Parallel Metaheuristics for Robust Graph Coloring Problem
نویسنده
چکیده
In this chapter a new formulation of the robust graph coloring problem (RGCP) is proposed. In opposition to classical GCP defined for the given graph G(V,E) not only elements of E but also Ē can be subject of color conflicts in edge vertices. Conflicts in Ē are assigned penalties 0<P(e)<1. In addition to satisfying constraints related to the number of colors and/or a threshold of the acceptable sum of penalties for color conflicts in graph complementary edges (rigidity level), a new bound called the relative robustness threshold (RRT) is proposed. Then three metaheuristics – SA, TS, EA and their parallel analogues PSA, PTS, PEA – for that version of RGCP are presented and experimentally tested. For comparison we use DIMACS graph coloring instances in which a selected percentage of graph edges E is randomly moved to Ē. Since graph densities and chromatic numbers of DIMACS GCP instances are known in advance, the RGCP instances generated on their basis are more suitable for testing algorithms than totally random instances used so far. The results of the conducted experiments are presented and discussed.
منابع مشابه
A new Simulated Annealing algorithm for the robust coloring problem
The Robust Coloring Problem (RCP) is a generalization of the well-known Graph Coloring Problem where we seek for a solution that remains valid when extra edges are added. The RCP is used in scheduling of events with possible last-minute changes and study frequency assignments of the electromagnetic spectrum. This problem has been proved as NP-hard and in instances larger than 30 vertices, meta-...
متن کاملParallel Jobs Scheduling with a Specific Due Date: Asemi-definite Relaxation-based Algorithm
This paper considers a different version of the parallel machines scheduling problem in which the parallel jobs simultaneously requirea pre-specifiedjob-dependent number of machines when being processed.This relaxation departs from one of the classic scheduling assumptions. While the analytical conditions can be easily statedfor some simple models, a graph model approach is required when confli...
متن کاملTesting of parallel metaheuristics for graph partitioning problems
In this paper we describe computer experiments while testing a family of parallel and hybrid metaheuristics against a small set of graph partitioning problems like clustering, partitioning into cliques and coloring. In all cases the search space is composed of vertex partitions satisfying specific problem requirements. The solver application contains two sequential and nine parallel/hybrid algo...
متن کاملA GRASP for Coloring Sparse Graphs
We first present a literature review of heuristics and metaheuristics developed for the problem of coloring graphs. We then present a Greedy Randomized Adaptive Search Procedure (GRASP) for coloring sparse graphs. The procedure is tested on graphs of known chromatic number, as well as random graphs with edge probability 0.1 having from 50 to 500 vertices. Empirical results indicate that the pro...
متن کامل