批量储运耦合多机器人制造单元的随机生产系统资源优化配置

Considering Stochastic Production Systems (SPS) consisted of multiple manufacturing cells interconnected with batch Material Handling System (MHS), this project deals with the Resource Allocation Problem (RAP), i.e. under the condition of stochastic arrivals of the orders, minimizing the total amount of investment subject to constraints of both average throughput and make-span of products. The content includes: setting up the queueing network models of flexible manufacturing cells with Dual Resource Constriants (DRC) in which robots are used to load/ unload the parts; building up the layered queueing networks of SPS consisted of multiple manufacturing cells interconnected with batch MHS; putting forward to algorithm to solve RAP with DRC; proposing a decomposition-coordination solution scheme to solve RAP of SPS. Creativities of this project lie in the following, open queueing network models with finite buffers are first extended to describe the beheaviors of manufacturing cells with DRC, state-dependent batch transportation and routing choices; A problem-features based branch-and-bound algorithm is put forward by means of decomposing the complicated problem into two interconnected sub-problems, one is a Capacity Configuration Problem (CCP) and the other a Buffer Allocation Problem (BAP), an iterative framework is presented between them, thus enhancing the efficiency of searching the optimal solution to RAP of a manufacturing cell with DRC; a decomposition- coordination solution scheme is then employed to solve RAP of SPS optimally. The output of this project can provide the order-driven manufacturers with scientific methodology for the resource allocation, which require to take capacity expansion or to build up a new factory.

针对具有批量储运耦合多机器人柔性制造单元的随机生产系统，本项目研究资源的优化配置方法，即在订单随机到达的情况下，配置各类制造资源的数量，使得生产系统以最小的投资满足产品的平均产出率和平均生产周期约束。具体包括：建立具有自动装卸工件机器人的制造单元双重资源排队网模型；建立具有批量储运环节的多单元生产系统分层排队网模型；提出单元双重资源优化配置问题的求解方法；提出生产系统资源配置问题的分解协调优化算法。特色与创新之处体现在：扩展了有限缓冲区容量开排队网模型的适用范围，使之能够描述双重资源约束、随机批量运输和状态相关的随机路径；提出了一种迭代求解制造单元双重资源优化配置问题的分支定界方法；提出了一种分解协调的策略解决随机大系统各部分之间的关联耦合效应。研究成果将为定制型制造企业，在扩大现有生产能力或建立新工厂时，提供科学的决策方法和分析手段，优化投资效益。

针对具有批量储运耦合多机器人柔性制造单元的随机生产系统，本项目研究资源的优化配置方法，即在订单随机到达的情况下，配置各类制造资源的数量，使得生产系统以最小的投资满足产品的平均产出率和平均生产周期约束。主要研究内容包括：建立具有自动装卸工件机器人的制造单元双重资源排队网模型；建立具有批量储运环节的多单元生产系统分层排队网模型；提出单元双重资源优化配置问题的求解方法；提出生产系统资源配置问题的分解协调优化算法。主要研究成果包括：扩展了有限缓冲区容量开排队网模型的适用范围，使之能够描述双重资源约束、随机批量运输和状态相关的随机路径；提出了一种迭代求解制造单元双重资源优化配置问题的分支定界方法；提出了一种分解协调的策略解决随机大系统各部分之间的关联耦合效应。研究成果将为定制型制造企业，在扩大现有生产能力或建立新工厂时，提供科学的决策方法和分析手段，优化投资效益。