Reiman 2011: Understanding maintenance work in safety-critical organisations

8/6/2019 Reiman 2011: Understanding maintenance work in safety-critical organisations

http://slidepdf.com/reader/full/reiman-2011-understanding-maintenance-work-in-safety-critical-organisations 1/29

This article was downloaded by: [VTT], [Teemu Reiman]On: 29 June 2011, At: 10:24Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Theoretical Issues in Ergonomics

SciencePublication details, including instructions for authors and

subscription information:

http://www.tandfonline.com/loi/ttie20

Understanding maintenance work in

safety-critical organisations – managing

the performance variabilityTeemu Reiman a

aVTT, PO Box 1000, FIN-02044 VTT, Espoo, Finland

Available online: 29 Jun 2011

To cite this article: Teemu Reiman (2011): Understanding maintenance work in safety-critical

organisations – managing the performance variability, Theoretical Issues in Ergonomics Science,

12:4, 339-366

To link to this article: http://dx.doi.org/10.1080/14639221003725449

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching and private study purposes. Any

substantial or systematic reproduction, re-distribution, re-selling, loan, sub-licensing,systematic supply or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.

http://dx.doi.org/10.1080/14639221003725449


http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/page/terms-and-conditions

http://dx.doi.org/10.1080/14639221003725449




Theoretical Issues in Ergonomics Science

Vol. 12, No. 4, July–August 2011, 339–366

Understanding maintenance work in safety-critical

organisations – managing the performance variability

Teemu Reiman*

VTT, PO Box 1000, FIN-02044 VTT, Espoo, Finland

(Received 7 October 2009; final version received 18 February 2010)

Human and organisational performance variability has been identified as a causeof many latent and active failures in maintenance. Seldom has the variabilitybeen considered as an integral and inherent part of the maintenance activity tobe managed by organisational means. The article deals with the challengeof understanding maintenance work in safety-critical organisations. The aim is toreview the current literature on maintenance work and illustrate the organisa-tional research challenges of managing performance variability in maintenance.This article presents six major research challenges in managing performancevariability in maintenance. The article concludes by noting that a holistic theoryon maintenance work is needed to manage the variability and turn it intoa positive force. Maintenance has the potential to produce positive performancevariability and guard against negative instability in complex sociotechnicalsystems.

Keywords: maintenance; human and organisational factors; system safety;

performance variability; organisation theory

1. Introduction

1.1. Maintenance work as safety-critical activity

Maintenance is a hazardous activity. In addition to hazards related to process or

production safety, maintenance work involves occupational safety hazards of various

kinds (physical forces, chemicals, radiation, etc.). Thus, maintenance organisations can be

considered as safety-critical organisations. This means that safety is a central goal of the

maintenance organisation and the organisation has to identify, remove, control and

prevent the various hazards associated with its work. Several accident investigations

have uncovered inadequate or faulty maintenance as one of the main contributors

to unanticipated events in various safety-critical domains, including the railway, offshore

oil drilling, chemical, petrochemical, aviation and nuclear industries (Department of

Transport 1989, Pate ´ -Cornell 1993, Marx and Graeber 1994, p. 88, Wright 1994,

Reason 1997, Hale et al. 1998, Kletz 2003, Reason and Hobbs 2003, Perin 2005, Baker

2007, Sanne 2008a). Thus, maintenance activities can be considered as having a highly

significant positive or negative impact on the effectiveness of the entire sociotechnical

system, including safety.

*Email: [email protected]

ISSN 1463–922X print/ISSN 1464–536X online

ß 2011 Taylor & Francis

DOI: 10.1080/14639221003725449

http://www.informaworld.com



Maintenance activities have been under various organisational changes and

restructuring initiatives, aiming at, e.g. reduced costs, increased availability of the

machines, better knowledge sharing and increased flexibility (Patankar 2005b, Reiman

et al. 2006, Pettersen 2008). This development has been going on for years in various

safety-critical domains. For example, in the aviation domain, the US Federal Aviation

Administration has identified a need for better efficiency in aviation maintenance to cope

with the ever-growing workloads caused by the increase in the number of passenger miles

flown by the airlines. The rise in traffic has not been accompanied by a similar increase

in resources, e.g. the number of maintenance technicians. Price-based competition

has reduced revenues and forced the companies to take various cost-cutting measures,

including outsourcing maintenance, leasing instead of buying aircraft and layoffs

(Patankar 2005b). In the nuclear industry, ageing plants and equipment (OECD/NEA

2000), the ongoing generation turnover (OECD/NEA 2001) and the deregulation of the

electricity market (Bier et al. 2001) have been the main drivers in the recent organisational

changes (see also IAEA 2001, OECD/NEA 2002). These changes put an even stronger

emphasis on understanding the maintenance work and its requirements as well asmanaging the performance variability in maintenance.

This article deals with the challenge of understanding maintenance work in

safety-critical organisations and its characteristic sources of performance variability.

The aim of this article is to review the current literature on maintenance work in

safety-critical organisations and to illustrate, with the help of the literature review and

theoretical work, the organisational research challenges of managing performance

variability in maintenance work.

1.2. Performance variability in maintenance

One of the challenges of maintenance work is that while it is necessary for the technology

in use, it can be a source of latent as well as active failures in the system. In addition, some

of the errors are hard to notice. Errors made during preventive maintenance can manifest

after a long time. Quick fixes in failure repairs might do more harm than good in the long

run. Event investigations almost always find actions outside the prescribed boundaries and

label these human errors or violations (Reason 1997). The foci of studies on human

performance variability are usually biased towards the negative effects of variability,

discounting the potential positive effects (cf. Hollnagel 2004, 2009b, Roth et al. 2006,

p. 181, Pettersen 2008, p. 84) or the fact that performance variability is found in all human

work (Hollnagel 2009b, p. 85), including maintenance. Human performance variabilityis as much a cause of safety as it is a cause of errors and accidents (Hollnagel 2009b).

Maintenance work is characterised by the requirement of acting under uncertainty

(cf. Norros 2004, Reiman 2007, Pettersen 2008). The amount of information, dynamic

relations between phenomena and the connections between environmental cues are so vast

that there can never be full certainty of the effects of the maintenance actions or of the

various phenomena associated with the object of work. It is important to understand

that uncertainty is never caused by an individual alone but is rather related to the object

of work, such as the condition of technical systems in nuclear power plants (NPPs) or the

reliability of measurement data in process control. The object of work contains

uncertainty; the progress and effects of work can never be fully predicted. This is whyemployees really should feel a suitable amount of uncertainty when dealing with them.

Recognising and coping with uncertainty is related to the development of expertise

340 T. Reiman



(Klemola and Norros 1997, Norros 2004) and decision making, in general. Thus,

maintenance work is by its very nature variable and requires variability also in human

performance.

In safety-critical organisations, rules and procedures are often considered to be a way

of improving the reliability of the activities of humans and organisations. This notion is

based on the (as such correct) notion that humans are forgetful and prone to error. Rules

and procedures try to control these ‘human’ characteristics. Rules and procedures are

considered as safety barriers to the troublesome variability of human performance

(cf. Hollnagel 2004, Oedewald and Reiman 2007a). Also, the rule designers often think of

procedures as tools for controlling the worker, not as tools for the worker to control his or

her work (Dien 1998, p. 181).

Performance variability has been studied from various perspectives, commonly placed

under the title of human and organisational factors. There are two main approaches to

counteracting and controlling performance variability: the first builds on restricting

and constraining human behaviour through rules and procedures, the second builds on the

strengths, competence and motivation of the personnel. The former approach is basedon the assumption that variability is inherently bad, whereas the latter approach

treats variability as a source of both successes and failures (cf. Hollnagel 2004, 2009b).

These same underlying assumptions can also be found in the maintenance literature.

Next, we will look at the current literature on maintenance work and illustrate the various

approaches taken towards managing the performance variability in the context of

maintenance.

2. Review of current literature2.1. Human errors as a source of performance variability

The focal point of concern for research in maintenance work has been on the performance

of individuals. Research has focused on unsafe acts, decision making and errors.

