Ch 10. Accident Analysis and Prevention ISE 412 Human Factors Engineering ISE

Top Causes of Death – US and Global 1. Heart disease 2. Cancer 3. Stroke 4. Accidents 5. Respiratory illnesses 6. Alzheimer’s disease Types of Accident Deaths (USA) Traffic -- 43% Home -- 20% Public situations -- 20% Occupational (incl. transport work) 16% These 4 are partially stress-related ISE

Distribution of occupational accident types Much of our understanding of accident analysis and prevention can be drawn from occupational accidents. Although there are differences in situations, there is generalizability to the other settings. Occupational environments are easier to study for some types of questions. Based on US insurance data from 2005 ISE

Accident Proneness Concept of accident proneness: “90% of the accidents are incurred by 10% of the people.” Such distributions can occur by chance, however, with a poisson distribution! Psychologists tried using personality and life-stress variables. Low correlations with individual behavior. Lifestyle, spillover are other explanations. However: “Lifestyle” correlations with occupational accidents are very low; poor prognosis for personnel selection. Work-home spillover -- more often in the direction of work spilling over to home or traffic, rather than the other way around. When a system is properly engineered for ergonomics, individual differences in personality should disappear. ISE

Risk Models Single cause model Multiple single causes Chain of events Multiple chain of events Factorial models Fault trees Flow charts and Petri nets Simulations Cusp catastrophe model Includes psychosocial and hazard variables Swiss Cheese model Resilience less analytic – considered here as part of safety climate and culture constructs ISE

Fault Tree Analysis (FTA) Fault Tree: A graphic “model” of the pathways within a system that can lead to a foreseeable, undesirable loss event. The pathways interconnect contributory events and conditions, using standard logic symbols. Numerical probabilities of occurrence can be entered and propagated through the model to evaluate probability of the foreseeable, undesirable event. ISE

FTA is best applied to cases with … Large, perceived threats of loss, i.e., high risk. Numerous potential contributors to a mishap. Complex or multi-element systems/processes. Already-identified undesirable events (a must!) Indiscernible mishap causes (i.e., autopsies.) Caveat: Large fault trees are resource-hungry and should not be undertaken without reasonable assurance of need. ISE

FTA produces: Graphic display of chains of events/conditions leading to the loss event. Identification of those potential contributors to failure that are “critical.” Improved understanding of system characteristics. Qualitative/quantitative insight into probability of the loss event selected for analysis. Identification of resources committed to preventing failure. Guidance for redeploying resources to optimize control of risk. Documentation of analytical results. ISE

Some definitions FAULT An abnormal undesirable state of a system or a system element induced 1) by presence of an improper command or absence of a proper one, or 2) by a failure (see below). All failures cause faults; not all faults are caused by failures. A system which has been shut down by safety features has not faulted. FAILURE Loss, by a system or system element, of functional integrity to perform as intended, e.g., relay contacts corrode and will not pass rated current closed, or the relay coil has burned out and will not close the contacts when commanded – the relay has failed; a pressure vessel bursts – the vessel fails. A protective device which functions as intended has not failed, e.g, a blown fuse. ISE

Assumptions and limitations Non-repairable system. No sabotage. Markov… Fault rates are constant. The future is independent of the past – i.e., future states available to the system depend only upon its present state and pathways now available to it, not upon how it got where it is. Bernoulli… Each system element analyzed has two, mutually exclusive states. ISE

The logic symbols (see also fig. 8.17, pg. 351) Most Fault Tree Analyses can be carried out using only these four symbols. Events and Gates are not component parts of the system being analyzed. They are symbols representing the logic of the analysis. TOP event – aka, FAULT event ISE

Steps in FTA Identify undesirable TOP event. 1 Identify first level contributors. 2 Link contributors to TOP by logic gates. 3 Identify 2 nd level contributors. 4 Link contributors to events by logic gates. 5 Repeat/continue. 6 ISE

Use FTA to … Identify probability of failures and faults. Identify candidates for engineering solutions. Identify common cause events … Events which, if they occur, will cause two or more fault tree events to occur. Typical common cause events include power failures, dust & grit, temperature effects (freezing/overheating), operator oversight, etc. Can be overcome through redundant systems, isolation or shielding, etc. ISE

Safety management & climate It should address all parts of the system ISE

Top-down Authoritarian vs. humanized management Work speed & profits vs. safety Keep track of automated processes Choice of accident prevention programs Efficacy of program implementations Maintenance policies Maintenance efficacy ISE

Bottom-up Perceptions of management concern Participation in safety functions Mutual monitoring of co-workers’ behavior Belief in controllability Supervisors’ autonomy Measured by questionnaire Meta-analysis showed positive and negative relationships with actual accident outcomes. ISE

Swiss Cheese Model System defenses all leak a little. When a risk trajectory gets through all the leaks, something bad happens ISE

For example … What you see is not always what you get. Change a parameter on a computer screen, but did the physical objects behave as intended? ISE

Resilience Engineering Move from explaining how accidents happened to anticipating them Stress-demand resilience function ISE

Calibrate risks Include surprise favorable events in the range of possibilities. ISE

Cusp model for safety climate, anxiety, and accidents Meta analysis showed safety climate can have positive and negative correlations with accidents. Here: Low safety climate could be dangerous in some respects, but could promote individual vigilance over risks. High anxiety: could induce errors, or increase vigilance over risks. ISE

Correspondence between the cusp model and resilience ISE

Intervention Effect Size Personnel selection 4.8% Technological interventions 54.4% Behavior modification 53.1% Poster campaigns -1.0% Installing safety committees 33.7% Medical or health mgmt 39.8% Near miss accident reporting 0.0% Comprehensive ergonomics 53.1% Other management interventions 55.0% Governmental interventions 9.7% See also: table 10.3, pg. 251 ISE