PARAMETER LESS HYBRID IG-JAYA APPROACH FOR PERMUTATION FLOW SHOP SCHEDULING PROBLEM

Abstract

Permutation flow shop scheduling problem (PFSP) is a well-known NP-hard problem with extensive engineering relevance. Consequently, various meta-heuristics have been proposed to obtain near optimum solutions. However, most of them involve tuning algorithm-specific parameters, which leads to excessive computational complexities. A recently developed meta-heuristic named Jaya algorithm is simple yet efficient as it is a parameter-less algorithm, thus does not require tuning of algorithm-specific parameters. However, as the problem size grows, Jaya algorithm loose solution diversity and tends to get trapped at local optima. To alleviate such drawbacks while retaining the parameter-less feature of Jaya, the present paper proposes a Hybrid Iterated Greedy based Jaya algorithm (HIGJ). The proposed approach combines a novel integration of Jaya’s population-based optimization with the single solutionbased iterated greedy algorithm for enhancing population members to retrieve improved solutions. The objective is to minimize the makespan. The destruction and construction phase of Iterated Greedy (IG) is embedded into the best solution of Jaya algorithm. Furthermore, the local search method is incorporated to improve the overall quality of candidate solutions resulting in faster convergence towards the optimal solution. An exhaustive comparative study along with statistical analysis is conducted with public benchmarks to realize the effectiveness of the proposed HIGJ algorithm. Computational results reveal that HIGJ yields improved solutions and outperforms some of the efficient meta-heuristics available in the literature.