市場指向プログラミングにおけるエージェントの戦略と資源配分特性-基本特性解析

抄録

Market-oriented programming is a multi-agent-based concept to facilitate distributed problem solving. In market-oriented programming we take the metaphor of an economy computing multiagent behaviour literally, and directly implement the distributed computation as a market price system. In a market-oriented programming approach to distributed problem solving, the resource allocation for a set of computational agents is derived by computing competitive market of an artificial economy. Since heterogeneous agent strategies have intensive effects on the resource allocation, dynamism, we analyse and systematise their relationship in supply chain simulation with a virtual market.The behaviour of the system can be analysed in agent interactions based on economic strategies.