MC环境下产品族零件工时定额制定方法研究

发布时间：2018-02-11 10:06

本文关键词： 工时定额制造特征虚拟零件工时描述平台神经网络　出处：《重庆大学》2012年硕士论文　论文类型：学位论文

【摘要】：随着经济的发展和顾客受教育水平的提高，顾客不再像过去那样只对产品的质量有要求，还要求产品具有新颖的外观和多样化的功能。因此，制造企业的市场主导已经从卖方市场转向买方市场，大规模定制(Mass Customization，MC)的生产模式应运而生。MC生产模式在一定程度上提高了制造企业对市场的应对能力，满足了顾客快速、多样的产品需求，但同时由产品种类的多样化带来的生产工艺的差异化与生产过程的动态化、复杂化，导致企业难以对产品交货期进行准确判断。造成这一问题的根本原因在于：复杂冗余、低效的传统工时制定方法已经不能精确地反映MC生产模式下产品的工时，进而导致生产计划无法按期执行，生产时间无法确定、产品交货期无法预算、产品交货期延迟的严重后果。因此，MC企业要在严峻的市场竞争中快速满足顾客需求，必须重新思考工时定额的制定方法，以满足MC生产环境下工时定额制定的高效性和准确性，提高企业制造系统对市场的反应能力。本文根据MC生产模式下制造系统的特点，在传统工时定额制定方法的基础上，以产品族零件为研究对象，提出了以零件制造特征为载体的工时定额制定新方法。该方法遵循由部分到整体、由具体到抽象的思想，以零件的制造特征为准则，运用零件相似性判定和神经网络等技术，，提高了工时制定的高效性和准确性。首先，通过分析MC环境下制造系统的特点，提出了零件制造特征的概念，并将其作为新方法的工时载体；其次，为了保证方法的高效性，根据MC生产模式下的零件设计和制造理论，建立了零件的制造特征模型，并研究了基于零件制造特征的零件相似性判定，该判定着眼于零件本身的两个信息层次：零件的类型和零件材料、零件制造特征模型；第三，定义了虚拟零件和工时描述平台的概念，给出了虚拟零件的构建规则，运用事物特性表技术为每一个虚拟零件建立零件工时描述平台，对每个零件组内所有零件的工时及其影响因素进行描述；第四，为了保证方法的准确性，运用神经网络技术，设计了零件的工时神经网络结构并建立了模型，并对运用MATLAB软件实现零件工时定额的估算进行了研究；最后，通过案例阐述了新方法的具体操作过程。
[Abstract]:With the development of economy and the improvement of the education level of customers, customers no longer only require the quality of products as they used to, but also require the products to have novel appearance and diversified functions. The market dominance of manufacturing enterprises has shifted from the seller's market to the buyer's market. The production mode of mass customization mass customization (MC) has emerged as the times require. The production model of MC has to a certain extent improved the ability of manufacturing enterprises to respond to the market and satisfied the rapid development of customers. The diversity of product demand, but at the same time, the differentiation of production process and the dynamic and complicated of production process brought by the diversification of product types, It is difficult for enterprises to accurately judge the delivery time of products. The root cause of this problem lies in the fact that the complicated redundancy and inefficient traditional man-hour formulation method can no longer accurately reflect the working hours of products under the MC production mode. As a result, the production plan can not be carried out on schedule, the production time can not be determined, the product delivery date can not be budgeted, and the product delivery date is delayed. Therefore, the MC enterprise should meet the customer's demand quickly in the severe market competition. In order to meet the efficiency and accuracy of the establishment of man-hour quota in MC production environment, it is necessary to reconsider the method of establishing man-hour quota, and to improve the ability of enterprise manufacturing system to react to the market. According to the characteristics of manufacturing system in MC production mode and on the basis of the traditional method of establishing man-hour quota, this paper takes the parts of product family as the research object. In this paper, a new method of making man-hour quota based on parts manufacturing features is put forward. The method follows the idea of from part to whole, from concrete to abstract, and takes the manufacturing feature of parts as the criterion, and applies the techniques of part similarity judgment and neural network, etc. It improves the efficiency and accuracy of man-hour making. Firstly, by analyzing the characteristics of manufacturing system in MC environment, the concept of part manufacturing feature is put forward and used as the man-hour carrier of the new method; secondly, in order to ensure the high efficiency of the method, According to the theory of part design and manufacturing in MC production mode, the manufacturing feature model of parts is established, and the similarity judgment of parts based on manufacturing features of parts is studied. The decision focuses on the two levels of information of the part itself: the type of the part, the material of the part, the manufacturing feature model of the part; thirdly, the concept of virtual part and man-hour description platform is defined, and the construction rules of the virtual part are given. In order to ensure the accuracy of the method, a part man-hour description platform is established for each virtual part by using the table-of-things technique to describe the man-hour and its influencing factors of all the parts in each part group. 4th, in order to ensure the accuracy of the method, the neural network technology is used. The man-hour neural network structure of the parts is designed and the model is established, and the estimation of the man-hour quota of the parts using MATLAB software is studied. Finally, the concrete operation process of the new method is explained through a case study.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH16;TP391.72

【参考文献】