Title: Developing a hybrid genetic algorithm solution to optimise multi-echelon supply chains

Authors: Haitham A. Mahmoud; Mohammed H. Hassan; Emad S. Abouel Nasr

Addresses: Faculty of Engineering, Mechanical Engineering Department, Helwan University, Cairo, Egypt ' Faculty of Engineering, Mechanical Engineering Department, Helwan University, on-leave to the British University in Egypt (BUE), Cairo 11732, Egypt ' Industrial Engineering Department, College of Engineering, King Saud University, P. O. Box 800, Riyadh 11421, Saudi Arabia; Faculty of Engineering, Mechanical Engineering Department, Helwan University, Cairo 11732, Egypt

Abstract: The design of dynamic supply chain (DSC) models has received a great attention in the last two decades. In this paper, a forward supply chain network (SCN) model is developed in which both dynamic facility locations and dynamic distributed quantities of materials and products are assumed. The problem is formulated mathematically using mixed integer linear programming (MILP). The solution methodology adopted is the hybrid genetic algorithm (hGA) comprising genetic algorithm (GA) and pattern search (PS) optimisation techniques. Two experimental studies are conducted to analyse the developed SCN model and the developed hGA. The results of the hGA analysis show that the accuracy of the proposed hGA is improved in the cases of large number of nodes of the SC's echelons, small number of manufactured products and large quantities of products' demands.

Keywords: dynamic supply chains; forward logistics networks; genetic algorithms; pattern search; hGAs; hybrid GAs; experimental study; optimisation; multi-echelon supply chains; supply chain management; SCM; dynamic modelling; mixed integer linear programming; MILP.

DOI: 10.1504/IJCENT.2013.058632

International Journal of Collaborative Enterprise, 2013 Vol.3 No.4, pp.287 - 313

Published online: 12 Jan 2014 *

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