Optimization with fuzzy linear programming and fuzzy knowledge base

摘要

Due to the complexity of modern systems, a decision process is most often dependent at least in part on qualitative reasoning, especially in the case where human involvement is important. The paper proposes an optimization methodology-fuzzy linear programming coordinated with a fuzzy linguistic knowledge base-to handle quantitative and qualitative information simultaneously. In this approach, fuzzy linear programming is used to model quantitatively the basic resource constraints and objective functions, while a linguistic knowledge base is used to model imprecise human expertise. Freight train dispatching, in which a physical rail network is very complex and human expertise of a yardmaster plays an important role in decision making, illustrates the application.