Managing the Risks of Organizational Accidents - Case Study Example

Background of the Study Human factors are defined as environmental, organisational, and job factors, and human and individual characteristics which influence behaviour at work in a way which can affect health and safety (The Health and Safety Executive, 2002). A simple way to view human factors is to think about three aspects the job, the individual and the organisation and how they impact on people’s health and safety related behaviour. Over the last 20 years we have learnt much more about the origins of human failure. We can now challenge the commonly held belief that incidents and accidents are the result of human error by a worker in the front line. Attributing incidents to human error has often been seen as a sufficient explanation in itself and something which is beyond the control of managers. This view is no longer acceptable to society as a whole. Organisations must recognise that they need to consider human factors as a distinct element which must be recognised and managed effectively in order to control risk. ( Health and Safety Executive HSG 48 Reducing error and influencing behaviour, 1999) Managers in industry know that accidents cost money. Whether people are injured, plant and machinery damaged or product wasted, organisations lose money. Large scale losses such as those arising from major fires or explosion, or involving loss of life, are very visible and some have been costed on an individual basis. For example the Piper Alpha explosion involved the loss of 167 lives and is estimated to have cost over £2 billion including £746 million in direct insurance payouts (Health and Safety Executive The costs of Accidents at wWork, 2000). Another illustration of major accidents which can be contributed to by human factors is the case of Three Mile Island when serious damage occurred to the core of a nuclear reactor due to operator failure to diagnose a stuck open valve due to poor design of control panel, distraction of 100 alarms activating and inadequate operator training (Health and Safety Executive, HSG48 Reducing aAccidents and iInfluencing Behaviour ,2000). Maintenance failures had occurred before, but no steps had been taken to prevent them recurring. Accidents can occur through peoples involvement with their work. As technical systems have become more reliable, the focus has turned to human causes of accidents. It is estimated that up to 80% of accidents may be attributed, at least in part, to the actions or omissions of people (Health and Safety Executive HSG48 Reducing aAccidents and Influencing Behaviour 2000). In fact, the onset of industrial research focusing on causal explanations for errors has been triggered by high accident rates in the manufacturing industry (Hale & Glendon, 1987). These researches presented various theories on accident causation, as well as traits that distinguished the injured (Hale & Glendon, 1987). The strong thrust on the investigation of ‘accident proneness’ has been sustained for several decades until methodological issues have been raised (Hale & Glendon, 1987; Weindling, 1985). Standardised assessment procedures have been designed to measure factors such as accident proneness, intelligence, manual dexterity, and job-fit (Weindling, 1985). With the ultimate goal of improving human performance, these assessments proliferated along with those presented by Taylor and Ford, among others (in Weindling, 1985). Two voluntarist approaches on health and safety management started to take shape, at time claiming exclusivity of solutions to the issue. The behavioral approach’s main point expresses the limitations of employees in modifying the variables that affect injury, accidents, weakness, and disease (Bohle, 1993). While the legislated-engineering approach emphasised hazards and unsafe conditions, the behavioral approach recognised the relevance of unsafe acts (Jones, 1985). Integrating these two paradigms, Rasmussen (1987) evaluates the root of untoward incidents to the human-system mismatch. In essence, errors may not be exclusively explained, defined and prevented with either behavioral or legislated-engineering approaches (Rasmussen, 1987). The current study is conducted to determine the perceptions of employees on _________’s safety systems, and recommend apt action plans for management on the basis of the study’s outcomes. Using Reason’s (in Stanhope, 1997) accident causation model, the organisation, task / environment, and individual characteristics may be assessed as factors composing safety systems. Review of Related Literature Data from the 1999 Health Survey for England suggest that men are at greater risk than women of having a major accident involving sport, a moving vehicle, tools/ implements or work. There is a specifically significant gender difference for work accidents - in 1999 there were 10 major accidents per 100 men in work in England, compared with a rate of just three per 100 women in work (http://www.statistics.gov.uk, retrieved September 10, 2005). The Health and Safety Statistics report for 2000-2001 states that the incident rate for non-fatal injuries among men aged 25-34 years is 875.1 (http: www.irata.org, retrieved September 10, 2005). Because accident rates have been increasing, human error analysis has been an area of growing emphasis from researchers. Human error analysis has received