Optimization RFID-enabled Retail Store Management with Complex Event Processing  

作　　者：Shang-Lian Peng Ci-Jian Liu Jia He Hong-Nian Yu Fan Li 

机构地区：[1]College of Computer Science and Technology, Chengdu University of Information Technology [2]The Southwest Jiaotong University(SWJTU)-Leeds Joint School, Southwest Jiaotong University [3]School of Computer Science and Network Security, Dongguan University of Technology [4]Faculty of Science and Technology, Bournemouth University

出　　处：《International Journal of Automation and computing》2019年第1期52-64,共13页国际自动化与计算杂志（英文版）

基　　金：supported by National Social Science Fund (No. 16CTQ013);the Application Fundamental Research Foundation of Sichuan Province, China (No. 2017JY0011);the Key Project of Sichuan Provincial Department of Education, China (No. 2017GZ0333)

摘　　要：Radio frequency identification(RFID) enabled retail store management needs workflow optimization to facilitate real-time decision making. In this paper, complex event processing(CEP) based RFID-enabled retail store management is studied, particularly focusing on automated shelf replenishment decisions. We define different types of event queries to describe retailer store workflow action over the RFID data streams on multiple tagging levels(e.g., item level and container level). Non-deterministic finite automata(NFA)based evaluation models are used to detect event patterns. To manage pattern match results in the process of event detection, optimization algorithm is applied in the event model to share event detection results. A simulated RFID-enabled retail store is used to verify the effectiveness of the method, experiment results show that the algorithm is effective and could optimize retail store management workflow.Radio frequency identification(RFID) enabled retail store management needs workflow optimization to facilitate real-time decision making. In this paper, complex event processing(CEP) based RFID-enabled retail store management is studied, particularly focusing on automated shelf replenishment decisions. We define different types of event queries to describe retailer store workflow action over the RFID data streams on multiple tagging levels(e.g., item level and container level). Non-deterministic finite automata(NFA)based evaluation models are used to detect event patterns. To manage pattern match results in the process of event detection, optimization algorithm is applied in the event model to share event detection results. A simulated RFID-enabled retail store is used to verify the effectiveness of the method, experiment results show that the algorithm is effective and could optimize retail store management workflow.

关 键 词：Complex event processing(CEP) radio frequency identification(RFID) Internet of THINGS data STREAM supply CHAIN RETAIL STORE 

分 类 号：TP[自动化与计算机技术]