Multi-objective optimisation using agent-based modelling

[摘要] ENGLISH ABSTRACT:It is very seldom that a decision-making problem concerns only a singlevalue or objective. The process of simultaneously optimising twoor more conicting objectives is known as multi-objective optimisation(MOO). A number of metaheuristics have been successfully adaptedfor MOO. The aim of this study was to investigate the feasibility ofapplying an agent-based modelling approach to MOO.The (s; S) inventory problem was chosen as the applicationeld forthis approach and Anylogic used as model platform. Agents in themodel were responsible for inventory and sales management, and hadto negotiate with each other in order tond optimal reorder strategies.The introduction of concepts such as agent satisfaction indexes,aggression factors, and recollection ability guided the negotiation processbetween the agents.The results revealed that the agents had the ability tond goodstrategies. The Pareto front generated from their proposed strategieswas a good approximation to the known front. The approach was alsosuccessfully applied to a recognised MOO test problem proving thatit has the potential to solve a variety of MOO problems.Future research could focus on further developing this approach formore practical applications such as complex supply chain systems, nancial models, risk analysis and economics.

[发布机构] Stellenbosch University