Safety Science 45 (2007) 373–396

www.elsevier.com/locate/ssci

An adaptive management system for hazardoustechnology organizations

Isaac J. Obadia a,¤, Mario C.R. Vidal b, Paulo F.F.F. e Melo c

a Instituto de Engenharia Nuclear, Caixa Postal 68.550, Ilha do Fundão,CEP 21.945-970 Rio de Janeiro, RJ, Brazil

b COPPE/UFRJ, Industrial Engineering Program, CEP 21.945-970, Rio de Janeiro, RJ, Brazilc COPPE/UFRJ, Nuclear Engineering Program, CEP 21.945-970, Rio de Janeiro, RJ, Brazil

Received 27 December 2004; received in revised form 19 June 2006; accepted 6 July 2006

Abstract

This paper introduces a management system suitable for hazardous technology organizationswhich has been developed based on the assumptions that in these organizations safety is a criticalstrategic factor, the existence of an enhanced safety culture is a crucial condition for safety and thatsafety culture enhancement implies in organizational changes. The management system was theoreti-cally developed and then implemented at a Brazilian nuclear research and development installation,as a case study, in order to validate the theoretical propositions assumed in the system development.The developed management system comprises a day-to-day based organizational framework whichtreats safety as one of the organization strategic perspectives and provides a continuous adaptationof the complex causal inter-relationships which occur between the implementation of new manage-ment practices – designed and implemented according to the requirements of the criteria of excel-lence of the Brazilian quality award management assessment model – and the organization safetyculture. The results achieved in the case study permitted to demonstrate the validness of all the sys-tem theoretical propositions and to conclude that the continuous and systematic operation of themanagement system makes an eVective safety culture enhancement possible and simultaneously facil-itates that the new management practices be eVectively implemented, thus making continuous orga-nizational improvement possible.© 2006 Elsevier Ltd. All rights reserved.

* Corresponding author. Present address: Brazilian Nuclear Energy Commission, Rua General Severiano 90,Botafogo, Rio de Janeiro, CEP 22.294-900, RJ, Brazil. Tel.: +55 21 2546 2211; fax: +55 21 2546 2316.

E-mail addresses: isaac@cnen.gov.br (I.J. Obadia), mvidal@ergonomia.ufrj.br (M.C.R. Vidal), frutuoso@con.ufrj.br (P.F.F.F. e Melo).

0925-7535/$ - see front matter © 2006 Elsevier Ltd. All rights reserved.doi:10.1016/j.ssci.2006.07.002

374 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

Keywords: Management systems; Hazardous technology organizations; Safety culture; Nuclear safety

1. Introduction

The complex environment that involves organizations nowadays, where changing ofscenarios occur under a dynamic, unpredictable and accelerated fashion, increases thediversity of internal and external factors that may inXuence organizational performance,while simultaneously maximizes their inter-dependence, thus submitting organizations to aset of conditions that represent threats or opportunities regarding their survival, competi-tiveness and success. The existence of a management system within the organization inorder to identify, treat and check these conditions in a continuous and systematic way has,therefore become a strategic need for organizational performance improvement towardsmarket competitiveness. In order to cope with this context, many organizations worldwidehave implemented total quality management (TQM). However, most of them did not suc-ceed in obtaining an eVective TQM implementation, mainly due to the lack of adequacy oftheir organizational culture regarding the changes required by TQM to occur within theorganization (Sousa-Poza et al., 2001; Vermeulen, 1997).

Within the context of hazardous technology organizations – where safety is a criticalstrategic factor – such as those of the nuclear, oil and aviation sectors, there is a possibilityof accident occurrence whose consequences may reach organization’s people and assets,the environment, and the society in general. Therefore, safety management must necessar-ily be part of the strategic policy of these organizations, besides quality management. Ana-lyses of hazardous technology industrial accidents have demonstrated that safety dependsnot only on the technical factors related to the operation of the industrial processes, but onissues related to human and organizational factors as well, giving rise to the need for thedevelopment of new safety management approaches (Perrow, 1984; INSAG-1, 1986; Rea-son, 1999; Llory, 1999).

Within the nuclear sector, since the accidents of Three Mile Island (TMI), occurred in1979 in the USA, and Chernobyl, occurred in 1986 in the Ukraine, safety became a para-mount concern for the nuclear organizations worldwide. Since that time, due to the possi-ble catastrophic consequences of a nuclear reactor accident, non-governmentalorganizations of environmental protection, as well as green political parties from manycountries, started to adopt a strong stand against the use of nuclear energy to electricalpower production, giving rise to serious public acceptance problems concerning the use ofnuclear energy worldwide. This context has caused a nuclear energy use discontinuity inseveral countries, highlighting therefore, the critical strategic condition of safety regardingthe sustainability of nuclear energy use to electrical generation, and therefore to the searchof improved competitiveness of these organizations.

The analyses of the Chernobyl accident accomplished by the International AtomicEnergy Agency (IAEA) through the international nuclear safety advisory group, led tothe development of the safety culture approach, understood as a reinforcement process toall other necessary safety measures, which must be constructed, maintained and continu-ously enhanced within nuclear organizations worldwide (INSAG-1, 1986). “Safety cultureis a necessary characteristic to reach safety in nuclear installations and therefore it mustbe possible to assess its status in order to improve it and maintain it in optimal levels”

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 375

(IAEA, 1994). Since then, several activities have been coordinated by IAEA aiming at theimprovement of methods for safety culture assessment as well as for the development ofan organizational framework capable of making an eVective safety culture enhancementpossible, which still represent research challenges (Carnino, 2002). Barraclough andCarnino (1998) observe that senior management should ensure that their organizationhas a safety management system implemented on a day-to-day basis that provides astructured and systematic means of achieving and maintaining high standards of safetyperformance.

The objective of this paper is to introduce an adaptive management system suitable fornuclear as well as other hazardous technology organizations, which has been developedbased on the assumptions that in these organizations safety is a critical strategic factor fororganizational competitiveness, the existence of an enhanced safety culture is a crucial con-dition for safety and that safety culture enhancement implies in organizational changes. Inorder to validate the theoretical propositions assumed in the system development, a fouryear case study was performed, consisting on the implementation of the developed systemat a Brazilian nuclear installation that deals with research and technological developmentfor health and industrial applications where several activities involving ionizing radiationare performed, including the operation of a research reactor and two cyclotron type parti-cle accelerators.

A discussion on the safety culture concept and on organizational culture change is pre-sented, followed by the theoretical development of the management system, where itsassumptions and propositions, theoretical supporting tools and adaptive framework arepresented. The developed adaptive management system is then introduced and described.The case study activities regarding a four year implementation of the developed system arethen described and the results achieved are presented and discussed, followed by the con-clusions of the paper.

2. Safety culture

It was after the Chernobyl accident that the safety culture concept was introduced intothe nuclear Weld as a proactive safety enhancement approach, quickly becoming a require-ment by the nuclear organizations worldwide. Nevertheless, the lack of a widely accepteddeWnition of safety culture still remains, bringing diYculties to the establishment of a prac-tical organizational framework for its enhancement (Farrington-Darby et al., 2004).

