Title: A novel swarm optimisation approach for risk-based partner selection in virtual enterprises

Authors: S.S. Mahapatra, N.R. Nayak, K. Prasanna, G.S. Beriha

Addresses: Department of Mechanical Engineering, National Institute of Technology, Rourkela – 769008, India. ' Department of Mechanical Engineering, National Institute of Technology, Rourkela – 769008, India. ' Department of Mechanical Engineering, National Institute of Technology, Rourkela – 769008, India. ' Department of Humanities and Social Sciences, National Institute of Technology, Rourkela – 769008, India

Abstract: Selection of right partners is vital for virtual enterprise (VE) to take the advantage of market opportunity. Partner selection problem belongs to combinatorial optimisation category and known as NP-hard. A modified discrete particle swarm optimisation (PSO) algorithm is proposed, in this work, to select alliances in a VE. The velocity terms, which is commonly used in the standard PSO algorithm, is avoided; rather, the concept of proportional likelihood with modifications, a method used in data mining techniques, is adopted to improve solution quality. A set of experiments have been conducted and results indicate PSO algorithm is an effective and efficient way to solve the partner selection problems with precedence, risk of failure and due date constraints. A design of experiment (DoE) approach is adopted to study statistical significance of various model parameters on solution quality.

Keywords: discrete PSO; particle swarm optimisation; DPSO; virtual enterprises; partner selection; ANOVA; DOE; proportional likelihood; modifications; data mining; design of experiments.

DOI: 10.1504/IJSOM.2011.041989

International Journal of Services and Operations Management, 2011 Vol.10 No.1, pp.52 - 73

Published online: 11 Mar 2015 *

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