Improvement of PSO algorithm by memory based gradient search - application in inventory management

TL;DR

This paper enhances the Particle Swarm Optimization algorithm with a memory-based gradient search method using Monte Carlo simulation, improving inventory management in stochastic supply chains.

Contribution

It introduces a novel memory-based gradient estimation technique for PSO applicable to uncertain systems, validated through benchmarks and supply chain optimization.

Findings

Improved convergence of PSO with memory-based gradients.

Effective application to complex inventory management problems.

Demonstrated robustness in stochastic supply chain scenarios.

Abstract

Advanced inventory management in complex supply chains requires effective and robust nonlinear optimization due to the stochastic nature of supply and demand variations. Application of estimated gradients can boost up the convergence of Particle Swarm Optimization (PSO) algorithm but classical gradient calculation cannot be applied to stochastic and uncertain systems. In these situations Monte-Carlo (MC) simulation can be applied to determine the gradient. We developed a memory based algorithm where instead of generating and evaluating new simulated samples the stored and shared former function evaluations of the particles are sampled to estimate the gradients by local weighted least squares regression. The performance of the resulted regional gradient-based PSO is verified by several benchmark problems and in a complex application example where optimal reorder points of a supply chain are determined.