In order to provide the safety culture theoretical basis used for the development of thiswork, the concepts of culture and organizational culture are Wrst presented and discussed.

2.1. Culture

Hall (as cited in Baumont, 1995) describes culture as an unspoken language to which wereact as if in response to some secret complex code that has never been written down but isunderstood by all. To Hofstede (1991), culture is deWned as the collective mind programthat distinguishes the members of one group or category of people from another. Culture isacquired and not inherited, being an eVect of the individual social environment and notfrom the genes. Studies by ethnologists and communication experts indicate that the acqui-sition of culture is manifested by the learning of patterns of behavior which become inte-grated into the unconscious of those involved (Baumont, 1995).

376 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

2.2. Organizational culture

Hofstede (1991) deWnes organizational culture as the collective mind program that dis-tinguishes the members of one organization from those of another. He comments thatalthough there is no standard deWnition to organizational culture, most authors agree thatit is: holistic, historically determined, related to anthropological studies, socially con-structed by the organization personnel, diYcult to change due to its intrinsic stability, andintangible but tangible measures are needed to change it. Schein (1989) developed adetailed organizational culture analysis model, composed by the three inter-dependent lev-els of artifacts, shared values and underlying assumptions, as illustrated in Fig. 1. He dis-tinguishes the elements of the organizational culture by treating basic assumptions as theessence – what culture really is – and by treating values and behaviors as observed manifes-tations of the cultural essence.

The most visible level of culture is composed by its artifacts that represent its con-structed physical and social environment. However, no one knows the reasons why theorganization was built in a certain way, or why people behave in one way or another.Moreover, whereas it is easy to observe artifacts, the diYcult part is to Wgure out what theartifacts mean, how they inter-relate, what deeper patterns, if any, they reXect. If one wantsto understand the cultural meaning of the artifacts one should analyze the central valuesthat provide the day-to-day operating principles by which members of the culture guidetheir behavior. A set of values that become embodied in an organizational philosophy willpredict much of the behavior that can be observed at the artifacts level. But if those valuesare not based on prior cultural learning, they may also come to be seen only as espousedvalues which predict well enough what people will say in a variety of situations but whichmay be out of line with what they will actually do in situations where those values shouldbe operating. Thus, the organization may say that safety is a value, but its decisions andattitudes in that regard may contradict what it says. Such inconsistencies mean that adeeper level of thought and perception is conducting behavior, and that the understandingof organizational culture depends on the identiWcation of what happens in this deeper level,which are the underlying assumptions. Therefore, in analyzing values, one must discrimi-nate carefully between those that are congruent with underlying assumptions and those

Fig. 1. Organizational culture levels, according to Schein (1989).

ARTIFACTS(Visible but often not

decipherable)

SHARED VALUES(Greater level of awareness)

BASIC UNDERLYINGASSUMPTIONS

(Taken for granted, invisible)

Visible products, practices, organizational structures

Strategies, objectives, philosophies

Beliefs, perceptions, unconscious thoughts and feelings accepted as true

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 377

that are, in eVect, either rationalizations or aspirations for the future. Espoused values aresometimes contradictory and incongruent with observed behavior, meaning that to get adeeper level of understanding of the culture we have to understand more fully the basicunderlying assumptions level, which actually guide behavior, and tell group members howto perceive, think, and feel about things. When a group faces a new issue, the Wrst solutionproposed can only have the level of a value because there is not yet a shared basis for deter-mining what is factual and real. If the proposed solution continuously works and the grouphas a shared perception of that success, the value gradually starts a process of cognitivetransformation into a belief and, ultimately, into an assumption. What was once a hypoth-esis, comes gradually to be treated as a reality (Schein, 1989). Cultural assumptions can bethought of as a set of Wlters that help us to focus on and perceive the relevant portions ofthe organization. They provide organizational stabilization, reducing the anxiety inherentin any new or unstable situation. These assumptions are, by deWnition, not confrontable ordebatable, meaning that if they are strongly held in a group, members would Wnd anybehavior based on any other premise inconceivable. This creates a stability to the organiza-tional culture, constituting an intrinsic diYculty to change it.

2.3. Safety culture approach

The safety culture concept was initially issued by the IAEA post-accident review meet-ing report of the Chernobyl accident which concluded that formal safety related proce-dures, properly reviewed and approved must be supplemented by the creation andmaintenance of a nuclear safety culture (INSAG-1, 1986). At that time, its meaning wasleft open to interpretation and guidance was lacking on how safety culture could beassessed and enhanced. Lately, INSAG-4 (1991) deWned safety culture as that assembly ofcharacteristics and attitudes in organizations and individuals which establishes that, as anoverriding priority, nuclear plant safety issues receive the attention warranted by their sig-niWcance. This concept was deWned in terms of the organizational framework in which peo-ple work and of the individual and group attitudes developed within this framework,considering the following components in each of the three organizational commitmentlevels (INSAG-4, 1991):

• At the policy commitment level – statement of the organization safety policy; establish-ment of the management structures; allocation of resources and performance of self-regulation.

• At the managers commitment level – deWnition of responsibilities; deWnition and controlof safety practices; qualiWcations and training; rewards and sanction; and audit, reviewand comparison.

• At the individuals commitment level – questioning attitude; rigorous and prudentapproach; and communication.

This deWnition remarks that safety culture relates both to organizations, concerningtheir style, and to individuals, concerning their personal attitudes and habits, and thatalthough these matters are generally intangible, their qualities lead to tangible manifesta-tions, and a key requirement is the development of means to use the tangible manifesta-tions to assess the associated safety culture underlying assumptions. This concept has beenpractically used by nuclear organizations worldwide as a basis for the assessment of the

378 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

safety culture eVectiveness and to identify opportunities for its enhancement. A safety cul-ture self-assessment method has then been developed by IAEA, consisting on an organiza-tion inquiry to assess personnel perceptions regarding a set of organizational factors, andon the implementation of a corresponding improvement actions plan in the organization(INSAG-4, 1991).

Fig. 2 illustrates the safety culture concept in light of the complex hazardous technologysystems defenses in depth approach, adapted from Reason’s (1999) “Swiss cheese model”(Obadia, 2004). Each layer of defense has weaknesses and gaps – which are in constantmobility according to system local conditions – creating dynamic windows of opportuni-ties for a possible accident trajectory to be completed when a rare conjunction of a set ofwindows in successive layers of defense happens. The conditions established by the layersof the defenses-in-depth and by their respective windows of opportunities are directlyrelated to the risk of an accident to occur in the system. The dotted line illustrates anincomplete trajectory, due to the eVective action of one of the layers of defense – where nowindow of opportunity was present – and the full line illustrates the safety culture contri-bution to a trajectory which would have culminated in a accident. Safety culture consti-tutes an additional defense-in-depth to the system, actuating across all system layers ofdefenses-in-depth related to a possible accident trajectory. An eVective safety culture con-tinuously contributes to the reduction of the windows of opportunities in each layer ofdefense, thus reducing the risk of accident in the system. “A good safety culture will beinherent in the thoughts and actions of individuals at all levels of an organization, creatinga high quality defense-in-depth against technical, human and organizational failures”(Barraclough and Carnino, 1998).

