Accident investigation models

The data in this paper show the total number of coal mine accidents happening in China from January 2006 to December 2010. (The data are from The report of national coal mine accidents analysis in 2010 published by the Department of Accident Investigation at State Administration of Coal Mine Safety). The data processed to model are from January 2006 to December 2009, and the data of 2010 aim to be compared with the forecasted data, see table 1. 

Part of the problem stems from the use of the chain-of-events model in accident investigation because it is difficult to find an event preceding and causal to the operator behavior, as mentioned earlier. If the problem is in the system design, there is no proximal event to explain the error, only a flawed decision during system design. 

If STAMP has been used as the basis for previous safety activities, such as the original engineering process or the investigation and analysis of previous incidents and accidents, a model of the safety-control structure may already exist. If not, it must be created although it can be reused in the future. Chapters 12 and 13 provide information about the design of safety-control structures. 

To its credit, the accident investigation board went beyond the usual chain-of-events model and instead interpreted the acddent in terms of a complex and flawed process ... 

The book is divided into three sections.The first part explains why a new approach is needed, including the limitations of traditional accident models, the goals for a new model, and the fundamental ideas in system theory upon which the new model is based. The second part presents the new, extended causality model. Ihe final part shows how the new model can be used to create new techniques for system safety engineering, including accident investigation and analysis, hazard analysis, design for safety, operations, and management. 

The number of conceptual accident models that drive government accident investigation programs seems unnecessarily diverse. Since they conflict, all models can not be valid [p. 124],... 

MORT Safety Assurance Systems by WilMam G. Johnson This text serves well both for incident causation model building and for incident investigation. The accident investigation chapter states that while accident investigation has always been a major element in safety, pre-accident hazard analysis is preferable (p. 347). 

The discussion suggested that the current discussion about epidemiological and systemic models might not be the reason why organizational factors are underrepresented. The setup and the mandate of maritime accident investigation may cause the focus of unsafe acts and preconditions for rmsafe acts only. 

Detailed investigation of accident scenarios modeled in Czech PSA projects has led to the conclusion that, at least for major subset of aU actions performed in response to initiating event occurrence and driven by symptom based procedures, lack of time should not be relevant issue. Thus, the time-versus-reliabiUty curves (for short time windows) are used just in some few very special cases of potential lack of time (with time windows shorter than 30 minutes), mostly connected with necessity to recover plant critical safety functions. 

Another example for an in-depth accident data base is the Pedestrian Crash Data Study (PCDS) from the US [29] (which is also described in Sect. 5.2.1) or accident investigations carried out by vehicle manufacturers. The latter ones have a very high level of detail but suffer even more from biases due to low case numbers, model selection criteria or geographic effects [16]. 

In accident reconstruction, someone first collects data at the accident scene. There are records of each type and model of vehicle There are photos and records of other data about the damage to each vehicle. There are records or plots of skid marks, impact points, final locations of vehicles, and vehicle components. There are records of pavements and surfaces involved in the accident. Investigators can log many of these data items directly on portable computers and devices. Often police record these data as part of on-scene reports. 

The theory of multicausation is that the contributing causes combine together in a random fashion to result in an accident. During accident investigations, there is a need to identify as many of these causes as possible. In reality, the accident model is an amalgam of both the domino and multicausality theories. 

Fault tree analysis is an analytical technique that is used to trace the chronological progression of factors (events) contributing to the accident situation, and is useful in accident investigation and as a predictive, quantitative model in risk assessment. Again,the principle of multicausality is utilised in this type of analysis. (A fuller treatment on fault tree analysis is given at section 10.6). 

An analysis of the incident by the Bureau of Air Safety Investigation (BASI, 1993) draws specifically on the accident analysis model developed by James Reasons in which he distinguishes between active and latent factors, which correspond broadly to the victim- and system-blaming explanations discussed above. (The following quotations from Reasons are found in the BASI report, p 31.)... 

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. 

Butgoyne, J.H. (1993). Reflections on Accident Investigation. Safety Sci., 16, 401-6. Cone, J.E., Makofsky, D. and Daponte, A. (1990). Fatal Occupational Injitry and Energy Exchange A Rephcation of Kriebers Model. Journal of Occupational Accidents, 12, 187. 

Included in these references are some which discuss accident causation and prevention generalfy. Purswell and Rumar and Hoyos and Zimolong discuss types of accident models while Benner evaluates the usefulness of different models and the use made of them by accident investigation agencies. 

---