DMBSI-CARE: A Causal Process-Mining Framework for Root-Cause Analysis and Bottleneck Attribution in Wafer Fabrication
Authors
Mohammad Sharifur Rahman1, *, M. M. Manjurul Islam1, Karl McCreadie1, Saugat Bhattacharyya1, Cormac McAteer2, Bryan John Baker3, Nuala Parker2, Girijesh Prasad1
1Intelligent Systems Research Centre, Ulster University, Londonderry, BT48 7JL, UK
2Seagate Technology, Londonderry, BT48 0BF, UK
3Seagate Technology, One Disc Dr, Londonderry, MN, 55435, USA
*Corresponding author.
Email: rahman-m11@ulster.ac.uk
Corresponding Author
Mohammad Sharifur Rahman
Available Online 29 October 2026.
- DOI
- 10.2991/978-94-6239-813-9_13How to use a DOI?
- Keywords
- root-cause analysis; causal inference; process mining; cycle time; semiconductor manufacturing
- ABSTRACT
Cycle-time root-cause analysis in wafer fabrication still relies on descriptive severity rankings, yet severity does not equate to causal impact, does not reveal how delays propagate, and cannot explain why a particular lot was slow. DMBSI-CARE closes these gaps by seeding a structural causal model with a process-mining Directly-Follows Graph, estimating per-step effects with double machine learning, decomposing each effect through severity-aligned mediators, and aggregating them into a per-lot causal cycle-time bill of materials. On a log of 1,406 lots and 7.9 million events, it surfaces leverage-step severity-ranking misses and attributes 73–99% of the indirect effect to processing variability rather than queueing.
- Copyright
- © 2026 The Author(s)
- Open Access
- Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
Cite this article
TY - CONF AU - Mohammad Sharifur Rahman AU - M. M. Manjurul Islam AU - Karl McCreadie AU - Saugat Bhattacharyya AU - Cormac McAteer AU - Bryan John Baker AU - Nuala Parker AU - Girijesh Prasad PY - 2026 DA - 2026/10/29 TI - DMBSI-CARE: A Causal Process-Mining Framework for Root-Cause Analysis and Bottleneck Attribution in Wafer Fabrication BT - Proceedings of the Global Symposium on Adaptive Manufacturing (GSAM 2026) PB - Atlantis Press SP - 37 EP - 39 SN - 2352-5401 UR - https://doi.org/10.2991/978-94-6239-813-9_13 DO - 10.2991/978-94-6239-813-9_13 ID - Rahman2026 ER -