3. Organizational culture change

According to Seel (2000), most of the organizational culture change programs concen-trate on the visible levels of culture manifestations, whose experiences show that theirresults present limited success. EVective and long-lasting organizational culture changesrequire modiWcations on the basic underlying assumptions level (Schein, 1989).

Fig. 2. Illustration of safety culture in light of the defenses-in-depth concept (Obadia, 2004), adapted fromReason (1999).

DANGERDefenses in depth

“Windows” ofopportunities

Organizational factors Local

workplace factors

Unsafeacts

SAFETYCULTURE

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 379

Hofstede (1991) has observed that the nature of the organizational culture diVerencesreside more in the practices and less in the values, and consequently the shared perceptionsof the organizational routine practices may be considered as the center of the organiza-tional culture. “There is no doubt that founders and leaders values contribute to the orga-nizational culture, but the way culture aVects organizational staV is through the sharedpractices”, remarks Hofstede (1991), meaning therefore that the organizational culturebasic underlying assumptions may be modiWed by the introduction of new managementpractices within the organization. According to Schein (1989), when the members of theorganization get a shared perception of the good results achieved by the new practices anda solution to a problem works repeatedly, the values begin to be taken for granted, becom-ing unconsciously accepted, thus modifying the underlying assumptions.

From these considerations, one can assume that the organizational culture and theorganization management practices and their corresponding results – which comprise anorganizational management process – are deeply inter-related, as shown in Fig. 3, whereorganizational culture basic underlying assumptions are inXuenced by the managementpractices and their results, and the implementation of the management process is inXu-enced by the organizational culture. Therefore, eVective organizational culture change pro-grams must be comprised of a framework that should be capable of treating these causalinter-relationships, which are complex in nature, due to the following main reasons:

• The management process is, in fact, embedded in the organizational culture.• The organizational culture and the management process inter-relate in a dynamic and

nonlinear fashion, acting simultaneously as cause and eVect.• The organizational agents are human beings, which are complex as well.

Sousa-Poza et al. (2001) have observed in their research including 133 organizationsfrom the USA, Switzerland and South Africa, the occurrence of these complex causal inter-relationships during the implementation of TQM. They have suggested the use of an adap-tive framework where the organizational culture is assessed to identify its favorable andunfavorable aspects regarding the TQM implementation which are then used as a designbasis to the new management practices implemented through TQM. This adaptive frame-work takes advantage of the organization strengths and considers its weaknesses concern-ing TQM implementation. It assumes that the new systematic organizational practicesbased on the identiWed unfavorable cultural aspects may, through its continuous use andcorresponding results, modify attitudes and behaviors of the organization and their indi-viduals, inXuencing the organizational culture basic assumptions, and thus, consequently,

Fig. 3. Inter-relationships between organizational culture and the management process implementation.

Culture

Practices

Results

ManagementProcess

380 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

facilitating an eVective implementation of TQM. It is claimed that this framework providesa continuous and simultaneous adaptation of the complex causal inter-relationships whichoccur between organizational culture and TQM implementation (Sousa-Poza et al., 2001).

4. System development

The adaptive management system introduced in this paper was theoretically developedbased on a set of assumptions and on the corresponding theoretical propositions. The sys-tem was then implemented at a Brazilian nuclear installation that deals with research andtechnological development for health and industrial applications where several activitiesinvolving ionizing radiation are performed, including the operation of a research reactorand two cyclotron type particle accelerators. The results obtained during a four year casestudy permitted to illustrate and evaluate its performance regarding the validation of thetheoretical propositions assumed.

4.1. Theoretical basis for system development

The theoretical propositions assumed in the system development were based on themodel of excellence of the Brazilian quality award (BQA) (FPNQ, 2002), used as the basisfor the development and implementation of the new organizational management practices;on the Balance Scorecard technique (Kaplan and Norton, 1996) used to develop the per-formance measurement system (PMS), which describes the organization strategic perspec-tives, objectives and performance indicators, enabling to monitor its strategic performance(FPNQ, 2001); and on the conceived adaptive framework of the system which makes thenecessary eVective organizational changes possible.

4.1.1. The model of excellence of the Brazilian quality award (FPNQ, 2002)Several countries such as the USA, Australia, Mexico and Japan have given rise to qual-

ity awards through the establishment of results-oriented organizational managementassessment models based on a set of managerial principles, considered as the fundamentalsof excellence, which are translated into criteria and requirements, and that according to adeWned assessment method the management level of excellence of the organization is mea-sured and classiWed. These models can be seen as an evolution of TQM where quality isseen as an strategic organizational management whose implementation presumes an eVec-tive incorporation of the fundamentals of excellence into the organizational culture anddue to its organizational learning process, composed by a continuous assessment of theperformance achieved by the management practices leading to the corresponding practiceimprovements, which is a part of the management dynamics of these models.

In Brazil, the model of excellence of the Brazilian quality award (BQA) was created in1991, based on the Malcolm Baldrige National Quality Award of the USA, mainly due tothe fact that this one is based in non-prescriptive management practices, diVerently frommost of the other existing quality awards (FPNQ, 2002).

The Model of Excellence of the BQA is a 1000 points holistic management assessmentmodel, based on the following 12 fundamentals of excellence: leadership and constancy ofpurpose, vision of future, focus on client and market, ethics and social responsibility, deci-sions based on facts, personnel valorization, process-based approach, focus on the results,innovation, agility, organizational learning, and systematic approach. This fundamentals

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 381

are translated into eight inter-dependent criteria of excellence which in turn are dividedinto 27 items, and a maximum assessment score is deWned to each item, as presented inTable 1. A set of non-prescriptive speciWc assessment requirements is established to each ofthese items, which must be as closely as possible fulWlled by the organization managementpractices and results. The results-based feature of the model is practically translated to theassessment method by the assignment of a maximum score of 450 points to the criterion ofresults, equivalent to 45% of the total maximum assessment score of the model.

The main core of the model is composed by the practices of items 2.1, 2.2, 2.3 and 1.3,which reXect de typical plan-do-check-action management cycle, providing feedback tocontinuous system improvement and organizational learning. A general requirementapplied to all items refers to the learning process, which is a basic feature of the modelmanagement dynamics. The organizational management evolution towards excellence is afunction of this continuous and systematic learning, comprised by the periodic assessmentof the set of practices of each item and by the corresponding implementation of practicesimprovements or innovations within the organization management system.

