Industrial-Grade Time-Dependent Counterfactual Root Cause Analysis through the Unanticipated Point of Incipient Failure: a Proof of Concept
This addresses root cause analysis for industrial systems, but it is incremental as it builds on existing causal technology with a proof-of-concept simulation.
The paper tackles the problem of identifying root causes of failures in industrial multivariate time series by focusing on the Point of Incipient Failure, where anomalies first appear, and demonstrates the approach experimentally in a simulated setting.
This paper describes the development of a counterfactual Root Cause Analysis diagnosis approach for an industrial multivariate time series environment. It drives the attention toward the Point of Incipient Failure, which is the moment in time when the anomalous behavior is first observed, and where the root cause is assumed to be found before the issue propagates. The paper presents the elementary but essential concepts of the solution and illustrates them experimentally on a simulated setting. Finally, it discusses avenues of improvement for the maturity of the causal technology to meet the robustness challenges of increasingly complex environments in the industry.