A Research On Work-in-Process State Trajectories Of Discrete Workshop Based On Real-time Manufacturing Data

The aggressive application of Internet of Things leads to huge amount of precious manufacturing data in the manufacturing environment,which lays a strong foundation for the research of complex manufacturing system.As the core factor of manufacturing system,the production process of Work-In-Process is the center of task dispatching.The manufacturing data of Work-In-Process provides valuable logic knowledge for the description and explanation of complex manufacturing system,which lays a foundation for production state prediction.This paper focuses on Work-In-Process state trajectories in discrete manufacturing shop floor and emphasizes on the manufacturing system knowledge in the collected data.A data model for Work-InProcess data management is proposed,based on which algorithms are proposed for Work-In-Process state trajectory pattern mining and state prediction.The contents mainly are:(1)Considering the knowledge in manufacturing data,this paper utilizes semantic coordinates and logic trajectories to describe Work-In-Process state chains and to achieve a higher abstraction of production states.(2)This paper analyzes Work-In-Process production state chains and proposes an algorithm for Work-In-Process state trajectory pattern mining based on the semantic coordinates.(3)Work-In-Process state trajectories provide logic knowledge for state prediction.This paper proposes an algorithm for Work-In-Process state prediction based on the trajectory patterns.(4)This paper introduces a fusion framework of physical and logical spaces for information interchange and data push.Moreover,this paper evaluates and examines the performance of the proposed approach.