Abstract: Business of any organization is continuous process and therefore supplier selection problem is dynamic in nature. And firm’s demand and supplier’s capacity for each part changes over time due to market fluctuation and customer demand. Thus supplier selection problem is also fuzzy in real practice. Based on dynamic and fuzziness in nature on supplier selection problem, considering supplier’s capacity and manufacturer’s demand for fuzzy variables in every time period, in this paper, the supplier selection problem for multi-period, multi-parts, and multi-source is considered as a 0-1 mixed integer fuzzy dynamic non-linear programming problem, objective function is minimization of total costs. And then a 0-1 mixed integer fuzzy dynamic non-linear program model is proposed. In order to solve this fuzzy dynamic model, the model is transformed into a deterministic 0-1 mixed integer dynamic non-linear model by converting fuzzy chance constraints to their respective deterministic equivalent formulation by Credibility Theory. A numerical example is illustrated to show that the proposed model is feasible and effective.

Key words: dynamic supplier selection, fuzzy chance constrained, mixed-integer non-linear program, credibility theory

摘要： 基于供应商选择问题的动态性和模糊性,考虑在每个周期内生产商的需求能力及供应商的供应能力为模糊变量,本文将一个多阶段多商品多渠道的供应商选择问题视为一个0-1混合整数模糊动态非线性规划问题,目标函数为总成本最小化。然后建立了0-1混合整数模糊动态非线性规划模型。为了求解该模型,通过可信性理论把模型中模糊机会约束清晰化,将该模型转化为一个确定型的0-1混合整数动态非线性规划模型。最后给出了一个数值算例验证了模型的可行性。

关键词: 动态供应商选择, 模糊机会约束, 混合整数非线性规划, 可信性理论