2015/11/13 by Shahriar Asta, Asta, Shahriar, Daniel Karapetyan +7
Decision Sciences · Engineering · #Artificial Intelligence (cs.AI) #BIM and Construction Integration #Data Structures and Algorithms (cs.DS) #FOS: Computer and information sciences #Neural and Evolutionary Computing (cs.NE) #Resource-Constrained Project Scheduling #Scheduling and Optimization Algorithms #Scheduling and Timetabling Solutions
paper · pdf · doi:10.48550/arxiv.1511.04387
openalex publication_date 2015/11/13 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Multi-mode resource and precedence-constrained project scheduling is a\nwell-known challenging real-world optimisation problem. An important variant of\nthe problem requires scheduling of activities for multiple projects considering\navailability of local and global resources while respecting a range of\nconstraints. A critical aspect of the benchmarks addressed in this paper is\nthat the primary objective is to minimise the sum of the project completion\ntimes, with the usual makespan minimisation as a secondary objective. We\nobserve that this leads to an expected different overall structure of good\nsolutions and discuss the effects this has on the algorithm design. This paper\npresents a carefully designed hybrid of Monte-Carlo tree search, novel\nneighbourhood moves, memetic algorithms, and hyper-heuristic methods. The\nimplementation is also engineered to increase the speed with which iterations\nare performed, and to exploit the computing power of multicore machines.\nEmpirical evaluation shows that the resulting information-sharing\nmulti-component algorithm significantly outperforms other solvers on a set of\n"hidden" instances, i.e. instances not available at the algorithm design phase.\n