A case-based reasoning approach for selecting risk management techniques

The management of risk in the built environment is critical and there is a wide range of techniques available to deal with the task. However, research has shown that only a small number of techniques are used by practitioners (Akintoye and MacLeod 1997, Wood and Ellis 2003). One reason cited for this is a lack of knowledge of the circumstances in which they can be used. The aim of the research in this paper is to produce a case-based reasoning (CBR) tool for decision support in selecting appropriate techniques for built environment problems. The tool uses a case-base developed from historic problems in the literature. A problem framework is used to characterise problems as 1) External or internal – described by the PESTLE Model, 2) The Risk Owner and 3) The Project Phase. A second stage defines the data used in a problem as fuzzy, incomplete or random. The methodology of CBR is heuristic and as such this work has investigated the effect of differing the retrieval mechanism, the inclusion of weights and the threshold value. The results demonstrated a tool which during validation predicted the correct technique up to 93% of the time; additionally it was seen that more complex CBR methods did not result in more accurate prediction rates. Overall the research has produced a simple tool to select appropriate risk management techniques and demonstrated the applicability of CBR to the problem – highlighting that the simplest methodology has proved the most effective.