Portal de Periódicos da CAPES

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

2011; Inderscience Publishers; Volume: 10; Issue: 1 Linguagem: Inglês

10.1504/ijsom.2011.041989

1744-2389

Siba Sankar Mahapatra, Niharranjan Nayak, K. Prasanna, G.S. Beriha,

Collaboration in agile enterprises

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.