The industry digitization is transforming its business models, organizational structures and operations, mainly promoted by the advances and the mass utilization of smart methods, devices and products, being leveraged by initiatives like Industrie 4.0. In this context, the data is a valuable asset that can support the smart factory features through the use of Big Data and advanced analytics approaches. In order to address such requirements and related challenges, Cyber Physical Systems (CPS) promote the development of more intelligent, adaptable and responsiveness supervisory and control systems capable to overcome the inherent complexity and dynamics of industrial environments. In this context, this work presents an agent-based industrial CPS, where agents are endowed with data analysis capabilities for distributed, collaborative and adaptive process supervision and control. Additionally, to address the different industrial levels’ requirements, this work combines two main data analysis scopes: at operational level, applying distributed data stream analysis for rapid response monitoring and control, and at supervisory level, applying big data analysis for decision-making, planning and optimization. Some experiments have been performed in the context of an electric micro grid where agents were able to perform distributed data analysis to predict the renewable energy production.