Table 1Criteria of excellence of the BQA model and their corresponding items and maximum assessment score (FPNQ,2002)

Criterion/maximum score Item/maximum score

1 Leadership/100 1.1 Leadership system/301.2 Culture of excellence/401.3 Global performance critical analysis/30

2 Estrategies and plans/90 2.1 Strategies formulation/302.2 Strategies deployment/302.3 Performance measurement system/30

3 Clients/60 3.1 Image and market knowledge/303.2 Relationship with clients/30

4 Society/60 4.1 Social and environmental responsibility/304.2 Ethics and social development/30

5 Information and knowledge/60 5.1 Organizational information management/205.2 Comparative information management/205.3 Intellectual capital management/20

6 Personnel/90 6.1 Working systems/306.2 Training and development/306.3 Quality of life/30

7 Processes/90 7.1 Products related processes management/307.2 Supporting processes management/207.3 Suppliers related processes management/207.4 Economics and Wnancing management/20

8 Results/450 8.1 Clients and market results/1008.2 Economics and Wnancing results/1008.3 Personnel related results/608.4 Suppliers related results/308.5 Products related processes results/808.6 Society related results/308.7 Organizational and supporting processes results/50

382 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

The level of excellence achieved by the organization is measured by an oYcial assess-ment team using a two steps method consisting Wrstly of the assessment of the managementreport of the organization which describes all its management practices related to each cri-terion of excellence and presents the corresponding results achieved, and secondly of a ver-iWcation of the information presented in the management report considered doubtful bythe assessment team, through a visit to the organization. The score assigned to each crite-rion of excellence depends on the degree of attendance to their corresponding require-ments, which is a function of the organization working practices and their correspondingresults.

At the end of the assessment process, the organization receives the assessment reportissued by the assessment team which contains all information related to the level of excel-lence achieved by the organization management system such as comments on strengthsand weaknesses and the assessment score of each item and criterion, as well as of the man-agement system as a whole. This report constitutes an extremely rich source of informationto the improvement of the organization management system.

4.1.2. The balanced scorecard technique (Kaplan and Norton, 1996)The balanced scorecard (BSC) is a technique used to establish a structure capable of

translating the organizational strategy into operational terms. The BSC translates theorganizational mission and strategy into a comprehensive set of performance measure-ments which is a basis for a strategic performance measurement system (Kaplan andNorton, 1996).

The BSC technique complements the Wnancial perspective measurements, which areclassical and prevailing in the decision-making process in many organizations, by consider-ing measurements of three other perspectives: clients, internal processes, and learning andgrowing. It assumes that the measurement of the strategic organizational performancethrough the Wnancial perspective only is not suYcient. The organization must search for anequilibrium condition among the four proposed perspectives, which represent the mini-mum necessary dimensions so that an strategy can be deWned, although other perspectivesmay be added, whenever needed. The BSC also permits to check and adjust the cause–eVectrelationships hypothesized among the performance indicators.

The perspectives to be utilized in the performance measurement system depends funda-mentally on the organization’s proWle. A well succeeded BSC is the one which communi-cates the organizational strategy through an integrated set of Wnancial and non-Wnancialmeasurements, whose importance to the organizational performance improvement can behighlighted by the following features:

• The BSC describes the strategic vision to all organizational levels, creating sharing aspi-rations within the organization.

• The BSC creates a model to represent the organizational strategy, illustrating to the taskforce how they can contribute to organizational success. Without this, people anddepartments may even improve their local performance, but certainly will not be able tofocus their contribution on the organizational strategic objectives.

• The BSC brings focus to the organization changing eVorts.

Therefore, the BSC is a suitable technique to develop the performance measurementsystem as required by item 2.3 of the model of excellence of the BQA, which comprises a

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 383

representation of the organization strategy, indicated by the perspectives, their correspond-ing objectives, performance indicators and planned goals.

4.1.3. The adaptive framework of the management systemThe adaptive management system developed assumes that safety culture corresponds to

that organizational culture which has safety as one of its values and whose enhancement isa function of the degree of how safety is eVectively perceived, shared and prevailing withinthe organization. According to Schein’s (1989) culture levels theory presented in Fig. 1, thismeans that safety culture will be as more eVective in an organization as more safetybecomes a value eVectively supported by its organizational culture underlying assump-tions. So, one can assume that safety culture enhancement implies in organizationalchanges in order to modify some organizational culture underlying assumptions which, inturn, according to the organizational culture change context of Fig. 3, requires the imple-mentation of new management practices in the organization. The adaptive framework ofthe developed management system, presented in Fig. 4, was then conceived through themodiWcation of the Sousa-Poza et al. (2001) adaptive approach, considering hazard tech-nology organizations context, being constituted by a safety culture self-assessment as pro-posed by IAEA (INSAG-4, 1991) and a management process based on the criteria ofexcellence of the BQA model. Due to the holistic management approach of the BQAmodel, it is possible to associate its criteria of excellence to the safety culture organiza-tional factors, thus allowing the development of new management practices aiming at theirimprovement, which, in turn, may aVect organization’s and individuals’ attitudes andbehaviors, and consequently modifying the safety culture underlying assumptions.

It is worth observing that within the scope of hazardous technology organizations,safety culture enhancement plays indeed a twofold crucial role regarding the enhancementof operational safety and the achievement of an eVective implementation of the new orga-nizational management practices as well.

4.2. System assumptions and theoretical propositions

Considering the strategic importance of safety to hazardous technology organizationsand the theoretical basis used for the management system development, the systemassumptions were deWned and the corresponding theoretical propositions established, aspresented in Table 2.

Fig. 4. Developed adaptive framework of the management system.

Safety Culture

The management practices affect attitudes and behaviors, and consequently, the safety culture underlying assumptions.

Safety culture affects which management practices will be emphasized.

BQAManagement

Model

384 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

4.3. The adaptive management system

The developed adaptive management system to hazardous technology organizations,presented schematically in Fig. 5, consists of an organizational management process thatfollows the requirements and the management dynamics of the model of excellence of theBQA, to which a safety culture self-assessment as deWned by IAEA (INSAG-4, 1991) isintegrated through the management improvement plan (MIP) – where the practicesimprovements are established – and a perspective of safety is prescribed to be part of the

Table 2Assumptions, theoretical considerations and propositions assumed in the development of the adaptive manage-ment system

System assumption Theoretical considerations Propositions assumed

Safety is a critical factor forhazardous technologyorganizations

Safety must be treated as anorganizational strategicdimension

The requirements of the criteria ofexcellence of the BQA assessmentmodel can be used to design theorganizational management systemwhere the adoption of a perspectiveof safety in the organizationalperformance measurement systemprovides an strategic treatment forsafety

An enhanced safety cultureis crucial for safety

Safety culture can be eVectivelyenhanced by the implementationof new management practices whoseresults are perceived as strengths bythe organization. The managementprocess must be able to provideimprovements to any of the IAEAsafety culture organizational factors

Due to the holistic managementapproach of the criteria of excellenceof the BQA model, these criteria canalways be associated to any IAEAsafety culture organizational factor,thus allowing for the development ofnew management practices aiming attheir improvement

Fig. 5. Schematic diagram of the developed adaptive management system.

