大数据下纺织制造执行系统关键问题研究-南京大学PASA大数据实验室.PDF

大数据下纺织制造执行系统关键问题研究 1,2 2 2 2 1 2 2 邵景峰 ，贺兴时 ，王进富 ，白晓波 ，刘晟 ，马创涛 ，刘聪颖 （1．长安大学信息工程学院，西安 710064 ； 2 ．西安工程大学管理学院，西安 710048 ） 摘要：为解决纺织制造执行系统设计过程中的数据融合问题，首先对纺织制造过程中产生的海量数据及其 相关性进行分析，并对纺织工艺流程进行了优化设计，然后，利用多色集合理论对纺织制造过程中产生的 海量数据进行了统一形式化表达，并针对异构纺织监测系统数据库间难以融合的问题，利用D-S 证据提出 了一种采用两级传感器信息融合的方法。最后，通过实验验证，结果表明：通过对制造过程数据的统一 描述，提出的信息融合方法有效解决了海量纺织数据的融合问题，实现了计划层与制造层之间信息的有效 衔接，有利于为构建大数据环境下的纺织制造执行系统提供技术支撑。 关键词：制造执行系统；纺织企业；数据融合；大数据；D-S证据 中图分类号： TS103.2 文献标识码：A Research on Key Problems of Textile Manufacturing Execution System Based on Big Data 1,2 2 2 2 2 1 2 Shao Jingfeng , He Xingshi , Wang Jinfu , Bai Xiaobo , Liu Sheng , Ma Chuangtao , Liu Congying (1. School of Information Engineering, Chang’an University, Xi’an 710064; 2. School of Management, Xi’an Polytechnic University, Xi’an 710048) Abstract: To solve the problem of data fusion in textile manufacturing execution system design process, firstly, the massive data which are from the textile manufacturing process and theirs correlations were analyzed, and the textile technology process was optimized. Then, massive amounts of data generated from the textile manufacturing process is formally expressed by the polychromatic sets theory, and aiming at difficult integration problem of the heterogeneous textile monitoring system databases, a two-level sensor information fusion method was proposed via D-S evidence. Finally, as verified in the experiment, the results have shown that, through a unified description of the manufacturing process data, the proposed information fusion method can effectively solve the problem of massive textile data fusion, realize the effective link between planning layer and manufacturing infor