An interactive tool for causal graph modeling for supervision purposes

This paper describes a software tool for causal graph model construction. The aim of a causal model is to capture the influences between variables of a continuous process expressed as qualitative and quantitative knowledge. This model can be used for different purposes related with supervision, such as fault detection, fault isolation and action advising for fault handling. The application of this tool to a four-tank system is briefly explained and its integration into MAGIC, a Multi-Agent based Supervisory System is also described.