A Case Study on Data Analytics Based on Edge Computing for Smart Manufacturing System

Competitiveness of an electronics manufacturing services (EMS) firm is being closely monitored and interrelated in production cost control and product reliability performance level. Data analytics in massive manufacturing data can extricate huge business opportunities and values to the firms. Major challenge in data analytics application is heterogeneous and enormous of which dynamic data generated from continuously running production line reflects real-time velocity of production environment, so that the factory management demands data analytics to provide real-time solution for the improvement right on the spot. Cloud-based data analytics exhibits problems such as data capturing, storage, transfer latency and data quality that hinders the advancement of big data analytics in smart manufacturing horizon. Selection of appropriate data mining algorithm or techniques has been challenging to industry leaders in deriving desired patterns or model solving the exact problem they are facing. The aim of this research study is to illustrate the edge-based intelligent integrated information framework (INFO-I2) for the improvement of data quality in relevancy and enabling cloud-based data analytics to focus on product performance augmentation (Pipino et al. in Commun. ACM 45:211–218, 2002 [1]). In the case study, edge devices had been used for not only real-time data collection but also localized failure analysis and predictive analytics to perform autonomous decision-making in different workstations through production process. Cloud-based computing performs efficient optimization analytics for product functionality performance with those processed data which is an integrated production management system and knowledge base generated from localized data analytics of edge devices. The implementation of edge-based information framework improves the workflow management, eventually reduces manufacturing cost and improved product reliability. The contribution of this paper is to demonstrate how the proposed cloud-based manufacturing system architecture adopted both Cloud and Edge Computing to enhance product reliability and pave the way for smart manufacturing.