Group-Driven Causal Modeling and Root Cause Analysis for High-Dimensional Time Series
- 17 June 2026
- journal article
- Published by World Scientific Pub Co Pte Ltd in International Journal of Software Engineering and Knowledge Engineering
- Vol. 37 (01) , 1-22
Abstract
Root Cause Analysis (RCA) is vital for industrial monitoring and network diagnosis. In high-dimensional scenarios, traditional causal modeling is susceptible to noise and redundant dependencies, leading to unstable causal structures and poor interpretability. This paper proposes a novel method, GCRA (Group-Driven Causal Modeling and Root Cause Analysis), to address these challenges. First, the method utilizes cross-correlation to construct variable groups, establishing a group-level causal graph to effectively manage complex dependencies. Second, a mapping mechanism gradually refines this group-level graph into a variable-level structure. This structure guides the reconstruction of time-series observations, while structure-guided predictors are designed to model exogenous interventions representing unpredictable external disturbances. Finally, fine-grained RCA is achieved by calculating attribution scores based on these exogenous interventions. Experimental results across three synthetic and one real-world dataset demonstrate that GCRA accurately captures causal relationships and identifies root causes. The findings validate that group-level construction and structure-guided modeling significantly stabilize causal structures and enhance localization accuracy in high-dimensional data.Keywords
Funding Information
- Tianjin Science and Technology Plan Projects (25YD TPJC00670)
- Tianjin Science and Technology Plan Projects (25ZXRGGX00220)
This publication has 17 references indexed in Scilit:
- Neural Time-Invariant Causal Discovery from Time Series Data2023
- MPGE and RootRank: A sufficient root cause characterization and quantification framework for industrial process faultsNeural Networks, 2023
- Root cause diagnosis for process faults based on multisensor time-series causality discoveryJournal of Process Control, 2022
- Root cause diagnosis for complex industrial process faults via spatiotemporal coalescent based time series prediction and optimized Granger causalityChemometrics and Intelligent Laboratory Systems, 2022
- MST-GAT: A multimodal spatial–temporal graph attention network for time series anomaly detectionInformation Fusion, 2022
- Joint learning of multiple Granger causal networks via non-convex regularizations: Inference of group-level brain connectivityNeural Networks, 2022
- Inferring the temporal structure of directed functional connectivity in neural systems: some extensions to Granger causality2019
- SWaT: a water treatment testbed for research and training on ICS security2016
- Causal discovery on high dimensional dataApplied Intelligence, 2014
- New Introduction to Multiple Time Series Analysis2005