A dual mutation operator to solve the multi-objective production planning of perishable goods

• Type: Chapter or Part of a Book
• Date: 2013

Authors

Bernardo Almada Lobo (Author) FEUP; Pedro Amorim (Author) FEUP; Carlos Henggeler Antunes (Author) Other

Book

Title: Advances in Metaheuristics

2º Edition

Pages: 77-97

ISBN: ISBN 978-1-4614-6321

Publisher: Springer

Scientific classification

FOS: Engineering and technology

CORDIS: Technological sciences

Other information

Authenticus ID: P-011-XE2

DOI: 10.1007/978-1-4614-6322-1_5

Abstract (EN): This work proposes a dual mutation operator that uses dual information to solve the multi-objective lot-sizing and scheduling problem arising in perishable goods production planning. This operator takes advantage of the hybrid structure of a multi-objective evolutionary algorithm, which combines the well-known NonDominated Sorting Genetic Algorithm and a mixed integer linear programming solver. Chromosomes are coded as strings of integer values ¿xing the production sequence. For each feasible individual, the reduced costs of the relaxed binary variables coming from the coding scheme are used to guide the mutation process. To assess the performance of this operator, a set of randomly generated instances based on a methodology reported in the literature is tested and evaluated with multi-objective performance metrics. The results indicate that the operator is able to achieve consistently better solutions in terms of an approximation to the Pareto solutions compared to the same solution procedure without the dual mutation.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 20