significant attention in safety critical industries, making noteworthy progress into identifying causes and measures for prevention. Such analysis have brought to the fore the notable frameworks of Rasmussen et al. (1987), Reason (1990), Hollnagel (1991) and Hale and Glendon (1997). They have presented systematic analysis of different forms or types of human errors and their corresponding contributory factors. Apart from individual human malfunction, organisational and managerial factors are evaluated as well representing more holistic, integrative approach to the study of accident causation. The Role of Human Factors in Socio-technological Systems Technological systems may be designed to cater to the limitations of human nature at all equally aspects of physical, psychological, team, organisational, and political (Vincente, in press). Error Reporting Biases The definition and components of an error has been an issue of strong contention among accident analysts. While this is the case, Reason (1997) cites the role of ‘violations’ rather than error, as causes of accidents (Reason, 1997). For instance, precautionary safety measures may be tedious or awkward, with individuals constantly resorting to omitting them to cater to more urgent task demands. In these cases, these are not to be considered errors because of a lack on intention on the part of the individual (Reason, 1997). Other plausible reasons for not reporting such incidents are fear of blame or reprisal, and lack of belief in or credence of the exercise (Reason, 1990). Moreover, accidents may be considered as mere task characteristics – something inevitable or bound to happen (Reason, 1990). With such a mindset, error detection and recovery tend to be neglected. Chappell (1996) also suggests that individuals my not be aware of implicit system faults as accident precursors. For instance, getting used to second-rate equipment or inadequate supervision are not reported as causes of accidents because the individuals have been accustomed to them (Chappell, 1996). Knowledge acquisition methods and direct observation methodologies have been acknowledged as measures for addressing accident reporting biases (Jordan, 1993; Woods, 1994). The above mentioned biases are those observed in incidents that transpire unreported; biases also exist in what is reported (Busse & Johnson,1998). The absolute incidence of accidents cannot be exactly determined from data from incident reports because reporters may not be the ones directly involved, or they may not be representative of the sample at hand; these are considerations that may possibly confound incident reports (Busse & Johnson, 1998). The Criticality of Incident / Accident Detection Specifying the factors that influence incident detection is deemed more effective than accident prevention or avoidance (Rasmussen & Vincente, 1989; Reason, 1990). Meanwhile, even if these factors are identified, they are not critically analysed. As suggested by Busse and Johnson (1998), such an analysis should encompass system factors, and cognitive aspects of the task and work environment. The parallelism between accidents (adverse events) and incidents (near miss occurrences) is often depicted through an iceberg metaphor (Van der Schaff et al., 1991; Battles et al., 1998). The implicit assumption is that such incidents and accidents are traced back to the same root causes. Accident causation research has emphasised on the categorisation and analysis of human error, featuring taxonomies that include behavioral categorisations (Hollnagel, 1991) to those underpinning cognitive processes (Reason, 1990). These all embody a critical first step in the analysis of incidents and accidents; however Rasmussen et al. (1997) states that behavioral classifications are not useful in expounding on the primary mechanisms that have caused these errors. The rationale behind identifying the underlying cognitive mechanisms in error causation is to customise remedial measures and error management methodologies (Maddox & Reason, 1996). Rasmussen’s Skill, Rule, Knowledge (SRK) Framework Rasmussen (1987) advocates that accidents are rooted on human-system mismatch, and presents this through a multifaceted framework, including factors affecting performance (subjective goals and intentions, mental load, resources); causes of human malfunction (external events such as distraction, task demand, operator’s being incapacitated, intrinsic human variability); situation factors (task characteristics, physical environment, work time characteristics); and personnel tasks (equipment design, procedure design, fabrication, installation, inspection, operation, test and calibration, maintenance and repair, logistics, administration, management), as critical influences to mechanisms of malfunction – that in turn determine both internal and external modes of malfunction (Rasmussen, 1982). Rasmussen’s Mechanisms of Malfunction According to Rasmussen’s (1982) human malfunction taxonomy, factors affecting performance, causes of human malfunction, and situation factors, in combination, determine the following mechanisms of malfunction: discrimination, input information processing, recall, inference, and physical coordination. Whereas, these mechanisms of malfunction, together with personnel tasks, influence internal human malfunction: detection, identification, decision, and action. Personnel