A Novel Two-level Causal Inference Framework for On-road Vehicle Quality Issues Diagnosis
This addresses a specific problem for the automotive industry by providing a systematic method to reduce investigation times from weeks, though it appears incremental as it builds on existing causal ML techniques.
The paper tackles the slow process of diagnosing on-road vehicle quality issues by proposing a two-level causal inference framework, which leverages causal machine learning to speed up root cause isolation and treatment evaluation.
In the automotive industry, the full cycle of managing in-use vehicle quality issues can take weeks to investigate. The process involves isolating root causes, defining and implementing appropriate treatments, and refining treatments if needed. The main pain-point is the lack of a systematic method to identify causal relationships, evaluate treatment effectiveness, and direct the next actionable treatment if the current treatment was deemed ineffective. This paper will show how we leverage causal Machine Learning (ML) to speed up such processes. A real-word data set collected from on-road vehicles will be used to demonstrate the proposed framework. Open challenges for vehicle quality applications will also be discussed.