基于特征与智能计算的工艺成本估算研究

发布时间：2018-02-24 13:40

本文关键词： 工艺成本特征智能计算工序成本估算　出处：《大连交通大学》2012年硕士论文　论文类型：学位论文

【摘要】：现代企业正在经历激烈的市场竞争，市场需求的不断变化以及技术创新、世界贸易和经济全球化的发展，使产品具有越来越短的生命周期和越来越快的研发速度。进行成本估算与控制，能有效减少人力、物力和财力上不必要的浪费，提高企业甚至整个制造业的技术水平。本文从成本精细化管理的角度，结合我国企业发展时间短，基础数据、人力、物力和财力均不充足的实际情况，利用成本会计学理论、企业资源计划理论和计算机技术，以机车车轴和主、从动齿轮为例，提出基于特征与智能计算的工艺成本估算方法。相对快速准确的估算产品成本，对提高企业的竞争力，改善企业生产环境具有深远的意义。文章首先论述了课题的研究背景、目的及意义，，成本计算的研究现状及其存在的问题等。接下来论述了工艺成本估算的基本定义和基于工序的工艺成本估算的计算流程，介绍了智能计算的方法、特点和产品特征的定义，阐述了基于特征与智能计算的工艺成本估算流程、成本特征模型、模糊集合、模糊隶属函数和模糊关系的表示方法、粗集理论及其基本思想和神经网络的基本模型。通过对基于工序的工艺成本估算方法进行了研究，用以作为基于特征与智能计算的工艺成本估算仿真和预测的依据。然后建立了特征—工艺成本模型，应用模糊理论的方法，消除工艺成本特征模型的模糊性。对工艺成本特征模型中不完整、不精确的数据预处理，进行特征属性的重要性分析及化简并用MATLAB实现程序化。最后总结了应用MBTAB—BP神经网络编程时容易遇到的问题。通过MATLAB神经网络编程实现模型的仿真和预测，并对不同参数的BP网络程序进行比较，最终使得到的预测结果能够达到较高的精度，满足工艺成本估算的需要。
[Abstract]:Modern enterprises are experiencing fierce market competition, changing market demand, technological innovation, the development of world trade and economic globalization. The cost estimation and control can effectively reduce the unnecessary waste of manpower, material and financial resources, and improve the technological level of the enterprise and even the whole manufacturing industry. In this paper, from the angle of cost fine management, combined with the fact that the development time of our country's enterprises is short, the basic data, manpower, material resources and financial resources are not sufficient, we make use of the theory of cost accounting, the theory of enterprise resource planning and the computer technology. Taking locomotive axles and main and slave gears as examples, a process cost estimation method based on feature and intelligent calculation is put forward. It is of great significance to estimate product cost quickly and accurately to improve the competitiveness of enterprises and improve the production environment of enterprises. This paper first discusses the research background, purpose and significance of the subject, the current research situation and existing problems of cost calculation, and then discusses the basic definition of process cost estimation and the calculation flow of process cost estimation based on working procedure. This paper introduces the method, characteristics and definition of product features of intelligent computing, and expounds the process flow of process cost estimation, cost feature model, fuzzy set, fuzzy membership function and fuzzy relation representation method based on feature and intelligent calculation. The rough set theory, its basic idea and the basic model of neural network are studied, and the process cost estimation method based on working procedure is studied. It is used as the basis for the simulation and prediction of process cost estimation based on feature and intelligent computing. Then the feature-process cost model is established and the fuzzy theory method is applied. Eliminate the fuzziness of the process cost feature model. Pretreatment of incomplete and inaccurate data in the process cost feature model, The importance of feature attributes is analyzed and simplified and programmed with MATLAB. Finally, the problems encountered in programming with MBTAB-BP neural network are summarized. The simulation and prediction of the model are realized by MATLAB neural network programming. By comparing the BP network program with different parameters, the predicted results can achieve high accuracy and meet the needs of process cost estimation.
【学位授予单位】：大连交通大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH16;F406.72

【参考文献】