Following Reason’s (1990, 1997) groundbreaking studies of human error and maintenance

in nuclear power and aviation, research on human errors in maintenance in various

safety-critical domains has aimed at classifying, predicting and preventing human errors

or minimising their consequences (Marx and Graeber 1994, Laakso et al. 1998, Fleishman

and Buffardi 1999, Isobe et al. 1999, Latorella and Prabhu 2000, Gibson et al. 2001, Pyy

2001, Svenson and Salo 2001, Toriizuka 2001, Hobbs and Williamson 2002, 2003, Reason

and Hobbs 2003, Dhillon and Liu 2006). Maintenance errors have been traced as a sourceof several aviation accidents and incidents, and it has been reported that the number of

maintenance-related accidents has been on the increase (McDonald et al. 2000, p. 154,

ATSB 2001, p. 1). All in all, maintenance has been identified as a major source of latent

failures in sociotechnical systems (Reason 1990, 1997).

The aim of many studies has been the identification of the most common types of

errors as well as the most effective countermeasures. For example, Hobbs and Williamson

(2003) studied 619 self-reported safety occurrences involving aircraft maintenance.

They categorised error types, outcomes and contributing factors. According to the

study, the most frequent error was a memory lapse in which a person forgot to perform an

intended action (Hobbs and Williamson 2003, p. 195). Memory lapses were associatedwith pressure, fatigue and environmental (e.g. noise, lighting) contributing factors. When

discussing their results, they note that ‘it is possible that aircraft maintainers routinely

Theoretical Issues in Ergonomics Science 341



perform their assigned tasks in the face of challenges such as fatigue or time pressure’

(Hobbs and Williamson 2003, p. 199).

With a similar aim as Hobbs and Williamson (2003), Laakso et al. (1998, see also Pyy

2001) reviewed approximately 4400 failure reports from a Finnish NPP from the period

of 1992–1994 and searched for human errors related to maintenance. They were especially

interested in human-related common cause failures (CCF)1 and the mechanisms causingthese failures. For single human errors, they (Laakso et al. 1998) identified instrumen-

tation and control (84 cases out of a total of 206 single human errors) plus electrical

equipment (40 cases) as being more error prone than other kinds of maintenance. Laakso

et al. (1998) identified 14 CCF from the failure reports. The distribution of CCF was

similar to single errors; they all occurred to either instrumentation or electrical equipment.

Also, similar to single human errors and contrary to many other studies (Hobbs and

Williamson 2003, Reason and Hobbs 2003) the most dominant error category was that

of commission. Weaknesses in work planning and in the design and layout of the

equipment from the maintainability point of view (cf. Seminara and Parsons 1982)

contributed to many of the human errors identified by Laakso et al. (1998). These authorsfurther identified that most errors had stemmed from the refuelling outage periods and

plant modifications and that they were discovered only after the outage during power

operation.

Reason and Hobbs (2003) note that the most common human errors in maintenance

in NPPs as well as in the aviation industry are errors of omission: failing to do something

that should have been done (see also Reason 1990, Hobbs and Williamson 2003, Patankar

and Taylor 2004b). They also note that these errors are commonly associated with

reassembly or installation activities. They have categorised the major types of unsafe acts:

. recognition failures, such as misidentification of objects or signals and

non-detection of problem states;. memory lapses, such as failure in encoding, storing or retrieving information;

. slips of action, such as executing a familiar task wrongly due to absent-

mindedness or external distraction or slipping into a familiar course of action

when required or intended to perform a less familiar action;

. errors of habit, such as applying bad rules and developing bad routines that have

no immediate negative effect;

. mistaken assumptions, such as applying a good principle or rule-of-thumb in a

situation for which it is not appropriate;

. knowledge-based errors, such as problem solving in new situations or unfamiliar

tasks;. violations, such as intentional deviations from procedures to save time, get

the job done or due to personal thrill-seeking (Reason and Hobbs 2003,

pp. 40–58).

Some of the types of unsafe acts identified by Reason and Hobbs (2003) such as bad

routines can be called performance-shaping factors that increase the probability of human

errors. The conditions and factors influencing human errors have also been extensively

studied. Suzuki et al. (2008) asked in their article on aviation maintenance related

incidents, ‘why cannot (the already established) safety procedures prevent human errors’.

Their study showed that coordination problems weakened these safety procedures. They

call for increased sense of responsibility for preventing future coordination failures.Patankar and Taylor (2004a, 2004b) have listed the ‘dirty dozen’ (originally developed by

Gordon Dupont in 1993 as the key component to the human performance in maintenance

342 T. Reiman



workshops), or 12 performance-shaping factors that cause human errors in maintenance

(Figure 1).

The factors in Figure 1 show that errors are ‘caused’ by qualitatively very different

phenomena, including issues at the individual, group and organisational levels. In order

to shed more light on performance variability in maintenance as well as on how safe orunsafe acts are decided on, studies have tried to identify sources of professionalism

in maintenance, including decision-making strategies, rule following and mental models or

conceptions concerning the maintenance work.

2.2. Professionalism and decision making as contributing to performance variability

A classic work that sheds light on the technician’s work within an organisational context

is that of Orr (1996). He has conducted ethnography on the work of field service

technicians at Xerox. He noted that the technical knowledge and the professional identity

of the technicians were strongly dependent on face-to-face encounters between thetechnicians and on the task-related stories (‘war stories’) that they shared (cf. Barley 1996).

These stories ‘combine facts about the (copying) machine with the context of specific

situations’ (Orr 1996, p. 127). He (Orr 1996, p. 91) also argues that ‘when technicians talk

about specific machines in their territories (areas of responsibility), it is clear that these

machines are individuals. Their different histories, different patterns of use, and different

social environments have given them each a distinct character for those who know.

Given this individuality, the machines may be discussed with as much ellipsis as any

mutual acquaintance’. He noted that the technicians consider the machines to be both

‘perverse and fascinating’. What really interests the technicians is a failure situation that

they do not understand (Orr 1996, pp. 95–97). They take pride in being able to cope with

the machines. The technicians must diagnose, repair, maintain and adjust the machines

in an environment that is ‘inherently unpredictable’. Orr (1996, p. 104) notes that ‘in all of

these activities, and perhaps most critically in diagnosis, the technicians must understand

the machines’. Understanding is central also for anticipating and preparing for future

problems. Orr points out that the corporation had a different view of the technicians’

work, one that emphasised not understanding but rather the following of directive

documentation (see McDonald et al. (2000), for a similar finding in aviation maintenance).

On the other hand, ‘in providing directive documentation, the corporation is assuming

responsibility for solving the machine’s problems, and in the eyes of the corporation,

technicians are only responsible for failure to fix a machine if they have not used the

documentation. However, while the technicians are quite willing to let the corporationassume any blame, their own image of themselves requires that they solve the problems

if at all possible’ (Orr 1996, p. 111).

Lack of resources

Distraction

Pressure

Complacency

Stress Unsafe norms

Fatigue

Lack of knowledge

Lack of awareness

Lack of

assertiveness

Lack of

communication Lack of teamwork

Figure 1. The ‘dirty dozen’ elements causing errors in aviation maintenance according to Patankarand Taylor (2004a, b).




Orr (1996) conducted his study in a non-safety-critical domain. However, somewhat

similar tensions and ambiguities towards rules have been identified by McDonald (2006)

in aviation, Bourrier (1996) and Reiman and Oedewald (2006) in nuclear and Lawton

(1998) and Sanne (2008a) in the railway context. Orr’s (1996) findings on the motivation

and interests of technicians also parallel those identified by Reiman and Oedewald (2006)

in NPP maintenance.McDonald (2006) summarises the results from a series of European projects

concerning aircraft maintenance. He notes that the technicians did not follow the

procedures routinely. They often justified their violations by reporting that there were

‘better, quicker, even safer ways of doing the task than following the manual to the letter’

(McDonald 2006, p. 161, see also McDonald et al. 2000, cf. Dekker 2005, pp. 134–138,

Hobbs and Williamson 2003, p. 196). Also, according to McDonald (2006, p. 163), for

many aircraft maintenance organisations, ‘there appears to be an unresolved tension

between effective planning and the requirement of flexibility to meet the normal variability

of the operational environment’. He then generalises from a number of surveys in different

organisations the core professional values of aircraft maintenance personnel. These valuesincluded the following characteristics:

. strong commitment to safety;

. recognising the importance of teamwork and coordination;

. valuing the use of one’s own judgement and not just following rules;

. being confident in one’s own abilities to solve problems;

. having a low estimate of one’s vulnerability to stress;

. being reluctant to challenge the decisions of others.

Taylor and Christensen (1998, pp. 83–84) have defined the characteristics of a

professional aviation maintenance technician as including competence, centrality, controland commitment regarding flight safety.

