Exploring Interactions among EPBM Features using Bayesian Networks

Abstract

This paper presents an implementation of Bayesian network (BN) and structure learning algorithm to explore the interactions and causal relationships among excavation features of an earth pressure balance tunnel boring machine (EPBM). This study was performed on a data set from the State Route 99 (SR99) tunnel construction in Seattle, WA. A score-based structure learning algorithm was used to construct the BN graphs. The effects of providing explicit information about geologic conditions in the interactions were evaluated. This study demonstrates that BN graphs could systematically model the interactions among EPBM features in a compact and interpretable representation. The score-based algorithm could successfully capture several true and sensible mechanisms of the feature interactions based on data. Furthermore, their dependencies might indicate how the operators controlled the EPBM.