Article

A multi objective volleyball premier league algorithm for green scheduling identical parallel machines with splitting jobs

Details

Citation

Salimifard K, Li J, Mohammadi D & Moghdani R (2021) A multi objective volleyball premier league algorithm for green scheduling identical parallel machines with splitting jobs. Applied Intelligence, 51 (7), pp. 4143-4161. https://doi.org/10.1007/s10489-020-02027-1

Abstract
Parallel machine scheduling is one of the most common studied problems in recent years, however, this classic optimization problem has to achieve two conflicting objectives, i.e. minimizing the total tardiness and minimizing the total wastes, if the scheduling is done in the context of plastic injection industry where jobs are splitting and molds are important constraints. This paper proposes a mathematical model for scheduling parallel machines with splitting jobs and resource constraints. Two minimization objectives - the total tardiness and the number of waste - are considered, simultaneously. The obtained model is a bi-objective integer linear programming model that is shown to be of NP-hard class optimization problems. In this paper, a novel Multi-Objective Volleyball Premier League (MOVPL) algorithm is presented for solving the aforementioned problem. This algorithm uses the crowding distance concept used in NSGA-II as an extension of the Volleyball Premier League (VPL) that we recently introduced. Furthermore, the results are compared with six multi-objective metaheuristic algorithms of MOPSO, NSGA-II, MOGWO, MOALO, MOEA/D, and SPEA2. Using five standard metrics and ten test problems, the performance of the Pareto-based algorithms was investigated. The results demonstrate that in general, the proposed algorithm has supremacy than the other four algorithms.

Keywords
Parallel machine scheduling; Splitting jobs; Wastes; Total tardiness; Multi-objective optimisation; Volleyball premier league

Journal
Applied Intelligence: Volume 51, Issue 7

StatusPublished
Publication date31/07/2021
Publication date online08/12/2020
Date accepted by journal16/10/2020
URLhttp://hdl.handle.net/1893/32140
ISSN0924-669X
eISSN1573-7497