McDonald (2006) noted that the above-mentioned professional values in many ways

matched the deficiencies found in the same organisations. Professionalism compensates for

organisational dysfunction. A problem is that the ‘double standard’ of work as formally

specified and unofficial ways of working is hidden (McDonald 2006). McDonald et al.

(2000) and McDonald (2001) argue that evidence from aviation maintenance indicates

that the current quality and safety management systems do not provide an adequate

picture of the way the work is actually carried out, partly due to this ‘double standard’.

Reiman (2007) studied maintenance culture at three Nordic NPPs. He carried out three

in-depth case studies with his colleagues (Reiman et al. 2005, Reiman and Oedewald 2006).

The case studies employed interviews, surveys, seminars, document analysis and group

work. The study consisted of an analysis of maintenance culture combined with a core task

analysis (cf. Norros 2004) of the maintenance core task. On the basis of core task

modelling, the maintenance core task was defined as balancing between three critical

demands: anticipating the condition of the plant and conducting preventive maintenance

accordingly, reacting to unexpected technical faults and monitoring and reflecting on the

effects of maintenance actions and the condition of the plant. The overall objective of the

maintenance activity was defined as follows: maintaining the operational reliability and

the economic value of the nuclear installation so that its power production can continue as

long as planned. The case plants differed in terms of their emphasis on, the interpretation

of and the culturally accepted means of carrying out the demands of the maintenance task.Despite this, they shared similar conceptions about the goals of maintenance and the

paramount importance of safety in the maintenance of an NPP. However, sense of control,

344 T. Reiman



sense of personal responsibility and organisational changes emerged as psychologically

challenging issues at all the plants. Critical attitudes towards management and the values

prevalent in the organisation existed at all the plants. The hands-on nature of maintenance

work was emphasised as a source of identity at the NPPs. Maintenance work produced

a feeling of meaningfulness, especially when there were technical problems to solve or

failure repairs to conduct. Overall, the importance of safety was taken for granted,

but there was little reflection on the cultural norms concerning the appropriate means to

guarantee it. Reiman (2007) proposes that maintenance work should be considered as

knowledge-intensive work and concludes by arguing that the nature and significance

of maintenance work should be better acknowledged by the maintenance workers

themselves and by other parties (e.g. operations and technical groups).

Carroll et al. (1998) have studied decision making in the context of maintenance.

They present evidence for the failure to give ‘due consideration for preventive

maintenance’ (cf. Seminara and Parsons 1982, p. 186) in two domains: nuclear and

chemical. They show how mental models and implicit assumptions influence decision

making. Both industries had trouble developing their programmes of preventivemaintenance. Carroll et al. (1998, pp. 109–110) argue that difficulties in managing

maintenance arise, in part, from limitations in mental models, which they define as

individual, shared and embedded beliefs and understandings. They write: ‘Preventive

maintenance is a prototypical activity that seems to be a low priority in the face of

immediate demands to keep the machines running at lower cost, and the ultimate effects

of deferred maintenance can be denied, ignored, or blamed on others’ (Carroll et al. 1998,

p. 110). They then demonstrate how a company in the chemical industry tried to change

from a culture of ‘corrective maintenance’ to a culture of ‘preventive maintenance’ with

a maintenance game. The mental models, however, proved very hard to change.

These studies have illustrated the complexities of the interaction between thetechnicians and the technical system as well as the challenges facing the personnel in

making decisions within the social context of the maintenance organisation. The findings

emphasize the need to take into account the social and organisational aspects in order to

better understand human performance and its variability in maintenance.

2.3. Performance variability in a social context

Rule bending and rule deviations are a typical management challenge in maintenance

contexts. There are some studies that shed light on the social factors affecting rule bending.

For example, in the maintenance organisation of an NPP studied by Reiman andOedewald (2006), rules and procedures were a source of tension and ambiguity for the

personnel. They were afraid of losing their professional identities as skilled craftsmen and

becoming ‘a small cog in a big machine’, but they felt this was the goal of the organisation

and also to some extent their daily reality (Reiman and Oedewald 2006). They felt that the

strong tendency to standardise and proceduralise tasks threatened their job motivation,

the meaningfulness of the work and their ability to carry out the daily work (cf. Hackman

and Oldham 1980, p. 75, Bourrier 1996, p. 106, Dekker 2005). Sanne (2008a, p. 647) also

points out a need in railway maintenance to identify the practices of risk taking and rule

bending and the process of how they are produced and reproduced in the social context of

maintenance.Often adaptation and interpretation of rules is considered an integral part of the

work. Pettersen (2008) has studied the human role in producing safety in aircraft




line maintenance. His interest was to study the social structure and human agency, and

their relationship in the production of safety. He provides empirical accounts of how the

technicians conduct their work and make decisions within the social context in which

they work. He utilised participant observation, document analysis and interviews in a

Norwegian aircraft maintenance company. His analysis emphasised the roles of human

agency – the capacity of the technicians to decide and act – and the unofficial social systemin creating safety. Pettersen argues that the technicians constantly sought to revise their

knowledge and their degree of certainty about it. He illustrates how the technicians both

valued and distrusted formal descriptions of work (procedures, rules and regulations)

as guides for work practice. The technicians adapted their practices and local organising

depending on the situation at hand. These adaptations were, however, supported by a

strong cultural imperative to achieve safety and by the technicians’ awareness of the

imperfection and uncertainty of both knowledge and practice. Pettersen (2008, p. 85)

notes that ‘technicians’ individual choices and actions take place within the context of a

historical system where, among other things, maintenance planning, materials supply,

a legal institutional framework and safety management systems constitute structuredrelationships of actions that constrain and enable how the technicians can act in ways not

regularly accounted for’.

Bourrier (1996, 1999) has compared practices in four maintenance units in France

and the USA. She spent between 3 and 4 months at each site and conducted a total of

300 interviews. She noted differences between the units in, e.g. the coordination of work,

the structuring of the tasks and the role of procedures during the annual outages of the

plants. Each plant had its own official or unofficial way of following the procedures and

acting when the procedures did not cover the work in question. For example, at one of the

plants, organisational reliability was based on situational improvisation when no suitable

procedures could be found. Foremen unofficially accepted the practice and trusted theexpertise of their workers and themselves. At another plant, reliability was based on

following the procedures strictly; for cases where no procedure existed, the plant had

a procedure with which the appropriate procedure could be quickly produced. According

to Bourrier, a drawback of this strategy was that it did not support individual decision

making on the part of the workers. In conclusion, Bourrier states that ‘local adjustments

to and re-arrangements of rules and, at times, even rule violations, are not only constant

but necessary for organizations to effectively pursue their goals’ (Bourrier 1996, p. 106).

Sanne (2008b) studied incident reporting and storytelling in railway infrastructure

maintenance. He was interested in the reasons for the low number of reported incidents

and its consequences for organisational learning. He illustrates how different accident

aetiologies shape incident reporting and storytelling. For the railway technicians, an

accident is seen as a ‘breakdown of occupational practices, skills, and values, rather than

as a system breakdown’ (Sanne 2008b, p. 1212). Consequently, this shapes what is

considered an incident in the technician’s community. On the other hand, incidents that

pose a threat to the technician’s identity as a responsible and careful professional can be

framed as insignificant – a one-off event – in terms of learning. Shame, blame and fear of

disciplinary actions decrease the willingness to submit reports. Some of the technicians had

never thought about what the organisation as a whole can learn from incidents. Also, the

modest and often negative feedback the technicians get from the reports they have made

does not promote further reporting – or learning. Incidents that do not result in injury are

normalised as ordinary, unproblematic practice, not requiring learning or other correctiveactions. In the occupational aetiology of accidents, the attention warranted by an event is

defined by the severity of its consequences rather than its causes and potential effects.

346 T. Reiman



Storytelling, in contrast, is a more attractive practice to technicians, since it promotes

the reproduction of occupational communities and provides a means of transferring

knowledge to, e.g. newcomers. However, storytelling is usually restricted to local practice

and does not address the systemic causes behind the accidents and incidents. The stories

are also often told as a way to justify and legitimise technicians’ practice rather than

to teach something new.

Studies on the social aspects of maintenance have clarified the role that group-level

factors play in the maintenance context. They have shown how the social context plays

a dialectic role with the formal ways of organising and managing maintenance. Local

practices influence and are influenced by the organisation-level practices, procedures and

rules of conduct.

2.4. The influence of the organisation and management of maintenance activities

on performance variability

Pettersen and Aase (2008) have studied safe work practices in aviation line maintenance.