tasks skill and internal human malfunction exert a pooled influence on external mode of malfunction, as manifested in failure to perform specified task, omission of act, inaccurate performance, wrong timing, commission of erroneous act, commission of extraneous act, or sneak path (Rasmussen, 1982). Reason’s Model of Accident Causation Reason (1990) states that an error must not be tagged as an accident or failure per se, but rather, as precedent or precursor to failure. Error may thus lead to active failure or latent failure (Reason, 1990). He relates active failure to the performance of front-line operators such as pilots and control room crew; it has an instantaneous and direct impact on the system. Whereas, latent failure is traced on back-end role players – designers, high-level decision makers, managers, among others. When compared with Rasmussen’s taxonomy, latent failure is reflected in the causes of human malfunction, situation factors, factors affecting performance and personnel tasks (Busse & Johnson, 1998). In combination with lapses or limitations in the human cognitive processing system, these lead to external modes of malfunction, that is, to an accident or incident (Reason, 1990). In Reason’s (1990) framework, the final event from either active or latent failure pathways need not necessarily lead to an accident. The accident path may stop at any level of the causation model, whose effects need not infiltrate all defense levels to cause maximal damage (Maurino, Reason, Johnston, & Lee, 1998). Nevertheless, it is imperative that these pathways, including their points of origin and termination, need to be examined for accident prevention in future, similar trajectories (Maurino et al., 1998). Simon’s Model of Decision Making ‘Descriptive’ and ‘normative’ have almost become synonymous in decision making literature. Descriptive theories have then began to introduce normative axioms, as in the Advantage Model (Shafir, Osherson & Smith, 1993). On the other hand, normative theories have also sought to fine-tune their jargon so that they are more exacting and accurate in describing decision making e.g. Prospect Theory (Kahneman & Tversky, 1979). Subjective Expected Utility (Von Neuman & Morgenstein, 1947), among others. A third model has recently been promoted – prescriptive theories – which presents what an individual can and ought to use, given his specific needs and circumstances. Simon (1960) advocates a triphasic model of decision making, composed of intelligence, design, and choice. Intelligence deals with determining the need for a decision. As Simon (1960) expresses, “searching the environment”. Once the need for a decision is determined, the design phase begins, involving the study of the scope of the problem and of alternatives. The last stage of Choice pertains to choosing the best alternative or course of action. Each phase is complex; with each representing an intricate decision making process (Simon, 1957). One of the central tenets of descriptive decision making theories is the concept of Bounded Rationality, stating that all deliberate rational behavior occurs within limits, including those which are cognitive in nature (Schoemaker, 1980). A rational decision maker is one who “has a well-organised and stable system of preferences and a skill in computation that enables him to calculate, for the alternative courses of action that are available to him, which of these will permit him to reach the highest attainable point on his preference scale” (Simon, 1955). He first cites human physiological and psychological constraints on his work on rational choice. He specifies, “the maximum speed at which an organism can move establishes a boundary on the set of its available behavior alternatives”; in simple terms, he suggests that humans have limits which they cannot surpass. Holistic and Non-holistic Descriptive Modes of Decision Making Perhaps, one of the earliest descriptive theories of decision making is the Satisficing model, which has surfaced at about the same time as the idea of Bounded Rationality. Simon (1957) expresses that decision makers select an alternative that surpasses some benchmark or standard. He further posits that decision makers, in most instances, do not and cannot maximise, that is, satisficing rather than optimising. Yet another descriptive decision making theory has been proposed by Cohen et al. (Cohen, March, & Olsen, 1972), the Garbage Can Model. Decision situations, described as organised anarchies, are characterised by the following traits: 1) problematic preferences, 2) ambiguous technology, and 3) dynamic or fluid participation (Cohen et al, 1972). This depicts the organisation as an unstructured, ambivalent group of ideas, distinguished by ‘hit or miss’ procedures and a ‘learn from mistakes’ philosophy. The Garbage Can model is distinct from most other descriptive theories; it states that organised anarchies do not necessarily seek the most appropriate alternative: “To understand processes within an organisation, one can view a choice opportunity as garbage can into which various kinds of problems and solutions are dumped by participants as they are generated” (Cohen, et al., 1972). A more recent theory has been authored by Beach and Mitchell, whose foundations are the Lexicographic Model (Beach & Mitchell, 1978) and the Strategy Selection Model. It represents an integration of