SYSTEMS METHODS AND THEIR APPLICABILITY TO ACCIDENT COSTING

Willemstyn, Claire Julie (1986). SYSTEMS METHODS AND THEIR APPLICABILITY TO ACCIDENT COSTING. BPhil thesis The Open University.

DOI: https://doi.org/10.21954/ou.ro.0000f90e

Abstract

This thesis concerns the subject and problem of accident costing, and an examination of the value and applicability of a holistic, systems approach to such issues. The initial impetus for the study arose through discussions between members of the Open University Systems Group and the Electricity Council Safety Branch. In considering particular classes of accidents that occur in the Electricity Supply Industry, it was realised that the Industry had no idea of the scale of monetary costs involved, and no mechanism for evaluating them. Thus the need to devise a costing framework was identified.

To meet the Industry’s need the problem was treated as a straight forward costing exercise. Following a study of the literature on valuing life and costing accidents (detailed in Section 1), and a period of field research at two Electricity Supply Industry Area Board districts, a costing framework, or model, was developed of areas of potential post factum costs for the Industry as a result of an accident (Section II). From the model, costs to the Industry both direct and indirect, were calculated, thereby illustrating areas of actual and potential high cost. Using the Industry’s analysis of accidents in terms of ’causation’, the class of accidents of major concern is that to which ‘Electricity' is stated as being causal. Although these accidents only form between four to seven percent of the total annually, their significance lies in the high proportion resulting in serious injuries and high costs.

The method adopted provided the Electricity Supply industry with a costing model, but was restricted by this desired result. Such a relatively narrow reductionist approach tends to rely on limiting assumptions and to ignore wider issues. In Section III of the thesis, therefore the problem is reexamined using two different systems methodologies in preliminary stages. Firstly, the Checkland methodology, a 'soft' systems methodology and secondly the de Neufville and Stafford methodology, a ‘harder' methodology. These methods lead to different conclusions and wider implications for implementation strategies. However, both the Hard and Soft Systems methodology used tended to broaden the understanding of some aspects of the problem whilst constraining others. Thus, the author of the thesis, together with two other members of the Systems Group developed the Hard Systems method in an attempt to bridge an apparent methodological gap. This method and other conclusions are to be found in the final (fourth) section.