They argue that the features of safe practices in aviation maintenance are dependent on

slack resources in the organisation. They conducted a qualitative study of the line

maintenance organisation of a Norwegian regional airport with the aim of understanding

the dynamics of safe work practices. They demonstrate how the line technicians

emphasised the importance of practical competence, e.g. in defining the ‘normal’ operative

state of a component or a technical system. The formal maintenance documentation

system had ‘grey areas’, where experience, practical skills, support from colleagues and

trial-and-error strategies were required in troubleshooting and problem solving.

Level-headedness and humility about one’s own skills and knowledge were considered

as desirable traits by the technicians. Further, as the technicians were under a constant

pressure to get airplanes operational within planned schedules, they had institutionalised

a way of slowing things down in order to create slack in the otherwise tightly coupled

system. Recent changes in the organisation further reduced slack and transformed, e.g.

communication practices between technicians and pilots. Pettersen and Aase (2008)

conclude that slack in organisations (e.g. in the form of time, knowledge, competence

and tangible assets) can be viewed as structural preconditions for the existence and

effectiveness of several forms of safe practices. They caution against treating slack solely as

a waste to be rid of in the organisation.

Herrera and Hovden (2008) have studied leading (proactive as opposed to reactive)

indicators applied to aviation maintenance. Their aim was to understand leadingindicators in the framework of resilience engineering and their ability to provide

information on changes in risk. They argue that leading indicators should provide a signal

of unintended system interactions and focus on the normal operation of the system instead

of failures. They give some examples of leading indicators: ‘the resources available, the

capacity to identify circumstances beyond the experience (of the maintenance personnel),

the possibility to reflect-on-action, openness, communication, the current technical state

of the aircraft, maintenance oversight, and implementation of preventive maintenance’.

Bier et al. (2001) have studied the effect of deregulation on safety in the US aviation

and rail industries and the UK nuclear power industry. They raised the issue that cutting

corners in maintenance in these industries is a troubling trend from the safety point of view. Another growing issue after deregulation was mergers and acquisitions. Bier et al.

point out that organisations cannot always accurately predict the (safety) impacts of




restructuring and downsizing. The safety impacts of organisational changes have been

surprisingly little studied in maintenance or other safety-critical fields (Reiman et al. 2006).

Ramanujam (2003, p. 614) argues that the ‘current explanations of the organizational

origins of accidents understate, or even ignore, the role of organisational change’.

According to the few studies that have been made, some common issues of concern

in organisational changes include vague responsibilities about safety matters during the

transition period from one organisational form to another, loss of competence,

deteriorating morale and employee motivation, deteriorating relations between employees

and managers, and stress or excessive workload among the personnel (Bier et al. 2001,

Kecklund 2004, Reiman et al. 2006, Pettersen and Aase 2008, Herrera et al. 2009). Herrera

et al. (2009) express in their study on changes in aviation a worry on the safety effects

of increased subcontracting of maintenance activities at airline operators. They recom-

mend that before outsourcing the companies should conduct risk analyses to consider

what parts of the maintenance function and what competence is needed to keep in-house in

order to ensure safety.

Pettersen and Aase (2008) studied the role of change and restructuring in the linemaintenance organisation of a Norwegian regional airport. The technical organisation,

including its line maintenance department, was merged with a larger airline’s

technical organisation, which functioned as a business unit separate from the airline.

They write:

Before parts of the technical organisation were sold and all technical functions were ‘underthe same roof’, all personnel and resources could be directed towards line maintenance(i.e. first-line operations) if needed. Enabled by the social structure and culture of theorganisation . . . these resources functioned as reservoirs of knowledge, competence andresources that could be used by line maintenance in their efforts towards creating safepractices. As soon as parts of the organisation were sold, for example the engine repair shop,structural ‘walls’ were built taking away slack and dwindling the knowledge and competenceboundaries of the line maintenance department. These changes do not directly produceaccident risk, but change the dynamics of operational practice.

They also noted changes in the information flow, opportunities for learning, communi-

cation between technicians and pilots and the amount of training (especially refresher

courses) offered. This means that change efforts alter the ways in which safety is achieved.

The line technicians expressed their concern about losing operational experience and

knowledge in the new organisation.

Reiman et al. (2006) have studied the safety effects of recently implemented changes

in four Nordic NPP maintenance organisations. Their analysis of selected changes showed

that all of these changes faced plenty of obstacles and had unforeseen or unintendedconsequences or side-effects on organisational practices and culture. Cost reduction,

enhancement of the efficiency of maintenance activities and maintaining and developing

competence were identified as goals of all of the reviewed changes. In addition to these

goals, the interviewees came up with many other secondary or implicit goals. Many of

these dealt with cultural issues such as communication, status, personnel issues and

stagnation (‘waking up’ the organisation). Some of these goals were explicit, some implicit,

but they all affected the way the change was carried out and eventually the outcomes of the

change process. Despite the prevalence of ‘soft’ goals, few organisational and personnel

development methods were used in these cases. Change management was approached

from a very technical standpoint, and Reiman et al. (2006) concluded that a more dynamicframework, including the consideration of human and organisational factors, is needed

for assuring safety during change in safety-critical organisations.

348 T. Reiman



Studies on the organisation and management of maintenance activities have illustrated

the effect of the organisational structural solutions and especially, change in those

solutions to the performance variability. Change creates instability in the organisation and

exposes the organisation to the negative effects of performance variability. These effects

are exacerbated if there is ambiguity among the personnel about the organisational goals

or priorities.

2.5. Goal conflicts creating conditions for performance variability

Due to the characteristics of the work (close contact with machinery or a running process,

exceptional work conditions, time pressures, etc.), maintenance operations are challenging

in terms of occupational safety. Maintenance activities can thus be viewed as posing a risk

to humans (Lind 2008, p. 928). Occupational safety has been studied, for example,

in aviation maintenance by Neitzel et al. (2008), in railway maintenance by Sanne (2008a)

and Farrington-Darby et al. (2005) and in Finnish industrial maintenance, in general,by Lind (2008). Research on occupational safety in maintenance often draws on some of

the theoretical approaches presented above – often on the human error view. Occupational

safety studies that consider the relation of occupational safety to the other goals of the

system are reviewed in this section.

Mercier (1988, pp. 86–87) characterises the maintenance work of an NPP as follows:

‘It is rare for so many non-repetitive tasks to be concentrated in an industrial environment

that is so very hostile to human activity. The forces in this environment are considerable.

Temperatures, pressures, the multitude of fluids, mechanical power, omnipresent

electricity, even the sheer weight of the equipment . . . all culminate to make maintenance

actions potentially dangerous and to weigh against success. The ‘‘nuclear’’ hazard and the

associated radiation protection restraints are simply one more risk, but a risk that is often

quite minimal compared to the others’. In all safety-critical domains, there are various

occupational risks that affect the way work is carried out and that have to be balanced

with the other goals of the organisation, such as efficiency, economics and system or

production safety.

Reason and Hobbs (2003, p. 59) argue that different forms of human errors are

associated with incidents threatening the safety of operation and with incidents threatening

worker safety. Thus, different remedies are needed to address both types of outcomes.

According to Reason and Hobbs, the three most common types of errors threatening

operation safety are (in descending order) memory lapses, violations and knowledge-based

errors. Worker safety is threatened especially by slips of action and violations. Lind (2008,pp. 929–930) found in her study of accident reports that the most important

latent conditions in fatal or non-fatal occupational accidents in industrial maintenance

were defects in planning or managing the work and defective work instructions

(cf. Reason 1997).

Sanne (2008a) conducted ethnography of railway infrastructure maintenance in

Sweden. He studied how the maintenance personnel framed occupational risk-taking.

Sanne (2008a, p. 645) notes that the railway technicians have a double-order relationship

to risk: ‘their job of achieving public safety and protecting the public from risk necessarily

exposes them to occupational hazards’. In addition, they ‘must trade train safety against

other ends, such as punctuality, time limits, economic constraints, and their own safety’(Sanne 2008a, p. 646). Sanne argues that the technicians took occupational responsibility

for transportation safety hazards since their tasks are safety-critical. They further




considered that it is important to understand the safety consequences of one’s actions in

a tightly coupled and complex railway context. He illustrates how the technicians bear

a sense of responsibility for each other’s safety. Occupational risks are constructed as

manageable with reference to technicians’ competence and to mutual responsibility and

trust in the team. Sanne (2008a, p. 652) writes that ‘interdependence in a team and fatal

consequences imply that to be a responsible railway technician, one must be able to take

care of oneself and ensure team safety, irrespective of formal roles and responsibilities’.