the current ideas on realistic decision making scenarios. It discusses progress decisions, that is, whether past decisions are being pursued, and adoption decisions, which make adjustments on previously made decisions that are more attainable and realistic. The cognitive/ disjunctive model of Einhorn (1970) explains that decisions are arrived at through a combination of data or information, as discussed by several of its proponents (Coombs & Kao, 1955; Dawes, 1964; Einhorn, 1970). The model seeks to identify the best solution or set of solutions from a universe of options; all those which have gone beyond some ‘aspiration level’ or threshold forms part of this group of solutions; those which fall below this standard are then excluded. The Disjunctive Model gauges alternatives based on its best trait, rather than on all its attributes. Its similarity to Simon’s (1955) Satisficing Model is noted, in that it seeks to identify the most appropriate solutions, not necessarily the best or most optimal. The Lexicographic model of decision making assumes that a decision maker ought to know the attributes that make up his options; with these, he then ranks them according to perceived importance (Tversky, 1969). Each pair of options is then compared in terms of these traits, until one is weighted more valuable than the other (Tversky, 1969). Trevsky (1972), on the other hand, authors a probabilistic model of choice, tagged the Elimination by Aspects (EBA) model, which also utilises intradimensions in evaluation of alternatives (Hogarth, 1980; Payne, Bettman, & Johnson, 1993). The primary activity is concealed elimination process, where each alternative is assessed as a group of attributes, which are then successively appraised. Modern Decision Making Theories A more modern theory is Klein’s (1989) Recognition Primed Decisions (RPD) model, which seeks to expound on decision making in ‘natural settings’, applicable to both organisational and real life scenarios. The model consists of four primary components: identifying cases as typical, situational understanding, serial evaluation, and mental stimulation. These components are conventionally engaged in a sequential manner, along with reviewing previously made decisions, and re-evaluating decisions and their probable outcomes (Klein, 1989). Two other models are noteworthy: The Additive and Additive Differences Models. The Additive Model independently and separately assesses multidimensional options, and comes up with forced decision as regards its value before proceeding to the next pairwise comparison. The approach is deemed more holistic; the procedure is carried out until the best option is selected. The Additive Difference Model (Tversky, 1969) is grounded on “comparisons of component-wise differences between the alternatives”, with only two options being compared at a time. Each comparison is then multiplied by the weight assigned for that component; the final decision is based on the summation of products for the alternative on several dimensions. Different people will react in different ways to the same situation; individual, characteristics (such as attitudes, abilities, and personality) will interact with features of the job and organisation, e.g job characteristics, work environment, and health and safety culture to affect a persons sense of well being and job satisfaction (Reducing error and influencing behaviour, 1999). In summary, descriptive theories describe the decision making process in terms of situation comparisons and non-situation comparisons. Situation comparisons consist of 1) comparing between situations, and 2) comparing between situations and alternatives. Meanwhile, non-situation comparisons are made up of 1) comparisons between an alternative, attribute, and some standard, 2) a comparison between alternatives, and 3) a comparison between attributes. References “Accident rate statistics: 2000 and 2001.” Retrieved on September 10, 2005 from http:// www.irata.org/uploads/healthandsafety/ “Annual major accident rates: by selected type of accident and gender, 1999: social focus on men.” Retrieved on September 10, 2005, from http://www.statistics.gov.uk/STATBASE/ssdataset.asp?vlnk=4449 “The Changing Picture of Assessment.” Retrieved September 13, 2005 from the eduplace.com website: http://www.eduplace.com/rdg/res/litass/chang.html Battles, J.B., et al. (1998). The attributes of medical event reporting systems.” Archives of Pathology and Laboratory Medicine Online, 122. Beach, L.R. & Mitchell, T.R. (1978). A contingency model for the selection of decision strategies. Academy of Management Review, 3, 439–449. Bohle, P. (1993). Work psychology and the management of occupational health and safety: a historical overview. In Quinian, M. (ed.). Work and health: the origins, management and regulation of occupational illness. Melbourne: Macmillan. Busse, D. & Johnson C. (1998). Modeling human error with a cognitive theoretical framework. ECCM ’98. Nottingham: Nottingham University Press. Chappell, S.L. (1996). Using voluntary incident reports for human factors evaluations. The International Society of Air Safety Investigators. Cohen M.D., March, J.P. & Olsen, J.P. (1972). A garbage can model of organizational choice. Administrative Science Quarterly, 17, 1–25. Coombs, C.H. & Kao, R.C. (1955). Nonmetric factor analysis. Research Bulletin, 38, Engineering Research Institute, University