A particle swarm optimization-based approach to tackling simulation optimization of stochastic, large-scale and complex systems

Authors:
Ming Lu;Da-peng Wu;Jian-ping Zhang
Affiliations:
Dept. of Civil and Structural Engineering, Hong Kong Polytechnic University, Hong Kong, China;Dept. of Civil Engineering, Tsinghua University, Beijing, China;Dept. of Civil Engineering, Tsinghua University, Beijing, China
Venue:
ICMLC'05 Proceedings of the 4th international conference on Advances in Machine Learning and Cybernetics
Year:
2005

Citing 1
Cited 3

Parameter Selection in Particle Swarm Optimization

EP '98 Proceedings of the 7th International Conference on Evolutionary Programming VII

A survey on metaheuristics for stochastic combinatorial optimization

Natural Computing: an international journal
Simulation-optimization integrated approach to planning ready mixed concrete production and delivery: validation and applications

Winter Simulation Conference
A faster path planner using accelerated particle swarm optimization

Artificial Life and Robotics

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this research, the methodology of particle swarm optimization (PSO) combined with discrete system simulation is described and employed for enhancing logistical and operational efficiencies of practical one-plant-multi-site concrete delivery systems. In a case study using data from a concrete plant in Hong Kong, PSO was compared with the genetic algorithms (GA) in assessing two mechanisms for optimizing stochastic simulation systems, namely, "steady, averaging" and "non-steady, stochastic". The results show our PSO-based approach could rapidly (5 minutes) converge at the minimum level for an output of the simulation model while GA failed to converge or required a long time (about 1.5 hours) in search of the minimum. In conclusion, the proposed optimization procedures hold the potential to provide a generic, efficient approach to tackling simulation optimization of stochastic, large-scale and complex systems.