On the other hand, the technicians’ occupational responsibility also entails responsibility

towards passengers. An attitude or collective aim of ‘making it work’ existed in the

technicians’ community, and it biased their actions, encouraging them to do what is

needed under given circumstances, even if this entails ‘rule bending and risk taking’

(Sanne 2008a, p. 653). They are sacrificing their own health and safety to assure the safety

of others. Sanne concludes his analysis by noting that technicians take occupational risks

to compensate for inadequate planning, time or resources: matters that should have been

handled by the corporation. Thus, risk taking is an expression of responsibility, skill and

control, and it is not conceived as ‘risk taking’ for the sake of thrills or adventure.Patankar (2005a, 2005b) presents an overview of ethical challenges in aviation

maintenance in a case study of ethical challenges faced by six mechanics working for a

major airline company in the USA. He distinguished several characteristics that inspired

one of the mechanics, ‘Joe’, to raise safety issues and violations: personal confidence in his

skills, professional pride, support from peers and an understanding that the general public

is his real boss; that he is working to guarantee their safety. Patankar identifies three main

ethical challenges in aviation maintenance: (1) ‘data smoothing’, where data is falsified

so that it is within certain allowable limits, (2) ‘pencil-whipping’, which means signing for

a job that has not been performed and (3) ‘not knowing when to act’, which refers to the

inevitability of numerous procedural violations in aviation maintenance on part of all theparties involved. Patankar (2005a, 2005b) then considers the transition of mechanics into

managers and raises the issue of why some managers who are familiar with the regulations

still put their mechanics into higher risk situations. He proposes that the reason is that

when a mechanic becomes a manager, his goals change from being primarily held

accountable for safety to being held accountable for on-time performance. For the

managers, the major ethical challenge is thus the issue of safety versus financial survival.

Patankar (2005b) points out how the ethical challenges in maintenance are exacerbated by

the economic and social challenges of the industry as a whole. These economic challenges

have led airlines, e.g. to outsource maintenance activities and lease airplanes instead of

buying them. The mechanics perceived that the industry is focusing more on cost-cuttingand profits than safety, and felt that this decreased their job satisfaction. Patankar (2005b)

recognised four common themes among the mechanics he interviewed: (1) passion for

aviation, (2) commitment to safety, (3) role models and defining moments and (4) ‘square

peg in a round hole’, in other words a sense of disconnection between the maintenance

workforce and the company. Patankar (2005b, p. 47) writes: ‘Ultimately, the above

individuals are committed to safety because of a strong sense of social responsibility

entrusted in them by the virtue of their profession’. He also shows the importance of role

models and stories in the character development of the mechanics. The different

orientations of the supervisors and the mechanics are illustrated by a quote from one

technician:I had a supervisor tell me once, that the airplanes come in for a ‘check and service’. I explainedto them, no they do not, they come in for ‘inspect and repair’. There is a major difference

350 T. Reiman



there. . . . I have very serious problems with that thought process (of avoiding repairs), theairplane is in, it’s down, it’s there to be repaired, and sent back out in as good a fashion aspossible. . . . (Patankar 2005b, p. 52)

The dual objectives of maintenance in ensuring safety and maintaining operations

or schedules and the general tension between safety and efficiency have been identified in

many studies (see, e.g. Endsley and Robertson 2000, Dekker 2005, Kettunen et al. 2007,Gomes et al. 2009).

Professionalism, including competence, personal responsibility and capacity for

situational judgement or adaptation, emerges as an important dimension in balancing

the different organisational goals. Thus, maintenance personnel seem to consider the

capability for performance variability an integral aspect of professionalism. However, it is

not clear what constitutes professionalism in a maintenance context and by what means

it can be achieved and developed.

2.6. Competence management and training as controlling for performance variability

Maintenance requires a multitude of skills and a wide range of knowledge. Still,

maintenance is often considered as mostly manual labour. Perin (2005) shows in her

analysis of event handling in the nuclear industry how the industry’s culture of control

is very engineering driven, emphasising probabilities, measures and risk estimates. This

culture largely discounts qualitative and experiential knowledge. These collective ways

of thinking influence, e.g. how the power plants convert ambiguity into certainty and

separate judgement from reason (Perin 2005, p. 225). She also emphasises the importance

of maintenance in designing and operating an NPP and agrees on the findings of, e.g.

Barley (1996) and Oedewald and Reiman (2003) on the unique knowledge of technicians

and on their work as being more similar to knowledge work or engineering work than

is commonly accepted. Technicians’ knowledge is more contextual than engineers’ or

designers’ knowledge, and it is based on real-time experience of equipment and its use.

Also Samurcay and Vidal-Gomel (2002, p. 159) have stressed the fact that (electrical)

maintenance work requires both technical knowledge derived from engineering science

and pragmatic knowledge including an understanding of the overall work process in the

organisation.

Endsley and Robertson (2000) have applied the concept of situation awareness (SA) to

aircraft maintenance teams. They define SA as a three-level phenomenon; level 1 SA

means being aware of the aircraft system one is working on. Level 2 SA involves the

technicians’ comprehension of the significance of observed system states. Level 3 SAdenotes the ability to project the state of the system in the near future. Technician with

level 3 SA would be able to project what effect a particular defect might have on the

performance of the aircraft in the future. Endsley and Robertson (2000) performed task

analysis to determine the specific SA requirements in the aircraft maintenance arena.

They also conducted an SA resource analysis to identify the resources used in the

maintenance environment to achieve the identified SA requirements. The results indicated

that the largest problem for team SA exists when there are gaps due to, e.g. mismatched

goals, lack of information or lack of understanding between organisations or individuals.

Based on the results, a team SA training programme was developed for the airline.

Training needs were found in five areas: shared mental models, verbalisation of decisions,shift meetings, feedback and general SA training. Endsley and Robertson (2000) conclude

by noting that providing personnel with knowledge is important but not sufficient for the








have been shown to be important for explaining performance variability in the

maintenance context. However, current research on maintenance work is fragmented in

terms of its focus; individual studies are seldom put into the larger context of safety science

or maintenance human factors.

3.2. Need for a theory to understand performance in maintenance context

Previous studies have tried to explain the performance variability mainly by looking at the

negative outcomes of the variability (errors), social factors affecting the variability

(Pettersen 2008, Sanne 2008b), the knowledge and psychological (mental) characteristics

of the maintenance personnel (Endsley and Robertson 2000, Patankar 2005a, 2005b,

Oedewald and Reiman 2007b, Rouse 1979) or the organisational functions such as safety

management or effects of outsourcing (Bier et al. 2001, Taylor and Thomas 2003,

Patankar and Taylor 2004b, Reiman et al. 2006, Herrera and Hovden 2008, Suzuki et al.

2008). Many studies have combined two or more of the above-mentioned dimensions (Orr

1996, Bourrier 1999, McDonald 2006, Reiman 2007, Pettersen and Aase 2008). Next, wewill look at the research challenges in combining these various foci into a holistic view on

maintenance work in safety-critical organisations.

Patankar and Taylor’s (2004a, 2004b) list in Figure 1 is focused on the dimensions

creating negative performance variability in maintenance. A similar list could be devised

on the elements creating positive variability in the maintenance organisation. Figure 2

presents a preliminary outline of a ‘pure dozen’ success factors in maintenance.

Figure 2 illustrates the multilevel nature of the challenges of effective maintenance

work. For example, motivation is an individual-level dimension (having to do with the

mental states and models of the personnel), whereas norms belong to the social level and

work design to the organisational level. The dimensions in Figure 2 are not exhaustive

in terms of describing a high reliability maintenance organisation, and they probably

partly overlap. In addition, the dimensions as such do not explain the mechanisms or

process by which, for example, norms, communication or complacency are created in the

organisation. Some indications of the mechanisms and various social processes can be

found in the literature, and these are dealt with in the next section.

The organisational challenges of maintenance stem from the nature of the maintenance

core task and its inherent variability. In order to understand the performance variability

in maintenance, the maintenance core task and its demands in various domains have to be

understood first. The core task of maintenance and the inherent hazards of the technology

that is maintained have to be taken into account (Reiman 2007).