Management Process (BQA)

Practices ResultsPerformance Measurement

System

CriticalAnalyses

Safety

Culture

Self-assessment

Criteria ofExcellence

MIP

Learning process

Organization Management

Report

Management external

evaluation

Strengths and opportunities for

improvement

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 385

organization strategic performance measurement system (PMS) which has the safety cul-ture self-assessment global mean index as one of its performance indicators. A detaileddescription of the system is presented below:

The management practices of the organization are designed and implemented based onthe requirements related to the criteria of excellence of the BQA model, in order to contin-uous and systematically improve organizational performance within the scope of the orga-nization proWle and core business.

A safety culture self-assessment, consisting on a questionnaire inquiry submitted to theorganization staV regarding a set of organizational factors associated to the IAEA safetyculture concept (INSAG-15, 2002) is performed so that those organizational factorsdeserving improvement are identiWed and taken into consideration in the learning processthrough the MIP. Due to the holistic management characteristic of the BQA model it isalways possible to associate any of the IAEA safety culture organizational factors to thecriteria of excellence, thus permitting that new management practices aiming at theimprovement of the organizational factors be designed within the scope of the correspond-ing criteria. The improvement of the safety culture organizational factors has a twofoldobjective of enhancing safety culture and facilitating an eVective implementation of theorganizational management practices.

The strategic objectives of the organization, their corresponding performance indicatorsand planned goals are established in the performance measurement system (PMS). In orderto assume the strategic condition that safety represents for hazardous technology organi-zations, and to create or enhance organizational commitment to safety issues, a perspectiveof safety is adopted in the PMS, which is developed by the BSC technique (Kaplan andNorton, 1996), thus providing systematic monitoring of the organizational safety perfor-mance indicators. The safety culture global mean index, obtained through the safety cul-ture self-assessment, is incorporated as one of the safety performance indicators, thusfostering a systematic organizational commitment to safety culture enhancement, reXect-ing its crucial importance to the organization safety improvement.

Periodically (usually annually, although several organizations are using shorter periods)the organizational performance is assessed through the critical analysis practice, whichanalyzes all kinds of results achieved by the organization as required by criterion 8 of theBQA model (see Table 1), and identiWes organizational management improvements. ACorrective Actions Plan is then established to be implemented in the organization aimingat improving its results.

The system is complemented by its learning process, implemented through the manage-ment improvement plan (MIP), which is deWned taking into consideration the results of thesafety culture self-assessment, the corrective actions plan obtained through the criticalanalysis practice, and the results of the external assessment of the organizational manage-ment level of excellence, performed according to the model of excellence of the BQAassessment method. The MIP establishes the new practices to be implemented in the orga-nization, related to each of the criterion of excellence of the BQA model, in order to conti-nuously improve the organizational performance.

It is presumed that the continuous operation of this management system provides in theorganization a simultaneous adaptation of the complex causal inter-relationships thatoccur between the safety culture and the management process implementation. In eachoperating cycle of the system – constituted by the phases of practices implementation,results achievement, critical analysis, management assessment and learning – the safety

386 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

culture underlying assumptions can be modiWed, making an eVective safety cultureenhancement possible, which, in turn, facilitates an eVective implementation of the organi-zational management system, thus providing the conditions for continuous organizationalperformance improvement.

5. The case study

The case study consisted on the implementation of the developed management systemat the Nuclear Engineering Institute (IEN), as part of its organizational change program inthe search for excellence. The results achieved during the period from year 2000 to 2003were analyzed and permitted to illustrate the system performance and validate the theoret-ical propositions assumed in its development.

All case study activities were coordinated by the quality management advisory body ofthe IEN, under the orientation of the Wrst author of this paper, in strict accordance to thedeveloped adaptive management system framework, in a continuous and systematic way.The method applied to perform the IEN safety culture self-assessments, the perspective ofsafety of the IEN performance measurement system, and the method used for assessing theIEN management system level of excellence as a measure of the degree of eVectivenessachieved by the system implementation, are described.

5.1. IEN proWle

The IEN is a Brazilian nuclear non-power research and development institution whichbelongs to the Brazilian Nuclear Energy Commission (CNEN). It performs scientiWc andtechnological research and development activities in accordance to its institutional mis-sion to contribute to the welfare of society by the development of the nuclear and corre-lated sectors and the safe use of nuclear energy through the provision of technology,products, services and training. The IEN proWle, management practices and organiza-tional results are presented in its management report (IEN, 2003), while the research anddevelopment activities are described in its progress report (IEN, 2002). The relation of theIEN activities which involve hazardous conditions during their operation are presented inTable 3.

5.2. IEN safety culture self-assessment

The IEN safety culture self-assessments were performed in 2001 and in 2003 by thesame method previously utilized by Eletronuclear, the operator company of the Braziliannuclear power plants Angra I and Angra II, which was based on the IAEA approach(INSAG-4, 1991) and was developed through IAEA consultancy (Eletronuclear, 1999).The self-assessment instrument was a questionnaire composed of six personal questions,which deWned the classes and subclasses of analysis, and of 70 statements related to the 22IAEA safety culture organizational factors presented in Table 4. For each statement, WvediVerent options related to the respondent level of perception were presented, establishedaccording to a summing Likert scale (Mattar, 2000), as shown in Table 5. The total punctu-ation of each statement corresponds to the summing of all values related to the selectedrespondents perception levels.

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 387

Table 6 presents the association of the assessed 22 organizational factors to the criteria ofexcellence of the BQA model, to provide the design basis for the new management practicesimplemented within the system in order to improve the safety culture organizational factors.

Table 3IEN activities involving hazardous operating conditions

Activity Hazardous condition

R&D of nuclear techniques using researchreactor

• Operation of the nuclear research reactor• Exposition to ionizing radiation

Irradiation services and samples analyses • Operation of the nuclear research reactor• Exposition to ionizing radiation

Radioisotopes and radiopharmaceuticalsproduction

• Operation of the cyclotron particleaccelerators CV-28 and RDS 111

• Handling of radioactive material• Exposition to ionizing radiation

Application of nuclear techniques toindustry

• Handling of radioactive material• Exposition to ionizing radiation

Low level radioactive waste management • Handling of radioactive material• Exposition to ionizing radiation

Radiation protection • Handling of radioactive material• Exposition to ionizing radiation

Table 4Organizational factors assessed and related number of statements in the assessment questionnaire

Number Organizational factor Number of statements

OF1 High-level management commitment to safety 2OF2 Evident leadership 3OF3 High priority to safety 3OF4 Systematic approach to safety 4OF5 Importance of safety on the organization’s strategic plan 2OF6 Lack of conXicts between production and safety 2OF7 Relationship with licensing and regulatory body 1OF8 Proactive and long-term perspective 5OF9 Management of changes 2OF10 Quality of documents and procedures 4OF11 Compliance with regulations and procedures 3OF12 QualiWed and well dimensioned personnel 3OF13 Well deWned tasks and responsibilities 4OF14 Transparency and communication 5OF15 Motivation and satisfaction at work 5OF16 Good working conditions related to execution time,

workload and stress3

OF17 Adequate resources allocation 2OF18 Collaboration and teamwork 4OF19 Error treatment at work 2OF20 Management of conXicts 2OF21 Organizational evolution through learning 5OF22 Commitment to performance and rewards 4

388 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

All safety culture self-assessments results – institutional, subclasses and organizationalfactors – have been analyzed by their corresponding mean index, calculated by the follow-ing expression and classiWed according to Table 7, which is the same classiWcation deWnedby IAEA for the eletronuclear safety culture self-assessment (Eletronuclear, 1999).

