing the range (R) by the number of classes. h= (X-S)/ K Now in this case,X=10S=0Therefore, class interval h = (10 – 0)/18 = 0.55We have chosen h = 0.5 for plotting the histogram. Number of classes, K= 18

Based on the above data and calculation done the following histogram was plotted

The Normal Distribution Curve

On the basis of data collected the following werecalculated.

Mean, M = 6.5593Standard deviation, S = 2.2291

Normal Distribution

User defined parameters

Mean 6.5593Standard Deviation 2.2291

x1 (lower limit) 6.5x2 (upper limit) 10

ProbabilitiesP(X <= 6.5) 0.489388324P(X >= 6.5) 0.510611676P(X <= 10) 0.938649822P(X >= 10) 0.061350178

P( 6.5 <= X <= 10) 0.449261498Normal Distribution  

     User defined parameters  

Mean 6.5593  Standard Deviation 2.2291  

x1 (lower limit) 8.5  x2 (upper limit) 10  

   Probabilities    P(X <= 8.5) 0.808019281  P(X >= 8.5) 0.191980719  P(X <= 10) 0.938649822  P(X >= 10) 0.061350178  P( 8.5 <= X <= 10) 0.130630541  

Normal distribution  

Control Charts

Xbar and R control chart:

X Bar Upper Control Limit, (XUCL), FormulaX Bar (XUCL)= X Double Bar + (A2 Factor) * R Bar XUCL = X Double Bar + A2R Bar

R Bar Upper Control Limit, (RUCL), FormulaR Bar (RUCL) = D4 * R BarRUCL = D4R Bar Therefore XUCL = 6.5593 + (0.31)* 6.17 = 8.475 (A2=0.31for sample size of 10)

RUCL = 1.78*6.17 = 11.05 (D4=1.78 for sample size 10)

Similarly lower control limit (LCL) for both the charts are calculated using the following formulae.X Bar Lower Control Limit, (XLCL), FormulaXLCL = X Double Bar - A2R BarR Bar Lower Control Limit, (RLCL), FormulaRLCL = D3R Bar Therefore, XLCL = 6.5593 – (.31)*6.17 = 4.64 RLCL= 0.22*6.17 = 1.38 (D2=0.22 for sample size 10)On the basis of the control limits defined above the following X bar and R chart were created.

Phase 3: Application of quality tools for identifying causes. Cause and Effect Diagram The 6 M’s category of causes was selected to

construct the diagram. It includes Machine, Method, Materials, Measurement, Man and Mother Nature (Environment). To construct the cause and effect diagram it was decided to develop a questionnaire which will be distributed to students who have been selected for our analysis. The questionnaire would allow the students to give their ideas and help identify causes that affect their academic performance.

Measurement Method Machine

Man Materials Environment

Abnormality in the performance of 1st

year students

Self-measurement

Time allowed on tests and assignment

Internal assessment

CGPA system

Attendance requirement

Teaching style

Handouts/notes used rather than text books

Practical application of

subjects

Curriculum

instructor not effective

Study habit

Excess Airconditioning

Absence of fans and open

windows in classroom

Poor quality laptop chargers

Slow Laptops

Personality

Student abilty and skill

Motivation

Attitude

Academic background

Composure

Instructors attitude

Hard working nature

Discipline

Communication skill

Poor sports facilities

Handouts and notes not effective

CD's, slides provided by

instructor

Reference books not used

Poor mess food

Space in indivisual study hall Teachers

pressure

Social pressure

Manipal culture

Friend circle

Poor peer relationship

Parental pressure

Cause and Effect Diagram

Pareto Diagram

• Based on the survey conducted the count for each of the causes was noted and the Pareto chart was created.

CONCLUSION

Based on our analysis and application of tools of quality on the CGPA samples of the students, we have arrived on the following conclusions:

•From the normal distribution curve and the histogram the mean M = 6.5593, was found to be very low. From the normal distribution table and Z values are under the curve i.e. the probability at different limits were calculated.

•It was observed that the probability of a student to get below 6.5 CGPA i.e. securing second class is as high as 0.489. Minimum eligibility criteria for most of the recruiting companies in MIT are that a student should have a CGPA greater than 6.5. At present 48.9% of 1st year students are ineligible.

•Probability of a student to get CGPA greater than 8.5 i.e. securing first class with distinction is equal to 0.192.

Inference from the control charts:

•All points lie within the control limits.

•Several points from sample number 9 to 16 line up consecutively on one side only of the central line (the median line). This is called a ‘run’. Seven consecutive points lie below the central line; hence the length of run is equal to seven points. Thus we conclude that there is an abnormality in the performance of 1st year students.

Based on the results obtained from the problem identifying tools we conclude that there is a significant drop in the performance of the first year students. To identify the causes cause and effect diagram and the Pareto chart were constructed. From these two tools various causes that might be affecting the performance of students were found out and it was conclude that Man i.e. Human factors are more likely to affect the performance of 1st year students here at MIT.

REFERENCES

• Kaoru Ishikawa (1989), ‘Guide to Quality Control’, Asian Productivity Organization ISBN 92-833-1035-7• Gerald M. Smith, ‘Statistical process control and quality improvement ’ISBN 0-13-095440-3• U.A Agarwaal, ‘Application of Quality tools’, National Centre for Quality Management journal, May-June 2007• Devendra Thakur, ‘ Research Methodology in Social Sciences’ ISBN 81-7100-616-X• Kultar Singh, ‘ Quantitative Social Research Methods’ ISBN 978-0-7619-3383-0• Edson Pacheco Paladini, ‘Quality tools in engineering education’, University of Santa Catarina, Brazil• Ahmad Ibrahim, ‘Current Issues in Engineering Education Quality’, DeVry Institute of Technology, Toronto• R. Radharamanan, ‘Use of Quality Tools and Standards for Continuous Improvement in Engineering Education’• www.wikipedia.com• www.sigmazone.com