Figure 3 presents a conceptualisation of the key content themes and researchchallenges of maintenance work in safety-critical domains. The content themes are based

on the analysis of the current literature and the theoretical framework of key

Flexible organizationand slack resources

Social permission tocarry work thoroughly

Self-criticism andreflection

Motivation and mentalresources

Norms supportingsafety

Vigilance and energy

Adequate task andsafety knowledge

Situation awareness

Assertive attitude tosafety issues

Clear communicationFunctioning teamwork

and cooperation

Good task and workdesign

Figure 2. Positive elements affecting maintenance, corresponding to and opposed to the ‘dirtydozen’ identified by Patankar and Taylor (2004b) arranged in order from individual to social toorganisational factors.

354 T. Reiman



organisational dimensions proposed by Reiman and Oedewald (2007). Figure 3 illustrates

the phenomena underlying the success factors of Figure 2, such as mental states

(vigilance), social processes (norms supporting safety) and organisational functions

(flexible organisation). The figure also includes technological hazards and the maintenance

core task as influencing elements, as well as performance variability as the main elementof concern in studies of human and organisational factors in maintenance to date.

The content themes of past research are categorised into four main organisational

dimensions in Figure 3: mental states and models, organisational functions, social

processes and performance variability. When considering safety and effectiveness of

maintenance, the technological hazards as well as the core task of the maintenance

function have to be taken into account in addition to the four dimensions. Next, the

research challenges are elaborated based on Figure 3.

4. Research challenges in maintenance organisations

4.1. Role of the individual in the maintenance function

More research is needed to better understand and conceptualise the role of individual

and social factors (the role of human agency, cf. Pettersen 2008) in performance variability

of the maintenance activities. This concerns questions such as individual assertiveness

concerning safety issues or the role of social networks and informal connections in daily

work. When assertive safety behaviour is against the norms of the workplace, what factors

define how the employee will act? How strong of a coupling exists between the official

system and the informal social organisation? How do the gaps in these two ‘systems’ affect

safety? Furthermore, research should address the tensions and differences in prioritiesand conceptions between maintenance and other functions of the company. Especially, the

significance of the differences to system safety as well as the well-being of the maintenance

Effectiveness andperformance

variability

Mental states and

mental models

Organisational

functions

Social processes

Maintenance

core task

3. Tools for the management of maintenance activities

1. Role of individual factors in maintenance

6. Leading indicators foreffective maintenance

5. Holistic evaluationmethods for maintenance

organisations

2. Effect of socialprocesses on performance

4. Maintenance coretask and its

requirements

Technological

hazards

Figure 3. The main dimensions of organizational factors and the associated research challengescritical to safety of maintenance.






motivation . . . in the process’. On the other hand, the maintenance personnel acknowledge

the impossibility of proceduralising all the aspects of the maintenance work and the

inadequacy of the procedures to cope with the realities and surprises of daily work

(cf. Hirschhorn 1993, p. 140, Bourrier 1996, Orr 1996, Carroll et al. 1998, Dien 1998,

Dekker 2005, Reiman 2007). The maintenance personnel seem to consider a key part of

one’s professionalism to be the knowledge of how to interpret, apply and neglect the

procedures in a manner that work can be carried out as thoroughly and as efficiently

as needed (cf. Hollnagel 2004, 2009b). Succeeding with this adaptive and variable way

of working requires a good knowledge of the task and its hazards.

An experienced maintenance team might manage daily tasks without a perfect

understanding of the fault mechanisms and details of the working methods. This is due to

the fact that their gradually formed tacit knowledge works well enough in most situations.

In spite of that, several event investigations have illustrated how the weaknesses of the

existing knowledge surface when established routines are challenged by unusual

disturbances or events (Hopkins 2000, Kletz 2003, Feldman 2004). One means of tackling

this problem is to analyse the maintenance personnel’s conceptions and mental models of their task. Norros (2004) has argued based on her studies in various safety-critical domains

that reflective as opposed to procedural orientation towards work facilitates learning.

The reflective orientation is also more flexible (i.e. variable) and adaptive to the situation

at hand than the procedural orientation. Organisational processes such as training,

learning practices and leadership should encourage this reflective orientation.

4.2. Social processes in maintenance

The second research challenge is related to the effects of social processes (e.g. formation

of norms, social identity) on performance variability in maintenance. In terms of social

processes, more information is needed on the ways technicians develop their practices and

learn from their daily work within the social structures of their company (Pettersen 2008,

Sanne 2008b). For example, how do the various safety management systems affect

field-level practices? How do the social identities of the maintenance personnel incorporate

models of errors, accidents and professionalism? What are the potential safety

consequences of strong professional identities that are based on local practice and

individual responsibility (Sanne 2008b) versus those based on collective responsibility and

structured practice (Reiman 2007)? Do these solutions depend on the nature of the core

task or the cultural features and history of the organisation? Both quantitative

questionnaire studies and qualitative case studies have a role in clarifying the contentsand effects of the various individual, social and organisational level dimensions.

Maintenance workers need to adapt to local circumstances and sometimes contradic-

tory goals, and work with the skills, resources, tools and time that they have. Norms

and local practices develop, and subcultures form based on technical disciplines, hierarchy

and physical location. The bending of rules or ‘innovative’ utilisation of tools is used to

compensate for organisational deficiencies and to accomplish goals deemed professionally

important. In many cases, minor adjustments to local procedures do not constitute

negligence but are done with good intentions (to get the job done, to save money). The

work and the organisational processes can be such that employees have to bend the rules

in order to get the work done. All this happens in a social context, where maintenancepersonnel jointly construct their view on the work and their image of the maintenance

profession, and both conceptions create and shape their practices. Pettersen et al. (in press)




remind that informal practices are as much part of the maintenance system as the formal

organisational structure.

Social processes and informal practices of the maintenance communities can lead to the

normalisation of small deviations and incidents as ordinary aspects of the work

(Sanne 2008b). Normalisation of deviance means a process where small changes – new

behaviours, technical anomalies or variations that are slight deviations from the normalcourse of events – gradually become the norm, providing a basis for accepting additional

deviance (Vaughan 1996). Normalisation of deviance produces disregard and misinter-

pretation – neutralisation – of potential danger signals. A signal of potential danger is

information that deviates from expectations, contradicting the existing worldview

(Vaughan 1996, p. 243). Since many maintenance actions and practices have effects that

carry over a long time period, the normalisation of deviance is a particularly significant

social process in the maintenance domain.

4.3. Management of the maintenance organisation

Many studies have shown that the achievement of effective and reliable maintenance

is dependent on good management of the maintenance organisations. Management does

not mean solely upper-level managers, but all activities geared towards ensuring that the

organisation is capable of functioning effectively. Maintenance organisations need to be

able to anticipate and plan for both expected events as well as unexpected disturbances.

Organisations have to be able to respond in a flexible manner to breakdowns and changes

in tasks, and they need to be able to update their knowledge of the sociotechnical system

that they are simultaneously part of as well as maintaining.

Two of the major challenges facing maintenance organisations are the almost constant

process of societal change and the various restructuring initiatives launched within the

companies owning the maintenance organisations (Kecklund 2004, Reiman et al. 2006).

These are reflected in the maintenance organisations as organisational changes. Changes

have generally been perceived as stressful and causing uncertainty among the workers

(Reiman et al. 2006). McDonald (2001, p. 223) warns that organisations that are based on

unofficial practices are especially vulnerable to changes (in technology, organisations and

personnel). Maintenance belongs to that category. Changes in maintenance have usually

been heavily technology driven (cf. Clegg and Walsh 2004). Human factors have been

considered only when problems occur, e.g. the personnel show ‘change resistance’ or do

not otherwise act as planned by the change agents (Reiman et al. 2006). Change

management is an organisational function that would benefit from a more explicit focuson human and organisational factors. Research should aim at providing tools and

methods to accomplish this.

Training is one of the instruments for creating an awareness of hazards as well as

sufficient skills for carrying out the work in a safe manner. An ongoing generational

change in the workforce calls for tools to analyse the existing know-how of the personnel

so that effective training programmes can be created. Maintenance work in safety-critical

organisations is not a routine-like activity that could be carried out just by following the

procedures. It requires different types of skills and knowledge. Practical craftsman skills,

overall understanding of the functioning of and couplings between the systems as well as

technical knowledge about the materials and equipments are needed in maintenance work.Thus, in addition to the tacit knowledge about maintenance practices and specific tasks,

understanding of the theoretical basis of technical phenomena and work processes are

358 T. Reiman



essential contents of the know-how of the personnel (Perin 2005, Oedewald and

Reiman 2007b).