of Michigan. “The costs to Britain of workplace accidents and work-related ill health in 1995/96.” ISBN 0717617092, UK: HSE. Dawes, R.M. (1964). Social selection based on multidimensional criteria. Journal of Abnormal and Social Psychology, 68, 104–109. Einhorn, H.J. (1970). The use of nonlinear, noncompensatory models in decision making. British Psychological Society Bulletin, 73 (3), 221–230. Hogarth, R.M. (1980). Judgment and choice: The psychology of decision. Chichester: Wiley. Hale, A. & Glendon, A. (1987). Individual behavior and the control of danger. Amsterdam: Elsevier. Hollnagel, E. (1991). The phenotype of erroneous actions: implications for HCIO design. In Weir, G. and Alty, J. (Eds.). Human-computer interaction and complex systems. London: Academic Press Ltd. Jones, H. (1985). An inspector calls: health and safety at work in inter-war Britain. In Weindling, P. (ed.). The social history of occupational health. London: Croom Helm. Jordan, B. (1993). Ethnographic workplace studies and computer supported cooperative work. Interdisciplinary workshop on informatics and psychology. Schaerding, Austria. Kahneman, D. & Tversky, A. (1979). Prospect theory. Econometrica, 47, 263–291. Klein, G.A. (1989). Recognition-primed decisions. In Rouse, W.B. (Ed.). Advances in man-machine research, 5, 47–92. Greenwish, CT: JAI Press. Levi, I. & Levi, S. (2000). Research on work-related stress. European Agency for Safety and Health at Work. Maddox, M.E. & Reason, J. (1996). Human factors guide for aviation maintenance. Washington, DC: Office of Aviation Medicine. Maurino, D., Reason, J., Johnston, N., & Lee, R. (1997). Beyond Aviation Human Factors. Brookfield, VT: Ashgate Publishing Company. Millanvoye, M. (1998). Ageing of the organism before sixty years of age. In Marquie J., Cau-Bareille D., & Volkoff, S. (Eds.) Working with age. London: Taylor and Francis. Payne, J.W., Bettman, J.R., & Johnson, E.J. (1993). The adaptive decision maker. Australia: Cambridge University Press. “Productive and safe workplaces for an ageing workforce.” Retrieved September 10, 2005 from http://www.apsc.gov.au/publications03/maturecomcare4.htm Rasmussen, J. & Vincente, K. (1989). Coping with human errors through system design: implications for ecological interface design. International Journal of Man-Machine Studies, 31, 517–534. Rasmussen, J. (1982). Human error: a taxonomy for describing human malfunction in industrial installations. Journal of Occupational Accidents, 4, 311–333. Rasmussen, J. (1987). The definition of human error and a taxonomy for technical system design. USA: John Wiley and Sons Ltd. Reason, J. (1990). Human error. London: Cambridge University Press. Reason, J. (1997). Managing the risks of organizational accidents. London: Ashgate Publishing Limited. “Reducing error and influencing behavior.” (1999). ISBN 0717624528.UK: Health and Safety Executive. Rogers, Don. (2005). Preparing for impact of top 10 megatrends. Retrieved from the EMA Inc. website: http://www.ema-inc.com/www/www_future.htm. Reducing Error and Influencing Behaviour H.S.E. Books HSG48. Schoemaker, P.J.H. (1980). Experiments on decisions under risk: The expected utility theorem. Boston: Martinus Nijhoff Publishing. Shafir, E.B., Osherson, D.N., & Smith, E.E. (1993). The advantage model: a comparative theory of evaluation. Organizational Behavior and Human Performance, 55(3), 325-378. Simon, H.A. (1955). A behavioral model of rational choice. Quarterly Journal of Economics, 69, 99–115. Simon, H.A. (1957). Rationale choice and the structure of the environment. In Simon, H.A. (Ed.). Models of man. New York: John Wiley. Simon, H.A. (1960). The new science of management decision. New Jersey: Prentice Hall. Spirduso, N. (1995). Physical dimensions of ageing. Human Kinetics Campaign. Stanhope, N., et al. (1997). Applying human factors methods to clinical risk management. British Journal of Obstetrics and Gynaecology, 104, 1125–1232. Stubbs, N., Fernandez, J., & Glenn, W. (1993). Normative data on joint ranges of motion of 25 to 54 year old males. International Journal of Industrial Ergonomics, 12, 265–272. The Cost of accidents at work H.S.E. Books. The Free Dictionary by Farlex. Retrieved September 13, 2005 from the freedictionary.com website: http://www.thefreedictionary.com/enthusiasm The Houghton Mifflin Company Dictionary. Retrieved September 13, 2005 from the Answers.com website: http://www.answers.com/topic/training The Institute of Nuclear Power Operations. (2002). Human performance fundamentals course reference. Tversky, A. (1969). Intrasensitivity of preferences. Psychological Review, 76 (1), 31–48. Tversky, A. (1972). Choice by elimination. Journal of Mathematical Psychology, 9, 341–367. Van der Schaaf, T.W., et al. (1991). Near miss reporting as a safety tool. Butterworth-Heinemann Ltd. Vincente, K.J. (In press). The human factor: Revolutionizing the way we live technology. Toronto: Knopf Canada. Von Neuman, J. & Morgenstein, O. (1947). Theory of games and ergonomic behavior, 2nd ed. Princeton, New Jersey: Princetown University Press. Weindling, P. (1985). Linking self help and medical science: The social history of occupational health. In Weindling, P. (ed.). The social history of occupational health. London: Croom Helm. Woods, D.D. (1994). The critical incident technique – 40 years later. The Human Factors and Ergonomics Society Annual Meeting. Read More