"Where is My Troubleshooting Procedure?": Studying the Potential of RAG in Assisting Failure Resolution of Large Cyber-Physical System
This program is tentative and subject to change.
In today’s complex industrial environments, operators must often navigate through extensive technical manuals to identify troubleshooting procedures relevant to some observed failure symptoms. These manuals, written in natural language, describe many steps in detail. Unfortunately, the number, magnitude, and articulation of these descriptions can significantly slow down and complicate the retrieval of the correct procedure during critical incidents. Interestingly, Retrieval Augmented Generation (RAG) enables the development of tools based on conversational interfaces that can assist operators in their retrieval tasks, improving their capability to respond to incidents.
This paper presents the results of a set of experiments that derive from the analysis of the troubleshooting procedures available in Fincantieri, a large international company developing complex naval cyber-physical systems. Results show that RAG can assist operators in reacting promptly to failure symptoms, although specific measures have to be taken into consideration to cross-validate recommendations using the sources used to obtain them.
This program is tentative and subject to change.
Thu 16 AprDisplayed time zone: Brasilia, Distrito Federal, Brazil change
14:00 - 15:30 | Dependability and Security 6SE In Practice (SEIP) / Research Track / SE in Society (SEIS) at Oceania X Chair(s): Elizabeth Dinellla | ||
14:00 15mTalk | "Where is My Troubleshooting Procedure?": Studying the Potential of RAG in Assisting Failure Resolution of Large Cyber-Physical System SE In Practice (SEIP) Maria Teresa Rossi University of Milano Bicocca, Italy, Leonardo Mariani University of Milano-Bicocca, Oliviero Riganelli University of Milano - Bicocca, Giuseppe Filomento University of Milano - Bicocca, Danilo Giannone University of Milano - Bicocca, Paolo Gavazzo University of Milano - Bicocca Pre-print | ||
14:15 15mTalk | Deploying SafeKAN for Anomaly Detection in Safety-Critical Satellite Operations: An Industry-Guided Study SE In Practice (SEIP) Alberto Petrucci Gran Sasso Science Institute (GSSI), Francesco Basciani Gran Sasso Science Institute (GSSI), Franco Raimondi Gran Sasso Science Institute (GSSI), Patrizio Pelliccione Gran Sasso Science Institute, L'Aquila, Italy | ||
14:30 15mTalk | FoundRoot: Towards Foundation Model for Root Cause Analysis via Structured Deep Thinking Research Track Zhe Xie Tsinghua University, Zeyan Li ByteDance, Xiao He Bytedance, Shenglin Zhang Nankai University, Longlong Xu Tsinghua University, Yuzhuo Yang Tsinghua University, Tieying Zhang ByteDance, Jianjun Chen Bytedance, Rui Shi Bytedance, Dan Pei Tsinghua University | ||
14:45 15mTalk | An Ontology-Based Approach to Security Risk Identification for Container Deployments in OT Contexts SE In Practice (SEIP) Yannick Landeck fortiss GmbH, Dian Balta fortiss GmbH, Martin Wimmer Siemens AG, Christian Knierim Siemens AG DOI Pre-print Media Attached | ||
15:00 15mTalk | PCICF: A Pedestrian Crossing Identification and Classification Framework SE In Practice (SEIP) Junyi Gu Chalmers University of Technology and University of Gothenburg, Beatriz Cabrero-Daniel University of Gothenburg, Ali Nouri Volvo cars & Chalmers University of Technology, Lydia Armini Chalmers University of Technology and University of Gothenburg, Christian Berger Chalmers University of Technology, Sweden | ||
15:15 15mTalk | Engineering Future Critical CPSs with Trustworthy GenAI Across the Lifecycle SE in Society (SEIS) Alessio Bucaioni Mälardalen University, Antonio Cicchetti Mälardalen University, Gordana Dodig Crnkovic Mälardalen University, Romina Spalazzese Malmö University, Emma Söderberg Lund University, Daniel Varro Linköping University / McGill University | ||