Learning from near misses, failures and events is important for the continuous

improvement of maintenance activities. This involves certain challenges. First, reporting

of events is often inadequate. Second, even if events are reported, there are difficulties

in analysing the significance of events in terms of what they tell about the effectiveness of maintenance activities. Third, even if analysed correctly, getting the results back into the

field presents its own challenges. Consequently, maintenance activities remain rooted

in local history and local adaptations, without necessarily taking the entire organisation

or the experience of other parties into account.

Another approach to competence management is to seek to decrease the probability

of the occurrence of human errors with various human performance management and

error prevention tools (Reason and Hobbs 2003, Patankar and Taylor 2004a). Reason and

Hobbs (2003, p. 95) list a number of error management techniques in maintenance,

including training, work planning, job cards, licence-to-work systems, licensing and

certification, audits, procedures, disciplinary procedures, human resource managementand total quality management. They note that these techniques have not been effective

in preventing a steady rise in maintenance-related errors during the past decade. They

comment on the techniques that ‘their limitations include being piecemeal rather than

principled, reactive rather than proactive, and fashion-driven rather than theory-driven’.

Safety management systems are an integrated formal way of managing organisations

and their safety. Many safety management systems are based on a rational or a

non-contextual image of an organisation (Reiman and Oedewald 2007). The role of

management in supervising and directing organisational behaviour is emphasised. Waring

and Glendon (1998, p. 175) criticise safety management systems that are based on an

overly rational image of the organisation and argue that they may be only partly effectivewhile creating an illusion that the risks have been fully controlled (see also Waring 1996,

p. 46, Dekker 2005, p. 2, Perin 2005). The reality of organisational life is usually very

different from the ideal set out in formal documents and systems. For example, in the

context of aviation maintenance, McDonald et al. (2000) and McDonald (2001) argue that

evidence indicates that the current quality and safety management systems seldom provide

an adequate picture of the way the work is actually carried out. The social structure and

the inherent performance variability need to be taken into account. The role of safety

management systems and human performance tools in steering the work in maintenance

and controlling unwanted performance variability is an important topic for future

research.

4.4. The core task of maintenance

The maintenance function is critical for the safety of any complex sociotechnical system.

Weick et al. (1999, p. 93) have aptly observed that ‘maintenance people come into contact

with the largest number of failures, at earlier stages of development, and have an ongoing

sense of the vulnerabilities in the technology, sloppiness in the operations, gaps in the

procedures, and sequences by which one error triggers another’. Maintenance is a key

function for a resilient organisation to use the term coined by Hollnagel et al. (2006) to

denote the intrinsic ability of an organisation (system) to ‘maintain or regain a dynam-ically stable state, which allows it to continue operations after a major mishap and/or in

the presence of a continuous stress’ (Hollnagel 2006, p. 16). Hollnagel (2006) argues that




the challenges to system safety come from instability which can be a result of, e.g. the

necessary adaptations or adjustments having consequences beyond the local and intended

effects. This is exactly where maintenance comes into play, as an adaptive force as well as

the last line of defence against the unseen effects of organisational instability.

The role of maintenance in increasing the resilience of the sociotechnical system is a

somewhat neglected aspect in research (cf. Reiman 2007). Maintenance is supposed to

have a crucial role in creating safety and resilience in the sociotechnical system as well as

in the negative sense, creating vulnerabilities and having a strong tendency towards

exhibiting performance variability. Maintenance personnel have hands-on experience with

the plant equipment and know its condition best. The maintenance function is able to

perceive new vulnerabilities in their development stages.

In the opinion of the author, being resilient means being aware of the boundaries

of safe activity, being able to recognise how the organisation is currently creating safety,

monitoring whether or not one’s model of safety and hazards is adequate and steering the

organisational processes in the necessary manner. These processes include recovery from

incidents or even accidents. The ability and willingness of the organisation to carry out itscore task is thus at the essence of resilience. Maintenance plays an integral role in this.

The challenge is in further specifying the core task of maintenance and its relation to the

overall task of the organisation, and in defining more specific criteria for a resilient

maintenance organisation. For example, how do the core tasks of aviation maintenance

and NPP maintenance differ, and how are the differences reflected in the maintenance

cultures? How much of this difference is explained by technological differences and how

much by historical reasons or differences in regulations? Research should also tackle the

outside influences on the maintenance task, such as deregulation or price competition.

For example, if a government-owned organisation, e.g. an NPP or a shipping company,

is turned into a private enterprise, does the task of the maintenance function also changeor have some new requirements?

4.5. Evaluation methods for maintenance organisations

The methods for evaluation should provide the organisations with information on how

they currently perceive and respond to their core task. It is not sufficient to rely on general

criteria such as safety attitudes or participative leadership. In addition to these, the task of

the organisation and the constraints and requirements that this task sets for maintenance

have to be taken into account. Organisational evaluation is one way of reflecting on the

ability of the organisation to carry out its task properly. This ability includes that theorganisation is able to monitor its current state, anticipate possible deviations, react to

expected or unexpected perturbations and learn from weak signals and past incidents

(cf. Weick and Sutcliffe 2007, Hollnagel 2009a, Reiman and Oedewald 2009).

Future research should aim at developing methods and approaches for evaluating

the functioning of the maintenance organisations holistically, taking into account the

individual, social and organisational elements. This includes understanding and explaining

the different trade-offs in maintenance work, for example between efficiency and

thoroughness (Hollnagel 2004), occupational safety and system safety (Sanne 2008a),

rule following versus rule bending (Bourrier 1999), certainty and uncertainty (Reiman

2007) and individual and collective action (Pettersen 2008).The development of methods for the evaluation of maintenance organisations is

connected to the debate on the integration of human and organisational aspects into safety

360 T. Reiman



auditing and technical safety assessments (cf. Le Coze 2005, Reiman and Oedewald 2007,

Mohaghegh et al. 2009). Work is needed in defining the significant features of complex

sociotechnical systems in terms of understanding their safety and effectiveness. This

research should draw also on the recent developments in the organisation science

paradigm.

4.6. Leading indicators of the maintenance function

It has been recognised that incident rates and other ‘lagging’ indicators such as personal

injuries do not provide an adequate picture of the ‘health’ of the system in relation to

major accidents (HSE 2006, Herrera et al. 2009). If one wants to evaluate the effectiveness

of development initiatives as well as the functioning of the current maintenance

organisation, reliable indicators are needed. Based on an understanding of the social

context and core task of maintenance, future research should identify the relevant leading

indicators for developing the safety and effectiveness of maintenance activities. These

indicators should be able to direct the development initiatives and provide indications

of whether the organisation is improving or not. These indicators can be used in a

feed-forward manner (cf. Hollnagel 2008) to adjust the functioning of the maintenance

organisation before its performance deteriorates.

5. Conclusions

Maintenance has too often been considered as mostly manual labour requiring little or no

mental work. This correlates also to maintenance quite often being at the bottom of the

hierarchy (in comparison to, e.g. technical support and operations) in terms of respect,influence and authority in organisations (Mercier 1988, p. 14, Perin 2005, p. 75, cf.

Hopkins 2005, p. 85). Mercier (1988, p. 14), for example, argues that NPP maintenance

work suffers from a ‘dirty hands’ image. Perin (2005, p. 262) states that ‘given the

significance of maintenance activities to risk reduction in all high hazard industries, in this

twenty-first century a ‘‘dirty hands’’ image marks a cultural lag of ‘‘gigantic’’ proportions’.

An emphasis on the manual labour requirement of maintenance is prevalent also inside the

maintenance organisations themselves (Reiman 2007).

Research on maintenance has focused mainly on human errors and individual-level

issues, even though social and organisational factors have received increasing attention in

recent years. Still, studies of normal work, practices and cultures of maintenance have been

scarce. Research and development in the maintenance context should acknowledge that

maintenance is a function that not only produces safety, but also gives rise to latent

failures. Maintenance personnel and the activity of the maintenance function can help the

entire organisation to be better aware of the boundaries of safe activity, the condition

of the technical equipment and the effectiveness of current practices and conceptions in

creating safety. In addition to its preparatory and anticipatory role, maintenance plays a

critical role in recovering from expected breakdowns and unexpected system perturba-

tions. A holistic theory on maintenance work is needed to manage the variability and turn

it into a positive force. At its best, maintenance produces positive performance variability

in terms of needed adjustments and adaptations to the condition of the technical

infrastructure and guards against negative instability in terms of equipment perturbations,safety system breakdowns, slow degradation of technical infrastructure and the other

changing vulnerabilities of the technology in use.








