Title: Hybrid multi-objective evolutionary algorithm for solving RALB-II problem

Authors: Venkataramanaiah Saddikuti; Mukund Nilakantan Janardhanan; Vigneshwar Pesaru

Addresses: Indian Institute of Management Lucknow, Operations Management, B1 Sector 62 Noida-201301, Uttar Pradesh, India ' School of Engineering, University of Leicester, University Road, Leicester, LE1-7RH, UK ' Fair Issac Corporation (FICO), Bengaluru, 560017, India

Abstract: In this paper, we propose an MIP model for minimisation of cycle time and total assembly line cost simultaneously. Due to NP-hard nature of RALB (Rubinovitz and Bukchin, 1991), and to avoid local minima, a hybrid multi-objective evolutionary (H-MOE) algorithm developed based on the features of NSGA-II and simulated annealing algorithm is used to solve the RALB-II problem. Performance of the proposed algorithm is evaluated using datasets from Mukund et al. (2017b) and it was found that H-MOE algorithm outperformed the algorithm by Mukund et al. (2017b) in five out of seven cases on saving in cycle time and four out of seven in terms of total cost saving. In terms of average improvement, the proposed algorithm outperformed in terms total cost saving and underperformed in terms of time cycle compared with the performance of algorithm by Mukund et al. (2017b). Conclusions and future scope are highlighted.

Keywords: hybrid algorithm; multi-objective; non-dominated sorting genetic algorithm; NSGA; robotic assembly line; RAL; parameter tuning.

DOI: 10.1504/IJOR.2022.121490

International Journal of Operational Research, 2022 Vol.43 No.1/2, pp.131 - 149

Received: 30 Jan 2020
Accepted: 19 Jun 2020
Published online: 16 Mar 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article