基于排队网的FFS设备配置近似优化方法研究

发布时间：2018-02-24 14:13

本文关键词： 设备配置排队网输出率平均生产周期　出处：《广东工业大学》2013年硕士论文　论文类型：学位论文

【摘要】：定制型装备制造系统是一种面向订单的制造系统,订单驱动的方式使得制造系统的生产过程复杂多变、生产过程中充满了随机因素,因此造成了生产不稳定、产能不匹配、拖期现象严重等诸多问题。在制造系统的生产过程中,生产设备是制造生产过程的核心资源,也是完成制造加工的功能单元,生产设备的配置尤为重要。本文针对定制型装备制造企业的多级柔性流水车间(FFS),以输出率和平均生产周期为约束条件,最小化制造系统的总投资成本。本文构建了排队网模型,但鉴于约束条件(输出率和平均生产周期)无法用决策变量(设备)的封闭形式加以表达,因此,本文通过扩展法近似求解和eM-Plant仿真统计求解的方式,获得约束条件的数值。本文主要研究内容如下：首先,对设备配置问题进行了描述与建模。从现实柔性流水制造系统中,提取产品类型,订单分布,工艺路径,工时分布等,构建了基于多级M/M/C/K的排队网模型和数学模型。其次,对性能指标进行近似求解。由于约束条件(输出率和平均生产周期)无法用决策变量的封闭形式加以表达,本文详细介绍了扩展法,并在MATLAB(?)不境下开发了一套求解性能指标的程序,并进行了算例对比以验证其有效性；其次,在eM-Plant软件平台上建立了实验仿真模型,给出了一组算例用于验证仿真模型的有效性。接着,对设备配置问题进行求解。建立了以优化阻塞率和以设备采购最低成本的二种启发式规则,针对给出的一个算例,运用扩展法解析计算和仿真统计的方法针对生产设备进行优化,通过对比,分析了四组数据的特性。最后,基于上述步骤所获得的理论研究成果,并结合合作企业的实际运作特点,自己开发了用于设备配置优化的软件包。总之,本文运用排队网及其相关理论,构建了一个随机环境下的多级柔性流水车间设备配置优化模型,并给出了相应求解算法。设计、开发了配套的软件工具,为企业的资源配置提供了一种有效的方法。
[Abstract]:The custom-made equipment manufacturing system is an order-oriented manufacturing system. The order-driven mode makes the manufacturing process complex and changeable, and the production process is full of random factors, which results in unstable production and mismatch of production capacity. In the production process of manufacturing system, production equipment is not only the core resource of manufacturing process, but also the functional unit to complete manufacturing process, the configuration of production equipment is particularly important. In this paper, the output rate and average production cycle are taken as constraints to minimize the total investment cost of manufacturing system in the multi-level flexible income workshop of custom-made equipment manufacturing enterprises. A queuing network model is constructed in this paper. However, since the constraint conditions (output rate and average production cycle) cannot be expressed in the closed form of decision variables (equipment), this paper uses the extended method to approximate the solution and the eM-Plant simulation statistical solution. The main contents of this paper are as follows:. Firstly, the problem of equipment configuration is described and modeled, and the queuing network model and mathematical model based on multilevel M / M / C / K are constructed by extracting product type, order distribution, process path and man-hour distribution from the actual flexible income manufacturing system. Secondly, the performance index is approximately solved. Because the constraint conditions (output rate and average production period) can not be expressed in the closed form of decision variables, this paper introduces the extended method in detail, and introduces it in MATLAB? In this paper, a program to solve the performance index is developed, and an example is compared to verify its validity. Secondly, an experimental simulation model is established on the platform of eM-Plant software, and a set of examples are given to verify the validity of the simulation model. Then, the problem of equipment configuration is solved, and two heuristic rules are established to optimize the blocking rate and the lowest cost of purchasing equipment, and an example is given. The extended method is used to analyze the calculation and the simulation statistics are used to optimize the production equipment. Through comparison, the characteristics of four groups of data are analyzed. Finally, based on the theoretical research results obtained by the above steps and the practical operation characteristics of cooperative enterprises, a software package for equipment configuration optimization is developed by ourselves. In a word, using the queuing network and its related theory, this paper constructs a multi-level flexible income workshop equipment configuration optimization model under random environment, and gives the corresponding solution algorithm. It provides an effective method for enterprise resource allocation.
【学位授予单位】：广东工业大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TH164

【参考文献】