AB - In this paper we propose a forward-backward improvement heuristic for the variant of resource-constrained project scheduling problem aiming to maximise the net present value of a project. It relies on the Lagrangian relaxation method to generate an initial set of schedules which are then improved by the iterative forward/backward scheduling technique. It greatly improves the performance of the Lagrangian relaxation based heuristics in the literature and is a strong competitor to the best meta-heuristics. We also embed this heuristic into a state-of-the-art CP solver. Experimentation carried out on a comprehensive set of test data indicates we compare favorably with the state of the art. © Springer-Verlag 2013. AU - Gu, H AU - Schutt, A AU - Stuckey, PJ DA - 2013/12/01 DO - 10.1007/978-3-642-38171-3_24 EP - 346 JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) PY - 2013/12/01 SP - 340 TI - A lagrangian relaxation based forward-backward improvement heuristic for maximising the net present value of resource-constrained projects VL - 7874 LNCS Y1 - 2013/12/01 Y2 - 2024/03/29 ER -