Resource-constrained project scheduling: an evaluation of adaptive controlschemes for parameterized sampling heuristics

For most computationally intractable problems there exists no simple heuris tic that consistently outperforms all other heuristics. One remedy is to bu ndle simple heuristics into composite ones in a fixed and predetermined way . Adaptive control schemes take this approach one step further by dynamical ly combining algorithms. Several such algorithms have been proposed recentl y in various settings, yet an experimental investigation comparing them wit h other contemporary methods has been lacking. The aim here is to close thi s gap by a comprehensive computational study on the field of resource-const rained project scheduling. Also shown is how to improve effectiveness of th e best algorithm by means of randomized sampling. Finally, revealed are sev eral advantages of adaptive control schemes over other algorithms which fac ilitate the OR practitioner's task of designing good algorithms for newly a rising problems.