| Resumo: | The ongoing 4th industrial revolution is characterized by the digitization of industrial environments, mainly based on the use of Internet of Things, Cloud Computing and Artificial Intelligence (AI). Regarding AI, although data analysis has shown to be a key enabler of industrial Cyber-Physical Systems (CPS) in the development of smart machines and products, the traditional Cloud-centric solutions are not suitable to attend the data and time-sensitive requirements. Complementary to Cloud, Edge Computing has been adopted to enable the data processing capabilities at or close to the physical components. However, defining which data analysis tasks should be deployed on Cloud and Edge computational layers is not straightforward. This work proposes a framework to guide engineers during the design phase to determine the best way to distribute the data analysis capabilities among computational layers, contributing for a lesser ad-hoc design of distributed data analysis in industrial CPS. Besides defining the guidelines to identify the data analysis requirements, the core contribution relies on a Fuzzy Logic recommendation system for suggesting the most suitable layer to deploy a given data analysis task. The proposed approach is validated in a smart machine testbed that requires the implementation of different data analysis tasks for its operation.
|
|---|