Risk, complex decision-making and the application of multiattribute utility analysis in the construction industry

The subject of this dissertation is the concept of risk as it relates to the Construction Industry. Its prominence in the Industry is documented and generic forms are identified. Conventional risk protection mechanisms are reviewed and found to be primarily defensive reactions which competitively shackle the Contractor. Risk and its counterpart, opportunity, are seen to be subjectively perceived concepts, which render objective measurement techniques, such as the standard deviation, to be of secondary value. Recognizing risk's intractable relationship to decision-making, one specific type of decision model is reviewed, namely Bidding Strategy Theory. Although these formulations reflect the probabilistic nature of risk, they normally fail as accurate descriptors of decision-making behavior. Field experiments conducted for this research indicate that decisions tend to be multiattributed in nature, perhaps explaining the inadequacy of these previous efforts which focus on but one decision variable; usually expected profit. Proceeding from this basis, a mathematical theory is presented which facilitates aggregation of preferences for all relevant decision variables, as they are valued by the individual under conditions of risk, into one global utility function. Four different formulations of Multiattribute Utility Analysis are presented, and the basic premises of each case are discussed in detail. Testing procedures for determining the correct formulation, as well as the limitations of each case and the Theory in general, are reviewed. Techniques are presented for structuring a decision problem which allow for closer approximation to the actual decision problem setting and for more valuable results. Finally, an illustrative problem is analyzed to demonstrate the requirements, mechanics, and solution format of the Theory. Since the roots of risk are grounded in the particular business market engaged, this analysis considers the Search for New Business Markets Problem. It is hypothesized that each business market contains an intrinsic and unique risk-opportunity structure and that the successful entrepreneur must seek the investment opportunity most compatible with his own preferences. The example problem is defined with three decision variables. Representative data and preference functions are used to determine what the consistent choice or choices would be, given the stated preferences of the decision-maker. The suggested course of action is based on a comparison of calculated global utility values. A sensitivity analysis is shown to be a valuable addition to the technique.