Globalization has enhanced the need for effective management of supply chain in global context. As the supply chain gets bigger, it becomes increasingly complex as the number of variables to handle increase along with the difficulty to coordinate between large number of members. Petroleum industry supply chain is one such complex supply chain because of the large number of processes involved right from extraction till customer delivery. Large number of companies are involved in such a chain, and its success or failure is detrimental to the economy of the nation. Multi-agent technology provides an optimal platform for effective and efficient supply chain management of large supply chains as it breaks the supply chain into discrete agents thus rendering easier coordination between the agents. In this paper multi-agent technology has been applied to make the supply chain faster as compared to conventional supply chain practices. The number of alternatives available in a petroleum supply chain is more and this also increases the number of agents. Here a representative but exhaustive model is considered, by taking into account the various options available at each agent. It has also been made additionally complex by including realistic assumptions like non-linear quantity dependence of cost of operation. Solving this problem requires an advanced Heuristic method like Co-evolutionary Particle Swarm Optimization based on Cauchy distribution which has been precisely made for cost and resource allocation. Cauchy provides a very rapid path towards approaching the solution. The previous work done on the Co-PSO using Gaussian approach is compared with that of Cauchy approach and relevant conclusions are drawn. Simulation results also indicate the better convergence characteristics of the Cauchy over Gaussian approach, hence proving its nobility. Comparative study of the final results of the two methodologies also indicates towards the superiority of the Cauchy approach.
- Co-evolutionary Particle Swarm Optimization
- Petroleum industry
ASJC Scopus subject areas
- Computer Science Applications
- Artificial Intelligence