面向企业生产集成系统的物料推荐系统

摘要/Abstract

摘要：

针对目前生产制造信息平台因多系统集成带来的物品检索量过大、信息冗余、人工筛选耗时、不支持近似查询等问题，提出一个可集成于当前工业信息系统的“物品-用户”组合推荐模型。该推荐模型组合了基于内容推荐和基于协同过滤推荐。在推荐项方面，考虑了推荐项的具体形态，全面分析物料清单 (含零部件、半成品、产品)所包含的特征，提出基于内容相似、物料清单结构相似及两者集成的相似距离定义方法，并按物品的不同选择不同的方法来定义物品间的近似距离，同时以距离为基准划分主题或分类。对于模型中的用户，主要针对生产系统特点来量化用户兴趣、定义用户近似公式。最后，以系统套件的方式实现了推荐平台的建立，并引入最新的研究成果解决了系统冷启动问题。实验表明，物料推荐平台有效填充了现有信息系统的空缺功能，能为用户准确推荐满足其需求的物品，提高信息获取的效率。

关键词: 相似性度量, 物料清单, 信息系统, 推荐系统

Abstract: Aiming at providing effect informations efficiently for appropriate users in the integration systems of enterprise manufacturing who was faced excessive search volume, information overload, time-consuming manual filtering, and no approximate query support problems, a product-user recommendation model was proposed and a recommendation system available for integrating was implemented. This recommendation model combined contents based on filtering and collaborative filtering. In the aspect of items, the specific forms of the recommended items were considered, all the characteristics of the BOM including parts, semi-finished products, products were comprehensively analyzed, and new metric methods were proposed based on content similarity, BOM-based similarity and the integration. Different items selected different methods to define the approximate distances which were the foundation of subject clustering or classification according to the specific circumstances. While in the aspect of users, the user interests were quantified and the user approximation formula were defined according to the characteristics of the production systems. Finally, this paper implemented the recommendation platform in the form of a system suite, considering cold start problem by introducing the latest research results. Experiments show that the system may accurately recommend the items that meet their needs for users and it is an important supplement for current manufacturing systems for obtaining information efficiently.

中图分类号:

TP391.7

吴文李1;范小朋2;周庚申1;黄羿1;曹旸1;林桂婵1. 面向企业生产集成系统的物料推荐系统[J]. 中国机械工程.

WU Wenli1;FAN Xiaopeng2;ZHOU Gengshen1;HUANG Yi1;CAO Yang1;LIN Guichan1. A Material Recommendation System for Enterprise Manufacturing Integrated Systems[J]. China Mechanical Engineering.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.cmemo.org.cn/CN/

http://www.cmemo.org.cn/CN/Y2019/V30/I15/1856

参考文献

[1]HAO Jia, YAN Yan, WANG Guoxin, et al. A Probability-based Hybrid User Model for Recommendation System[J]. Hindawi Publishing Corporation Mathematical Problems in Engineering, 2016,1:1-10.
[2]WILMANN R, KASTN W. Product Ramp-up for Semiconductor Manufacturing Automated Recommendation of Control System Setup[M]. Lübeck:RonPub UG, 2016:219-254.
[3]刘庭煜,汪惠芬,贲可存,等. 基于多维情境本体匹配的产品开发过程业务产物智能推荐技术[J]. 计算机集成制造系统, 2016, 22(12): 2727-2750.
LIU Tingyu, WANG Huifen. BEN Kecun, et al. Intelligent Business Artifact Recommending Approach for Product Development Processes Based on Context Ontology Matching[J]. Computer Integrated Manufacturing Systems, 2016, 22(12): 2727-2750.
[4]FAJAR J E, MARTA S, ESTEFANIA S, et al. Ontology-based Data Integration in Multi-disciplinary Engineering Environments[J].Journal of Information Systems, 2017,4(1):1-26.
[5]YEHUDA K, ROBERT B. Advances in Collaborative Filtering Recommender Systems Handbook [M]. Boston:Springer, 2017:77-118.
[6]SON J, KIM S B. Content-based Filtering for Recommendation Systems Using Multiattribute Networks[J]. Expert Systems with Application, 2017,89(15):404-412.
[7]BLERINA L, KOSTAS K, STATHES H. Facing the Cold Start Problem in Recommender Systems [J]. Expert Systems with Applications， 2017, 41(2014):2065-2073.
[8]JYOTIRMOY G, SANJAY K J. A Survey on Solving Cold Start Problem in Recommender Systems [C]∥International Conference on Computing, Communication and Automation. Greater Noida,2017：1-13.
[9]WEI Jian, HE Jianhua, CHEN Kai, et al. Collaborative Filtering and Deep Learning Based Recommendation System for Cold Start Items[J]. Expert Systems with Applications. 2017, 69:29-39.
[10]SHIH H M. Product Structure BOM-based Product Similarity Measures Using Orthogonal Procrustes Approach[J]. Computers & Industrial Engineering, 2013, 61(3):608-628.
[11]GENTE V, KENNY G, MEYLER A, et al. Combining Expert Forecasts: Can Anything Beat the Simple Average?[J].International Journal of Forecasting, 2013，29(1):108-121.
[12]AMITA M, KALYANI K, POOJA K. Prediction of Heart Disease Using K-Means and Artificial Neural Network as Hybrid Approach to Improve accuracy[J]. International Journal of Engineering and Technology, 2017, 9(4): 3081- 3085.
[13]GENG Zhiqiang, LI Yanan, HAN Yongming, et al. A Novel Self-organizing Cosine Similarity Learning Network[J]. Energy, 2018: 142(1): 400-410.
[14]SIMOPOULOS C, WERETILNYK E, GOLDING G B. Prediction of Plant IncRNA by Ensemble Machine Learning Classifiers[J]. BMC Genomics, 2018,19(1):316.
[15]吴旭,张雄. 一种融合云模型和新人的推荐算法[J].信息技术,2016,2552(5):140-146.
WU Xu, ZHANG Xiong. Recommendation Algorithm Integrating Cloud Model and Trust[J]. Information Technology, 2016, 2552(5):140- 146.
[16]雒兴刚,杨勇.基于联合分析的可配置产品优化选择方法[J]. 中国机械工程, 2012, 23(5): 578-584.
LUO Xingang, YANG Yong. Optimization Method for Configurable Products Based on Conjoint Analysis[J]. China Mechanical Engineering, 2012, 23〖HJ1.65mm〗(5): 578-584.
[17]ISRAT C, RICHI N. Identifying Product Families Using Data Mining Techniques in Manufacturing[C]∥Proceedings of the Twelfth Australasian Data Mining Conference. Paradigm， 2014：113-120.