Multistage heuristic approach for solving infrastructure investment decision problems

The solution of an infrastructure investment decision problem is highly complex. The mathematical model developed in this paper represents the investment decision as a 0-1, nonlinear, multiobjective knapsack selection problem that discourages straightforward, exhaustive solving approaches because of high computational complexity. This paper introduces a staged heuristic approach to tackle this problem. The practical significance of the proposed approach is twofold, including (1) the inclusion of the gray relation concept for determining the rank order among alternatives; and (2) the introduction of a set of heuristics for alternative selection for removing the nonlinearity in the investment problem. Based on experience of implementing this approach in real-life problems, both the computational effort required and the quality of the results are satisfactory. It is suggested that the proposed approach be implemented by public agencies responsible for routine infrastructure investment decisions.