Predictive Fail-Safe Improving the Safety of Industrial Environments through Model-based Analytics on hidden Data Sources
This paper explores how the functional safety of industrial deployments can be improved through emerging Industrie 4.0 approaches. We discuss how new sources of data, that are becoming accessible through advancing digitalization, can be used for this purpose, and how principles from predictive maintenance systems can be applied to industrial fail-safe applications: based on data from the industrial components themselves and from their environment as well as on metadata about interactions between these systems and people, we propose to create a model-based monitoring and controlling system that focuses on preserving the functional safety of the installation as a whole. We expect such a Predictive Fail-Safe system to mitigate or even prevent unsafe consequences of failures even in highly dynamic “smart factories”, thereby reducing or preventing harm to other equipment, the environment, and the involved people.