A Negotiation-Based Approach to Supply Chain Planning and Scheduling Problems in a Fully Distributed Environment

摘要

This paper presents an agent-based fuzzy constraint-directed negotiation mechanism for planning and scheduling in supply chain. The supply chain scheduling problem is modeled as a set of fuzzy constraint satisfaction problems (FCSPs), interlinked together by inter-agent constraints. For converging each distinct firm's interests, the conflicts among the set of FCSPs are resolved through negotiation by iteratively exchanging offers/counters with limited sharing of their perspectives and preferences. During the negotiation, proposing offers/counter-offers takes not only firm's self-interest and preferences but also opponents' perspectives into consideration. By sharing perspectives between agents to gradually uncovering the intent of opponents, consensus can be obtained and the quality of consensus can be guaranteed at satisfactory level. Experimental results suggest that the proposed approach obtains a superior solution for supply chain scheduling than other negotiation models in a fully distributed processing.