Abstract: To effectively integrate internal resource in industrial chain,the coordinated scheduling optimization for production and distribution of industrial chain were researched.After analyzing three value steps,the related mathematical model was established,and an uncertainty compensation model was established to solve the uncertainty in production scheduling.On this basis,an uncertain programming optimization system integrated stochastic simulation,fuzzy inference,neural network and heuristic algorithm was proposed,which could fit the randomness,concurrency and fuzziness in uncertain programming,and could obtain an optimal solution of coordinated scheduling of production and distribution of industrial chain under uncertain environment.Simulation experiments were conducted on instances of different sizes,and the results showed that the uncertain environment in the scheduling process could be well fitted by the proposed system.The adaptable solution was obviously superior to the genetic algorithms and the stage optimization algorithm.

Key words: industrial chains collaboration, uncertain programming, neural network, genetic algorithms

摘要： 为有效整合产业链内部资源,提升产业链竞争优势,对产业链生产与配送协同调度优化问题进行了研究,分析问题涉及的3个价值环节,建立其数学模型,并针对调度中存在的生产不确定性问题,进行模型的不确定性补偿,在此基础上,提出一种基于随机模拟、模糊推理、神经网络、启发式算法融合的不确定规划最优化系统。基于该系统,对不确定规划中存在的随机性、并发性与模糊性进行拟合,并通过启发式寻优求解出不确定环境下产业链生产与配送协同调度问题的最优方案。通过对不同规模问题实例进行仿真实验,结果表明所构造的系统可较好地拟合调度过程中的不确定环境,在此基础上可得出适应性强的方案解,解的质量明显优于生产与配送协同调度遗传算法和生产与配送阶段优化算法。

关键词: 产业链协同, 不确定规划, 神经网络, 遗传算法