• Mean index (%)D100£ total obtained punctuation/maximum possible punctuation.

In the case of the safety culture institutional mean index, it was calculated consideringthe punctuation of all statements of all respondents. In the cases of each subclass meanindex, they were calculated considering the punctuation of all statements of the corre-sponding subclass of respondents, and the 22 organizational factors mean indexes werecalculated considering the punctuation of all respondents for the set of statements relatedto each one of these organizational factors.

5.3. IEN performance measurement system

The IEN performance measurement system (PMS), developed by the BSC technique(Kaplan and Norton, 1996) was composed by 29 performance indicators distributed along

Table 5Respondents perception levels and their corresponding values

Option Value

Number Perception level

1 Agree 42 Disagree 13 Partially agree 34 Partially disagree 25 Do not know 0

Table 6Main associations of the 22 assessed organizational factors with the criteria of excellence of the BQA model

Organizational factor assessed Criteria of excellence of the BQA model

OF1, OF2, OF4, OF6, OF9, OF14, OF20, OF21 1 – LeadershipOF4, OF5, OF8, OF17, OF21 2 – Strategies and plansOF7, OF11, OF21 4 – SocietyOF7, OF14, OF21 5 – Information and knowledgeOF3, OF4, OF6, OF8, OF9, OF11, OF12, OF13, OF14,

OF15, OF16, OF18, OF19, OF20, OF21, OF226 – Personnel

OF4, OF6, OF10, OF11, OF17, OF19, OF21 7 – Processes

Table 7ClassiWcation of the IEN safety culture mean indexes

Mean index (%) ClassiWcation

Between [85 and 100] OptimalBetween [75 and 85) GoodBetween [65 and 75) SatisfactoryBetween [50 and 65) RegularBetween [0 and 50) Unsatisfactory

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 389

the six following perspectives of results: Financial/Clients and Market/Knowledge/Safety/Processes/Personnel. The perspective of safety is highlighted in Table 8. The complete PMSof the IEN is described in its management report (IEN, 2003).

5.4. Assessment of the IEN management system level of excellence

This assessment was performed through the participation of the IEN in the Excellencein Technological Research Project, coordinated by the Brazilian Association of Techno-logical Research Institutions (ABIPTI), where most of the participating institutions areIEN peers of the Brazilian science and technology sector (Albuquerque and Giesbrecht,2000). This project has the general objective of stimulating the search for the continuousperformance improvement in the participating organizations, through the implementa-tion of an organizational management process based on the criteria of excellence of theBQA model, and more speciWcally through the performance of the management level ofexcellence assessment, based on the method of the BQA model. This assessment was per-formed uninterruptedly from 2000 to 2003 by an external assessment team, established byABIPTI, and is comprised by an analysis of the IEN management report (IEN, 2003) –issued annually according to a standard format deWned by the BQA model – followed bya visit of the ABIPTI assessment team to IEN in order to check the actual implementa-tion of some management practices as presented in its management report, as well as toverify some information considered doubtful by the assessment team. Considering theIEN organizational proWle, institutional mission and core business, as well as the BQAmodel assessment requirements established to each item of the criteria of excellence, allimplemented management practices were assessed based on their degree of: adequacy,proactive approach, reWnement, innovation, dissemination, and continuity; and the orga-nizational results achieved were assessed based on the degree of relevance, level of perfor-mance compared to benchmarks, and tendency, as prescribed by the BQA modelassessment method (Obadia, 2004). This assessment has provided the following informa-tion to IEN through the ABIPTI’s Final Assessment Reports regarding the level of excel-lence of its management system (Lima and Araújo, 2000; Marques et al., 2002; Silva et al.,2002; Soares et al., 2003):

• The score related to each criterion of excellence.• The global score related to the level of excellence of the IEN management system.• Comments and remarks on strengths and opportunities for improvement related to the

practices of each criterion of excellence and to the system as a whole.

Table 8Perspective of safety of the IEN performance measurement system

Perspective Strategic objective Outcome indicator Critical factors for success Driver indicator

Safety To cultivate an imageof safe institution

• Accidents causingpersonal medicallicenses

• Accidents withionizing radiation

Proactive and preventiveattitudes related to safetyissuesFulWllment of legal safetyrelated obligations

• Safety cultureinstitutional meanindex

• Accidents notcausing personalmedical licenses

390 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

• General conclusions about the management system, in light of the criteria requirementsand the fundamentals of the BQA model of excellence.

• ClassiWcation of the level of excellence achieved by the IEN management systemaccording to the BQA model classiWcation table (FPNQ, 2002).

6. Results and discussions

6.1. Results of the IEN safety culture self-assessments

The comparative results of the IEN safety culture self-assessments of 2001 and 2003are presented. The assessment participation rate was of 73% of IEN personnel in 2001and of 52% in 2003. The evolution of the safety culture institutional mean index as wellas of each speciWc assessed subclass mean indexes are presented in Fig. 6, which showsthat:

• The safety culture institutional mean index (number 1 in Fig. 6) increased its classiWca-tion from regular to satisfactory.

• In all the other six assessed classes – education, sex, working time, age, salary and exter-nal experience, corresponding respectively to numbers 2, 3, 4, 5, 6 and 7 in Fig. 6 – atleast one of their corresponding subclasses had the safety culture mean index increasedfrom regular to satisfactory. Although all subclasses showed a percent increase, eight ofthem maintained their previous regular classiWcation while nine evolved from regular tosatisfactory (2.2, 2.3, 3.1, 4.2, 4.3, 5.2, 6.3, 7.1 and 7.2). It is worth remarking that some ofthese nine subclasses, such as 2.2 and 2.3 which relate to the employees with graduateand post-graduate education, 4.2 and 4.3 which relate to the employees with more than10 years working time, and 6.3 which relates to the employees with higher salary, consti-tute IEN personnel representative segments which highly inXuence organizationalbehavior. Therefore one can consider that these results indicate that signiWcant organi-zational changes towards safety culture enhancement were achieved.

Fig. 6. Evolution of the safety culture self-assessments mean indexes.

