Hybrid biased random sampling for multiple resource-constrained project scheduling problems

In this paper we propose a new heuristic to solve the well-known multiple resource-constrained project scheduling problem. The method is basically a biased random sampling procedure which shows extremely good results by use of the following features: A problem-based selection of the solution space, a sample-size-based guidance of the search, application of a priority rule superior to so-far existing rules, and finally the application of global and local (lower) bounds. Evaluating this new heuristic on a set of widely used benchmark-instances we show that it derives superior results than all other existing polynomially bounded algorithms.