Causation models importance

The use of event-chain models of causation has important implications for the way engineers design for safety. If an accident is caused by a chain of events, then the most obvious preventive measure is to break the chain before the loss occurs. Because the most common events considered in these models are component failures, preventive measures tend to be focused on preventing failure events— increasing component integrity or introducing redundancy to reduce the likelihood of the event occurring. If corrosion can be prevented in the tank rupture accident, for example, then the tank rupture is averted. 

In the past it has been irsual for the safety practitioner to concentrate much effort on the hazard and the methods by which it can be removed or controlled. This approach has been very sirccessfirlfy applied in industry, but the irrjrrry causation model now provides the safety practitioner with another new field of investigation in determining the reasons for human error. The human error is not considered to have arty element of blame. What is important is to determine the cause of the human error and to identify the stimtrlus that produced the behaviour pattern which resulted in the error. 

An important part of aity safety strategy is the provision of an effective accident investigation methodology. For accident investigation to be meaningful, selecting the appropriate accident causation model is essential. 

We also know that incentives and motivational factors play an important role in causal attributions and the selection of remedies. An illustration of this is when decision-makers attribute accidents to unstable causes rather than stable causes. In the ILCI causation model, we find examples of both types of causes. Mental stress is an example of an unstable factor as opposed to inadequate equipment, which is a stable characteristic. Unstable causes produce less certain predictions about the efficiency of remedial actions. By selecting such causes, the decision-maker may escape from the obligation to decide on binding and resource-demanding solutions to the safety problem (Kjellen, 1993). 

In this chapter a simple model of incident causation is presented and the relative importance of three groups of factors contributing to industrial safety will be discussed Technical, Organisational and Behavioural Factors. Historic trends or fashions focussing on one of these thpee factors will be described, followed by recent results of the situation in the chemical process industry in the Netherlands. 

Experimental design is a large topic and we can only mention several of the important issues here. To keep this discussion focused on parameter estimation for reactor models, we must assume the. reader has had exposure to a course in basic statistics [4]. We assume the reader understands the source of experimental error or noise, and knows the difference between correlation and causation. The process of estimating parameters in reactor models is part of the classic, iterative scientific method hypothesize, collect experimental data, compare data and model predictions, modify hypothesis, and repeat. The goal of experimental design is to make this iterative learning process efficient. 

Figure 10.1 highlights the luck factor that determines the outcome of an undesired event, the important role that near-miss incidents play in loss prevention, and the importance of the accident root causes and their antecedents. This model will help managers look at the bigger picture of accidental loss—the causes, effects, and luck factors—and explains the importance of risk identification and mitigation in the safety process. Similar models on loss causation can be referred to as long as they focus on the bigger picture and do not indicate unsafe employee actions as the main accident causes. 

---