50%

55%

60%

65%

70%

1 2.1 2.3 3.1 3.2 4.1 4.2 4.3 5.1 5.2 5.3 6.1 6.2 6.3 7.1 7.2 7.3

2001

2003

1-INSTITUTIONAL2-EDUCATION 2.1 High School 2.2 Graduation 2.3 Post graduation3-SEX 3.1 Male 3.2 Female4-WORKING TIME (years) 4.1 Until 10 4.2 Between 10 and 20 4.3 Over 205-AGE (years old) 5.1 Until 40 5.2 Between 41 and 50 5.3 Over 506-SALARY (U$) 6.1Until 700 6.2 From 701 to 1,300 6.3 Over 1,3007-EXTERNAL EXPERIENCE 7.1 until2 years 7.2 From 2 to 4 years 7.3 Over 4 years

Mea

n In

dex

Classes/subclasses

2.2

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 391

The results of the safety culture self-assessments related to the 22 assessed organiza-tional factors are presented in Fig. 7. The evolution of their mean index indicate that four-teen organizational factors (OF 2, 3, 6, 7, 8, 9, 10, 15, 16, 17, 18, 19, 20 and 21) maintainedtheir previous classiWcation – although they showed some percent improvement – while theother eight evolved to the immediately superior classiWcation. This represented a decreasefrom thirteen to nine in the number of organizational factors classiWed as unsatisfactory orregular, and an increase from nine to thirteen in the number of organizational factors clas-siWed as satisfactory or good.

In order to illustrate the operation of the adaptive system framework, the actions asso-ciated to the evolution achieved in organizational factor 1 (OF1) – high-level commitmentto safety – are described. In the 2001 safety culture assessment, OF1 was classiWed as regu-lar. It was then associated to the leadership criterion of excellence of the BQA model, giv-ing rise to the creation of the IEN safety culture committee, composed by onerepresentative of each sector of the Institute, including the administrative ones, under thecoordination of the head of the Safety and Radiation Protection Division. The creation ofthis committee emerged during the establishment of the 2001 management improvementplan, with the objective of performing open institutional meetings to discuss, in a transpar-ent way, all sort of issues concerning the safe operation of the IEN’s hazardous activities,based on the results of the safety culture self-assessments, and mainly those related to theorganizational factors classiWed as unsatisfactory or regular. The safety culture committeemeetings were held with the participation of the managers and the operational teaminvolved in the discussed issues, thus providing a transparent communication processabout safety issues within the Institute, where human, technical and organizational factorswere considered under managerial and operational views. These discussions have provedto constitute an eVective management practice to identify the IEN safety culture underly-ing assumptions, bringing therefore, valuable contributions to the improvement of theorganizational factors – OF1 was improved to satisfactory in the 2003 self-assessment –and hence to the safety culture enhancement.

6.2. Evolution of the IEN safety performance indicators

The performance indicators of the perspective of safety of the IEN performance mea-surement system and their corresponding calculation method are presented in Table 9.

Fig. 7. Evolution of the 22 assessed organizational factors mean indexes.

30%

40%

50%

60%

70%

80%

0 2 4 6 8 10 12 14 16 18 20 22Organizational Factors

2001

2003

Unsatisfactory

Regular

Satisfactory

Good

Mea

n In

dex

392 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

The following results were achieved in the case study for these safety related indicators:

• No accident with ionizing radiation occurred.• Safety culture institutional mean index increased its classiWcation from regular to satis-

factory, as previously presented in Fig. 6.• The number of accidents causing personal medical licenses per million working hours

had a very signiWcant 77% decrease from 1999 to 2003, as presented in Fig. 8.• The number of accidents not causing personal medical licenses per million working

hours had a signiWcant 45% decrease from 2000 to 2003, as presented in Fig. 9.

Table 9Indicators of the IEN perspective of safety

a These performance indicators are recommended by the Brazilian Association of Technical Standards to mea-sure personal health related accidents (ABNT, 2001).

Indicator Calculation method Unit

Accidents causing personal medicallicensesa

106 £ Number of accidents/totalstaV annual working hours

Accident per millionworking hours

Accidents with ionizing radiation Number of accidents UnitsSafety culture Self-assessment global mean index %Accidents not causing personal

medical licensesa106 £ Number of accidents/totalstaV annual working hours

Accident per millionworking hours

Fig. 8. Number of accidents causing personal medical licenses.

Acc

iden

ts p

er m

illio

n w

orki

ng h

ours

0

5

10

15

20

1999 2000 2001 2002 2003

Good

Year

Fig. 9. Number of accidents not causing personal medical licenses.

01020304050

2000 2001 2002 2003

Good

Acc

iden

ts p

er m

illio

nw

orki

ng h

ours

Year

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 393

6.3. Results of the IEN management system level of excellence assessments

The level of excellence achieved by the management system is a measure of the degree ofattendance of the requirements of the BQA model criteria of excellence by the imple-mented management practices as well as by the organizational results achieved. Theseassessments were used to evaluate the degree of eVectiveness achieved in the managementsystem implementation.

The annual mean assessments of the ABIPTI project participating institutions manage-ment systems were considered an adequate benchmark to evaluate the improvements ofthe IEN management system, since it reXected the level of excellence achieved by IENpeers institutions from the science and technology sector, including the other Braziliannuclear research and development organizations. The use of such a benchmark was meantnot only to attend to assessment requirements, but also to establish an external reference tothe IEN personnel so that an objective perception of the evolution achieved by the IENmanagement system could be obtained.

Fig. 10 presents the criterion eight – organizational results – assessments from 2000 to2003, compared to the ABIPTI’s project mean assessments, where the seven diVerent typesof results, as presented in Table 1, were considered. One can observe that within theassessed period a continuous improvement tendency was achieved in the IEN performanceresults, with the assessment score increasing from 12% (of the 450 maximum score of thiscriterion) in the year 2000, to 37% in the year 2003, which means that a very signiWcantimprovement of more than 200% in the IEN performance results was achieved.

Fig. 11 presents the evolution achieved in the assessments of the IEN management sys-tem as a whole, from 2000 to 2003, compared to the ABIPTI’s project mean assessments.These results reXect the total score of the management assessment achieved each year inthe eight criteria of excellence of the BQA model.

The results of Fig. 11 indicate the positive tendency achieved by the level of excellence ofthe IEN management system. It is worth remarking that from 2000 to 2003, the manage-ment system assessment score increased from 158 points (after one year of implementation)to 346 points (after four years of continuous and systematic implementation), meaning

Fig. 10. Evolution of the IEN organizational results assessments.

0

10

20

30

40

2000 2001 2002 2003

%IEN

ABIPTI

Year

394 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

a 119% net increase in the level of excellence of the developed adaptive managementsystem.

The results achieved during the four year case study showed that the safety culture wasenhanced, the safety performance indicators were gradually improved, and the organiza-tional performance results assessments as well as the management system assessmentsgradually signiWcantly improved, demonstrating that an eVective implementation of themanagement system was achieved. Therefore, one can consider that the theoretical propo-sitions assumed in the system development were validated.

7. Conclusions

This paper introduced an adaptive management system suitable for hazardous technol-ogy organizations which has been developed based on the assumptions that in these orga-nizations safety is a critical strategic factor, the existence of an enhanced safety culture is acrucial condition for safety and that safety culture enhancement implies in organizationalchanges. These assumptions gave rise to the following basic design features of the devel-oped management system:

• The requirements of the criteria of excellence of the BQA model were used as designbasis to develop and implement the system management practices, although they havebeen originally conceived as management assessment requirements;

• The developed system makes safety culture enhancement possible on an integrated basisthrough the association of the BQA criteria of excellence to the safety culture organiza-tional factors assessed in the safety culture self-assessments. This integration of thesafety culture to the BQA management process comprises an adaptive framework totreat the complex causal inter-relationships that occur between the implementation ofnew management practices and the organization safety culture;

• A perspective of safety is incorporated to the performance measurement system, whichestablishes the organization safety related objectives and the corresponding safety per-formance indicators – the safety culture institutional mean index is one of this indicators –thus assuring an strategic treatment to the safety related issues, including safety culture.

