RBU: A model for reducing bias and uncertainty in multi-evaluator multi-criterion decision making

Decision-making in fields such as politics, engineering and healthcare, shapes the world and how it evolves. Both public and private organizations face challenges when making decisions. Two examples occurred with the Minnesota Department of Transportation in 2007 and the U.S. Department of Energy in 2008. Losing bidders for a bridge-rebuilding contract and a liquid-waste cleanup project, respectively, protested the agencies’ awards on the basis of evaluation methods and selection criteria. Multi-evaluator multi-criterion (MEMC) decision making can be controversial if bias or uncertainty find their way into the final decision. In a previous study, the authors of this paper developed a model that reduces the effect of uncertainty resulting from an evaluator’s insufficient expertise in a particular criterion. This paper builds on the previous study by also correcting for potential favoritism or bias. It presents a more comprehensive mathematical model that supports MEMC decisions and protects decision-makers from criticism. The methodology includes: (1) proposing a probabilistic model and its assumptions; (2) developing an iterative algorithm; (3) testing the algorithm and analyzing its convergence; and (4) revising the model based on the results. Tests of the model show it performs better than the simple average method on 100% of the simulations.