Detection Of Abnormal Event In Manufacturing Workshop Based On RFID

There are many uncertain factors in the production process of the manufacturing enterprise, which may lead to unexpected anomalies in personnel, materials, equipment, etc. If not timely detection and treatment of these abnormal events will affect the production schedule for business losses. With the development of information technology, Radio frequency identification (RFID) technology as a kind of automatic identification technology, because its non-contact identification, fast read and write, can also identify multiple targets and other advantages, is more and more application in data collection and management in logistics, supply chain, transportation and other fields. How to combine the RFID technology to the management of the workshop production site,accurate and efficient transformation of massive RFID data into the event information for enterprise decision making, to improve the response ability of the enterprise to the workshop production abnormal event has become a question which is worth studying.On the basis of analyzing the RFID technology principle, the data characteristic and the workshop production process, this thesis studies the method for detecting abnormal event in workshop production based on RFID. In the research of the RFID simple event general detection model in the manufacturing workshop, three types of simple events are defined, such as Appearance event, Deletion Event, and Continued Event. And the method of obtaining these simple events from RFID data is analyzed. On the basis of summarizing the research of the logic rules of complex events, the formal method of complex event representation is constructed to deal with the complex and changeable business logic requirement in the practical application. Then this paper discusses how to effectively deal with the massive RFID data in the practical application, discusses the idea of using Rete algorithm based on MapReduce framework to extract the RFID complex events from the massive RFID data.