Klemola, U.-M. and Norros, L., 1997. Analysis of the clinical behaviour of the anaesthetics:

recognition of uncertainty as a basis for practice. Medical education, 31, 449–456.

Kletz, T., 2003. Still going wrong! Case histories of process plant disasters and how they could have

been avoided . Oxford: Butterworth-Heinemann.

Laakso, K., Pyy, P., and Reiman, L., 1998. Human errors related to maintenance and modifications.

STUK-YTO-TR 139. Helsinki: Edita.Latorella, K.A. and Prabhu, P.V., 2000. A review of human error in aviation maintenance and

inspection. International Journal of Industrial Ergonomics, 26, 133–161.

Lawton, R., 1998. Not working to rule: understanding procedural violations at work. Safety Science,

28, 77–95.

Le Coze, J.-C., 2005. Are organisations too complex to be integrated in technical risk assessment and

current safety auditing? Safety Science, 43, 613–638.

Lind, S., 2008. Types and sources of fatal and severe non-fatal accidents in industrial maintenance.

International Journal of Industrial Ergonomics, 38, 927–933.

Marx, D.A. and Graeber, R.C., 1994. Human error in aircraft maintenance. In: N. Johnston,

N. McDonald, and R. Fuller, eds. Aviation psychology in practice. Aldershot: Ashgate,

87–104.McDonald, N., 2001. Human systems and aircraft maintenance. Air and Space Europe, 3, 221–224.

McDonald, N., 2006. Organisational resilience and industrial risk. In: E. Hollnagel, D.D. Woods,

and N. Leveson, eds. Resilience engineering: concepts and precepts. Aldershot: Ashgate,

205–221.

McDonald, N., et al., 2000. Safety management systems and safety culture in aircraft maintenance

organisations. Safety Science, 34, 151–176.

Mercier, J.-P., 1988. Nuclear power plant maintenance. English adaptation of ‘‘La maintenance des

centrale s nucleaires’’ published for Electricite de France (1987). 2nd ed. Paris: Editions Kirk.

Mohaghegh, Z., Kazemi, R., and Mosleh, A., 2009. Incorporating organizational factors into

Probabilistic Risk Assessment (PRA) of complex socio-technical systems: a hybrid technique

formalization. Reliability Engineering and System Safety, 94, 1000–1018.

Neitzel, R.L., et al., 2008. Exposure to fall hazards and safety climate in the aircraft maintenance

industry. Journal of Safety Research, 39, 391–402.

Norros, L., 2004. Acting under uncertainty. The core-task analysis in ecological study of work.

VTT Publications 546. Espoo: VTT.

OECD/NEA, 2000. Nuclear power plant life management in a changing business world. Workshop

proceedings, 26–27 June 2000 Washington, DC, USA. Issy-les-Moulineaux: OECD Nuclear

Energy Agency.

OECD/NEA, 2001. Assuring future nuclear safety competencies. Specific actions. Issy-les-

Moulineaux: OECD Nuclear Energy Agency.

OECD/NEA, 2002. Regulatory aspects of management of change (NEA/CSNI/R200220). OECD/

CSNI workshop, 10–12 September 2001 Chester, UK. Issy-les-Moulineaux: OECD Nuclear

Energy Agency.

Oedewald, P. and Reiman, T., 2003. Core task modelling in cultural assessment: a case study

in nuclear power plant maintenance. Cognition, Technology and Work, 5, 283–293.

Oedewald, P. and Reiman. T., 2007a. Special characteristics of safety critical organizations. Work

psychological perspective. VTT Publications 633. Espoo: VTT. Available from: http://

www.vtt.fi/inf/pdf/publications/2007/P633.pdf [Accessed 23 March 2010.

Oedewald, P. and Reiman. T., 2007b. Measuring conceptual knowledge among NPP

maintenance personnel – a tool for knowledge management. The proceedings of the joint

8th conference on human factors and power plants and 13th annual meeting of the HPRCT

organization, 26–31 August 2007 Monterey, CA. doi:10.1109/HFPP.2007.4413235.

Orr, J.E., 1996. Talking about machines: an ethnography of a modern job. Ithaca, NY: ILR Press.

Patankar, M.S., 2005a. Ethical challenges in aviation maintenance. In: M.S. Patankar, J.P. Brown,and M.P. Treadwell, eds. Safety ethics. Cases from aviation, healthcare and occupational and

environmental health. Farnham: Ashgate, 23–36.

364 T. Reiman



Patankar, M.S., 2005b. Aviation maintenance environment. The context for ethical decisions.

In: M.S. Patankar, J.P. Brown, and M.P. Treadwell, eds. Safety ethics: cases from aviation,

healthcare and occupational and environmental health. Farnham: Ashgate, 37–55.

Patankar, M.S. and Taylor, J.C., 2004a. Risk management and error reduction in aviation

maintenance. Aldershot: Ashgate.

Patankar, M.S. and Taylor, J.C., 2004b. Applied human factors in aviation maintenance. Aldershot:Ashgate.

Pate ´ -Cornell, M.E., 1993. Learning from the piper alpha accident: a post mortem analysis of

technical and organizational factors. Risk Analysis, 13, 215–232.

Perin, C., 2005. Shouldering risks: the culture of control in the nuclear power industry. New Jersey:

Princeton University Press.

Pettersen, K.A., 2008. The social production of safety: theorising the human role in aircraft line

maintenance. Thesis (PhD) UiS no. 59. University of Stavanger.

Pettersen, K.A. and Aase, K., 2008. Explaining safe work practices in aviation line maintenance .

Safety Science, 46, 510–519.

Pettersen, K.A., McDonald, N., and Engen, O.A., in press. Rethinking the role of social theory

in socio-technical analysis: a critical realist approach to aircraft maintenance. Cognition,Technology and Work. doi:10.1007/s10111-009-0133-8.

Pyy, P., 2001. An analysis of maintenance failures at a nuclear power plant. Reliability Engineering

and System Safety, 72, 293–302.

Ramanujam, R., 2003. The effects of discontinuous change on latent errors in organizations: the

moderating role of risk. Academy of Management Journal , 46, 608–617.

Reason, J., 1990. Human error. Cambridge: Cambridge University Press.

Reason, J., 1997. Managing the risks of organizational accidents. Aldershot: Ashgate.

Reason, J. and Hobbs, A., 2003. Managing maintenance error. A practical guide. Hampshire:

Ashgate.

Reiman, T., 2007. Assessing organizational culture in complex sociotechnical systems – methodological

evidence from studies in nuclear power plant maintenance organizations. VTT Publications 627.Espoo: VTT.

Reiman, T. and Oedewald, P., 2006. Assessing the maintenance unit of a nuclear power plant –

identifying the cultural conceptions concerning the maintenance work and the maintenance

organization. Safety Science, 44, 821–850.

Reiman, T. and Oedewald, P., 2007. Assessment of complex sociotechnical systems – theoretical

issues concerning the use of organizational culture and organizational core task concepts.

Safety Science, 45, 745–768.

Reiman, T. and Oedewald, P., 2009. Evaluating safety critical organizations. Focus on the nuclear

industry, Research Report 2009:12, Swedish Radiation Safety Authority.

Reiman, T., Oedewald, P., and Rollenhagen, C., 2005. Characteristics of organizational culture at

the maintenance units of two Nordic nuclear power plants. Reliability Engineering and System

Safety, 89, 333–347.

Reiman, T., et al ., 2006. Management of change in the nuclear industry. Evidence from maintenance

reorganizations. MainCulture Final Report, NKS-119. Roskilde: Nordic Nuclear Safety

Research.

Roth, E.M., Multer, J., and Raslear, T., 2006. Shared situation awareness as a contributor to high

reliability performance in railroad operations. Organization Studies, 27, 967–987.

Rouse, W.B., 1979. Problem solving performance of maintenance trainees in a fault diagnosis task.

Human Factors, 21, 195–203.

Samurcay, R. and Vidal-Gomel, C., 2002. The contribution of work process knowledge to

competence in electrical maintenance. In: N. Boreham, R. Samurcay, and M. Fischer, eds.

Work process knowledge. London: Routledge, 148–159.

Sandberg, J., 2000. Understanding human competence at work: an interpretative approach.

Academy of Management Journal , 43, 9–25.




Reiman 2011: Understanding maintenance work in safety-critical organisations

Documents