Risk analysis in construction: A paradigm shift from a hard to soft approach

Risk analysis in construction is becoming more popular as competition, project size and complexity increase. Traditional risk analysis relies on rules of thumb. While this approach can work, it is neither robust nor reliable. The development of information technology has made the use of probabilistic estimating and simulation a practicable alternative. These techniques are, however, precise but not necessarily accurate as simplification and many assumptions are used. Further information technology development in expert systems and the paradigm shift in systems engineering from a hard to soft approach has promoted an alternative approach to risk analysis utilizing the theory of fuzzy sets which translates linguistic expressions (such as highly likely) into numerical membership functions. This paper introduces risk analysis in construction against characteristics of the industry. It is argued that normative theories in probability are not as applicable in the construction industry as first perceived. The concept of fuzzy sets is described and the difference in approach in modelling risk between probabilistic and fuzzy methods will be presented. Initial empirical study reveals the applicability of fuzzy sets theory in risk analysis. This is important in that a model of risk analysis can be developed requiring the user to express risk attitude linguistically. Moreover, the results are also in linguistic terms and can, therefore, be understood and interpreted correctly by the user.