MTO模式下的制造企业稳健型调度问题研究

发布时间：2018-09-09 11:57

【摘要】：面对生产资源的限制和市场竞争的压力,MTO生产模式下的制造企业需要从内、外部环境角度统一考虑市场战略和生产管理,将企业的物流仓储、生产制造与业务管理高度集成,使个性化的产品按时、按量的交到客户手中。然而,市场形势的变化莫测以及生产系统固有的复杂性,都会导致实际生产过程中各种扰动因素的产生,一旦某个环节出现错误,都会给企业带来不必要的损失。因此,设计合理的扰动检测方法以及建立稳健型调度问题模型,能够有效地应对各种扰动、保持生产系统的稳定性、提高企业的生产效益。重调度作为稳健型调度的一种,能够有效地消除扰动因素给生产系统带来的影响,因此,受到当今工程界和学术界的广泛关注。本文在详细分析已有重调度相关理论的基础上,围绕动态不确定环境下制造企业扰动因素检测和处理方法,从重调度因素、重调度策略以及重调度方法这三方面,较为全面系统地研究制造企业稳健型调度问题。首先,本文详细描述了生产制造车间常见的扰动因素,根据其对生产系统的影响程度进行分类,并采用模糊数学理论和概率基神经网络相结合的方法,设计了一种模糊神经网络算法,来量化评估扰动因素对生产系统的影响程度。然后,针对现有混合型重调度策略存在的不足,提出了一种改进的混合型重调度策略,在使用模糊神经网络对扰动因素量化评估的基础上,选择合适的响应方式,并通过引入最小重调度时间间隔min?T进行约束,协调了周期性重调度策略和事件驱动型重调度策略之间的关系。其次,针对动态不确定环境下,生产过程中需要重新生成调度方案的情况,以目前制造业广泛存在的柔性作业车间为研究对象,构建了一种具有自适应能力的重调度模型,并提出一种基于双层编码的遗传算法对模型进行求解。最后,总结归纳了一种制造系统的自适应重调度流程,并在生成调度方案时,引入智能优化算法与人工调度相结合的人机协同策略,有效地应对制造系统中各种常见的扰动因素,保证生产的连续性与均衡性。
[Abstract]:Faced with the limitation of production resources and the pressure of market competition, manufacturing enterprises under MTO production mode need to consider market strategy and production management from the angle of internal and external environment, and integrate the logistics warehousing, manufacturing and business management of enterprises. Make personalized products on time, according to the quantity of the hands of the customer. However, the unpredictable market situation and the inherent complexity of the production system will lead to a variety of disturbance factors in the actual production process. Once a link is wrong, it will bring unnecessary losses to the enterprise. Therefore, the design of reasonable disturbance detection method and the establishment of robust scheduling problem model can effectively deal with all kinds of disturbances, maintain the stability of the production system, and improve the production efficiency of enterprises. As a kind of robust scheduling, rescheduling can effectively eliminate the influence of disturbance factors on production system. Based on the detailed analysis of existing rescheduling theories, this paper focuses on three aspects: detection and processing of disturbance factors in dynamic uncertain environment, rescheduling factors, rescheduling strategies and rescheduling methods. The robust scheduling problem of manufacturing enterprises is studied comprehensively and systematically. First of all, this paper describes the common disturbance factors in manufacturing workshop in detail, classifies them according to their influence on production system, and adopts the method of combining fuzzy mathematics theory with probabilistic neural network. A fuzzy neural network algorithm is designed to quantitatively evaluate the influence of disturbance factors on production system. Then, aiming at the shortcomings of the existing hybrid rescheduling strategy, an improved hybrid rescheduling strategy is proposed. Based on the quantitative evaluation of disturbance factors by using fuzzy neural network, the appropriate response mode is selected. The relationship between periodic rescheduling policy and event-driven rescheduling policy is coordinated by introducing minimum rescheduling interval min?T. Secondly, a rescheduling model with adaptive ability is constructed to solve the problem that scheduling schemes need to be regenerated in the production process under dynamic uncertain environment, and the flexible job shop, which is widely existed in manufacturing industry, is taken as the research object. A genetic algorithm based on double-level coding is proposed to solve the model. Finally, an adaptive rescheduling process of manufacturing system is summarized, and a man-machine coordination strategy combining intelligent optimization algorithm and manual scheduling is introduced in the process of generating scheduling scheme. Effectively deal with the common disturbance factors in manufacturing system to ensure the continuity and balance of production.
【学位授予单位】：重庆理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F273;F425

【参考文献】