个性化定制是制造业未来的发展趋势。本项目以个性化定制型多层单元流水式车间布局问题为对象，以其中的立体物料储运系统为中心，研究其布局的随机优化问题。研究内容有五点：分析并优化车间立体物流网络的拓扑结构；针对多层单元流水式车间物流储运系统，提出布局方案的评价体系与车间性能指标的计算方法；研究在制品存储单元吞吐能力及容量优化方法；优化车间层内存储单元及水平物流路径的布局；优化多层单元流水式车间的垂直物流布局。特色与创新之处在于：鉴于个性化定制型生产车间物流强度时空分布的非均衡性，充分考虑生产过程中出现的在制品各类“等待”现象，优化车间布局；采用多重有向封闭图的边连通性分析并优化立体物流网络拓扑结构的路径柔性；采用分解协调的算法结构求解多层车间的布局随机整数非线性规划。本项目的研究成果有助于个性化定制型多层单元流水式车间布局优化过程的建模与求解，为多层智能车间的设计与规划提供有效的理论与方法。

Individualized production is the developing tendency of the manufacturing industry in the future. For the multi-floor flowshop in individualized production, being focus on the intelligent material handling system which is consisted of many automatic guided vehicles (AGVs) driving in the unidirectional guided-path networks to have multiple loops configuration and autonomous vehicles storage and retrieval system (AVS/RS) in each floor, and lifts moving amang different floors, this project is aiming at the stochastic multi-floor layout problems. The research contents include the following 5 topics: in view of the unidirectional guided-path network of AGV, which has the topological configuration of multiple nested loops, this project puts forward the path-flexibility analysis and optimization methods based upon the edge-connectivity analysis of multiple closed digraph; presents the indices to evaluate the layout and the algorithms to calculate the performance indices; investigates the algorithms to optimize indices of storage cells for the work-in-pieces; gives out the approach to solve stochastic layout problems of each layer in the flowshops; puts forward the decomposition-and-coordination structure of the optimization algorithm to solve the layout optimization of the multi-floor flowshop. The creativities are as follows: the flowshop layout is optimized fully oriented to the various kinds of “waiting” phenomena of the work-in-pieces in the production due to the imbalance of workload distribution in space and time; the graph theoretic model is used to analyze and optimize the path-flexibility of topological configuration of the AGV-path networks; the decomposition-and-coordinate structure of algorithms is built up to solve the layout optimization of the multi-floor workshop, i.e. a stochastic integer nonlinear programming model. The research outputs of this project will provide with effective methodology to the design and planning theory for the multi-floor intelligent production workshops.