Fig. 11. Evolution of the level of excellence of the IEN management system assessments.

0

50

100

150

200

250

300

350

400

2000 2001 2002 2003

Poi

nts

IEN

ABIPTI

Year

I.J. Obadia et al. / Safety Science 45 (2007) 373–396 395

The developed management system was continuous and systematically implemented,from 2000 to 2003, at the Nuclear Engineering Institute, a Brazilian nuclear research anddevelopment installation, as a case study, whose results permitted to validate the theoreti-cal propositions assumed in the system development and to conclude that the developedmanagement system comprises a day-to-day based integrated adaptive framework whosecontinuous and systematic operation provides the following contributions to the continu-ous performance improvement of hazardous technology organizations:

• The management system considers safety as a strategic dimension of the organizationperformance, thus strengthening the organizational commitment to safety by the estab-lishment of systematic practices for its continuous management and improvement.

• The management system makes an eVective safety culture enhancement possible, thuscontributing in a proactive way to safety improvement and therefore to reduce theoccurrence of human or organizational failures that might contribute to possible acci-dents.

• The management system facilitates an eVective implementation of the organizationalmanagement practices designed according to the requirements of the criteria of excel-lence of the BQA model, which constitutes a method presently considered as a nationaland international reference to foster world-class organizations.

• The management system provides a localized and speciWc treatment of the complexcausal inter-relationships that occur between the implementation of the managementpractices based on the requirements of the criteria of excellence of the BQA and theorganization safety culture. Therefore, it can be applied to any organization, regardlessits industrial sector and geographic location.

Although this work has been mainly developed based on the nuclear sector context, it isexpected that it shall encourage the use of the adaptive management system by hazardoustechnology organizations from other industrial sectors.

The generalization of the presented conclusions must take into consideration the limita-tions imposed by the speciWc conditions under which the case study was performed.

References

ABNT, 2001. Working accidents – procedure and classiWcation (in Portuguese). Standard NBR-14280, BrazilianAssociation of Technical Standards, Rio de Janeiro.

Albuquerque, L.C., Giesbrecht, H.O., 2000. Performance assessment of technological research institutions: theexperience of the Excellence in Technological Research Project (in Portuguese). In: proceedings of the ABI-PTI congress, pp. 141–148, Fortaleza.

Barraclough, I., Carnino, A., 1998. Safety culture. Keys for sustaining progress. IAEA Bulletin 40, 27–30. Vienna.Baumont, G., 1995. Culture and safety culture: individual and collective attitudes. In: Experience with Strengthen-

ing Safety Culture in Nuclear Power Plants, IAEA-TECDOC-821, Vienna, pp. 103–112.Carnino, A., 2002. Historical development of safety culture. In: Proceedings of the IAEA International Confer-

ence on Safety Culture in Nuclear Installations, Rio de Janeiro, pp. 1–8.Eletronuclear, 1999. Safety Culture Self Assessment. In: O Átomo, vol. 15, Rio de Janeiro (in Portuguese).Farrington-Darby, T., Pickup, L., Wilson, J.R., 2004. Safety culture in railway maintenance. Safety Science 43, 39–60.FPNQ, 2001. Global performance measurement system planning . Thematic Committee Report (in Portuguese).

Fundação Prêmio Nacional da Qualidade, São Paulo.FPNQ, 2002. Criteria of Excellence 2003. The state of the art in the management for the excellence in performance

and for competitiveness improvement. Fundação Prêmio Nacional da Qualidade, São Paulo (in Portuguese).

396 I.J. Obadia et al. / Safety Science 45 (2007) 373–396

Hofstede, G., 1991. Cultures and Organizations: Software of the Mind. Mc Graw-Hill, London.IAEA, 1994. ASCOT Guidelines. TECDOC-743, International Atomic Energy Agency, Vienna.IEN, 2002. Progress Report. Instituto de Engenharia Nuclear, Rio de Janeiro.IEN, 2003. Management Report. Instituto de Engenharia Nuclear, Rio de Janeiro (in Portuguese).INSAG-1, 1986. Summary Report on the Post-Accident Review Meeting on the Chernobyl Accident. Safety

Series No. 75, International Nuclear Safety Advisory Group, IAEA, Vienna.INSAG-4, 1991. Safety culture. Safety Series No. 75, International Nuclear Safety Advisory Group, IAEA,

Vienna.INSAG-15, 2002. Key Practical Issues in Strengthening Safety Culture, International Nuclear Safety Advisory

Group, IAEA, Vienna.Kaplan, R.S., Norton, D.P., 1996. The Balanced Scorecard. Harvard Business School Press, Boston.Lima, A.A., Araújo, G.N.F., 2000. Final Assessment Report of the Nuclear Engineering Institute. ABIPTI, Brasí-

lia (in Portuguese).Llory, M., 1999. Acidentes Industriais. O custo de silêncio. Ed. MultiMais, Rio de Janeiro.Marques, R.A., Inácio, J.J., Filho, M.N.N., 2002. Final Assessment Report of the Nuclear Engineering Institute.

ABIPTI, Brasília (in Portuguese).Mattar, F.N., 2000. Marketing Research. Atlas, São Paulo (in Portuguese).Obadia, I.J., 2004. Adaptive management system for hazardous technology organizations: safety culture as

an assumption of nuclear excellence. D.Sc. Thesis, COPPE/Universidade Federal do Rio de Janeiro, Rio deJaneiro (in Portuguese).

Perrow, C., 1984. Normal Accidents: Living with High Risks Technologies. Basic Books, New York.Reason, J., 1999. Managing the Risks of Organizational Accidents, third ed. Ashgate Publishing Limited,

England.Schein, E.H., 1989. Organizational Culture and Leadership, eighth ed. Jossey-Bass Inc. Publishers, California.Seel, R., 2000. Culture and complexity: new insights on organizational change. Organizations & People 7 (2), 2–9.Silva, J.E., Campos, C.O.N., Rosek, R., Chagas, F.A., 2002. Final Assessment Report of the Nuclear Engineering

Institute. ABIPTI, Brasília (in Portuguese).Soares, I., Stivi, S.A., Schlemper, A.P., 2003. Final Assessment Report of the Nuclear Engineering Institute. ABI-

PTI, Brasília (in Portuguese).Sousa-Poza, A., Nystrom, H., Wiebe, H., 2001. A cross-cultural study of the diVering eVects of corporate culture

on TQM in three countries. International Journal of Quality & Reliability Management 18 (7), 744–761.Vermeulen, W., 1997. Cultural change: crucial for the implementation of TQM. Training for Quality 5 (1), 40–45.