某仪表制造业通用零件生产预测系统的研究与实现

发布时间：2018-07-12 10:48

本文选题：离散制造业 + 预测　；参考：《宁夏大学》2017年硕士论文

【摘要】：为了实现对企业的信息进行动态管理,企业一般采取高级计划排程(Advanced Planning and Scheduling)方法。这是一种可将时间、订单、库存、预测产量等生产中的重要因素考虑进去,在企业生产过程中随时获取各种动态变化从而调整生产去迎合市场变化的方法。它解决了企业产能和资源平衡的问题,为离散制造业各种资源的高速流通带来了极大的便利。在预测中使用合理的数学模型来预测零件产量能给企业的生产计划带来极大的帮助,但是企业的生产预测却具有复杂、多层次这些特点,这些特点给建模带来了层层困难。本文的研究主体是基于宁夏吴忠仪表责任有限公司(简称吴忠仪表)的,该企业是一家离散型的阀门制造企业,该企业生产的产品的特点是品种多样、批量少、批次繁多。所以在满足客户需求的情况下实现企业资源合理的利用,就必须采取更加有效的生产组织方式。为了完善企业自身已有的预测系统,提高预测能力,企业需要建立一套新的预测模式。通过分析企业的历史生产数据,并对数据进行清洗然后根据数据的特点选择合适的算法来构建预测模型。在算法上本论文提出了一种用遗传算法优化的支持向量回归机来对数据进行预测,在预测模式上提出了两种预测模式,通过算法和模式的组合来更加准确的对企业的通用零件的产量进行预测。然后使用MATLAB对构建的预测模型进行仿真模拟验证模型的准确性,最后通过编程实现了预测系统,有效的实现了对企业通用零件的预测。企业通用零件预测系统的实现具有以下两方面的意义:第一,提高了对订单的处理能力和响应能力;第二,提高了该企业的市场竞争力。
[Abstract]:In order to realize the dynamic management of enterprise information, enterprises usually adopt the method of Advanced planning and scheduling. This is a method that can take into account the important factors in production such as time, order, inventory, and forecast output, and obtain all kinds of dynamic changes at any time in the production process of the enterprise to adjust the production to meet the market changes. It solves the problem of enterprise capacity and resource balance, and brings great convenience to the rapid circulation of various resources in discrete manufacturing industry. The use of reasonable mathematical model to predict the production of parts can bring great help to the production plan of the enterprise, but the production forecast of the enterprise has the characteristics of complex and multi-level, which brings many difficulties to the modeling. The main body of this paper is based on Ningxia Wu Zhong instrument liability Co., Ltd. (Wu Zhong instrument), the enterprise is a discrete valve manufacturing enterprise, the characteristics of the products produced by the enterprise is a variety of products, fewer batches and lots of batches. Therefore, it is necessary to adopt more effective production organization mode to realize the rational utilization of enterprise resources under the condition of satisfying customer demand. In order to perfect the existing forecasting system and improve the forecasting ability, enterprises need to establish a new forecasting model. By analyzing the historical production data of the enterprise and cleaning the data, the prediction model is constructed by selecting the appropriate algorithm according to the characteristics of the data. In this paper, a support vector regression machine optimized by genetic algorithm is proposed to predict the data, and two prediction models are proposed in this paper. Through the combination of algorithms and patterns to more accurately predict the production of common parts of the enterprise. Then MATLAB is used to simulate and verify the veracity of the model. Finally, the prediction system is realized by programming, which effectively realizes the prediction of the common parts of the enterprise. The realization of the general part prediction system has the following two meanings: firstly, the ability to deal with and respond to the orders is improved; secondly, the market competitiveness of the enterprise is improved.
【学位授予单位】：宁夏大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F426.46;TP311.52

【参考文献】