Comparative experiments with GRASP and constraint programming for the oil well drilling problem

Abstract

Before promising locations become productive oil wells, it is often necessary to complete drilling activities at these locations. The scheduling of such activities must satisfy several conflicting constraints and attain a number of goals. Here, we describe a Greedy Randomized Adaptive Search Procedure (GRASP) for the scheduling of oil well drilling activities. The results are compared with those from a well accepted constraint programming implementation. Computational experience on real instances indicates that the GRASP implementation is competitive, outperforming the constraint programming implementation. © Springer-Verlag Berlin Heidelberg 2005.