Supply Forecast Service in Cyber Physical Production Systems: A Case Study of Food Business

Abstract

The global demand for food value chains is expanding as the world population grows. It necessitates that food manufacturers streamline their supply chain to satisfy shorter product life cycles, also known as horizontal integration in an Industry 4.0 perspective. On the other hand, a vertical integration from farm production to supply chain planning involves constantly communicating their performance status and responding autonomously via Cyber-Physical Systems (CPS) concept for dynamic production requirements. The design and implementation of three layers of CPSs is proposed including farm CPS, farm Cyber-Physical Production Systems (CPPS) and Cyber-Physical Supply Chain Planning Systems (CPS2). The farm CPS was managed on the OSIsoft-PI platform, which is based on service-oriented architecture (SOA) and then integrated into the Enterprise Cloud (Microsoft Azure). The CPPS was implemented with data integration, which include farm management such as biosafety, quality control, maintenance management and inventory management. The prediction service in CPPS was also implemented on Azure machine learning service, using data from corporate IT and real-time performance from farm CPS. The case of supply forecast, weekly-monthly pig body weight forecast, was studied. The results reveal that the prediction model can forecast pig body weight and body size ahead from the beginning to the end of the crop farms, separated by seasonal. Finally, the cyber physical system in supply chain planning (CPS2) will be collecting the supply forecast information from all sites and then transform into supply information in corporate level.