A framework for shopfloor material delivery based on real-time manufacturing big data

Although auto-identification devices such as radio frequency identification and smart sensors have been widely used in shopfloor management and control, major challenges still hinder the vision of real-time and multi-source data-driven material delivery in a manufacturing big data environment. For instance, how to collect manufacturing big data in a timely and accurate manner, and how to discover the hidden pattern from the manufacturing big data rapidly to improve the efficiency of material delivery. To address these challenges, in this paper, a framework for shopfloor material delivery based on real-time manufacturing big data is proposed. Key technologies of the proposed framework are investigated. Firstly, a solution of data sensing and acquisition is designed. Secondly, the methods of manufacturing big data preprocessing and storage are developed to integrate and share the manufacturing data in a unified data format, and to ensure the reusability of the data. Thirdly, a graphic model for the manufacturing big data mining is presented. An improved Apriori-based association analysis model is exploited to identify the frequency trajectories of material delivery. In order to demonstrate the implementation of the proposed framework, a proof-of-concept scenario is designed. The key findings and insights from the experimental results are summarized as managerial implications, which can guide manufacturers to make more informed decisions for the shopfloor management.