{"id":{"repo_id":"auckland-ms","oai_identifier":"oai:researchspace.auckland.ac.nz:2292/71212"},"canonical_url":"https://search.dev.ndltd.org/etd/auckland-ms/oai:researchspace.auckland.ac.nz:2292/71212","repository":{"repo_id":"auckland-ms","name":"University of Auckland","base_url":"https://researchspace.auckland.ac.nz/server/oai/request"},"display":{"title":"Intelligent Industrial Maintenance: Using Natural Language Processing Technology and Domain Knowledge for Decision Support","abstract":"Intelligent maintenance system (IMS) is crucial for optimizing maintenance strategies, reducing downtime, and improving overall equipment effectiveness in modern manufacturing systems. However, existing maintenance systems often struggle to effectively integrate and analyse multi-modal and heterogeneous data sources, leading to suboptimal maintenance decisions. This thesis presents a comprehensive and integrated framework for intelligent maintenance system that leverages advanced natural language processing techniques, particularly large language models (LLMs), and domain-specific knowledge to provide decision support on predictive maintenance, fault diagnosis, and maintenance task planning. The proposed framework consists of three key components: 1) a multi-modal fusion methodology for integrating sensor data and event logs to improve the accuracy of predictive maintenance and fault diagnosis models; 2) a task-centric knowledge graph (TCKG) schema and extraction framework for capturing maintenance knowledge from unstructured documents; and 3) an intelligent question-answering system that integrates LLMs and TCKGs to provide accurate, context-aware answers to maintenance queries and support task planning. The effectiveness and potential impact of the proposed framework are demonstrated through extensive experiments and case studies using real-world maintenance datasets. The results show substantial improvements in fault prediction accuracy, maintenance knowledge extraction, and question-answering performance compared to state-of-the-art baselines. The thesis also discusses the limitations of the current work and identifies future research directions, such as the need for explainable AI techniques, industry-specific ontologies, and the integration with other Industry 4.0 technologies. This thesis contributes to the field of intelligent maintenance system by proposing new methods for data integration, knowledge representation, and intelligent question answering for maintenance task planning. The framework's modular architecture allows for seamless integration with existing systems and adaptation to evolving industry needs. The research outcomes have the potential to drive improvements in operational efficiency, cost reduction, and asset performance, aligning with Industry 4.0 objectives.","abstract_html":"Intelligent maintenance system (IMS) is crucial for optimizing maintenance strategies, reducing downtime, and improving overall equipment effectiveness in modern manufacturing systems. However, existing maintenance systems often struggle to effectively integrate and analyse multi-modal and heterogeneous data sources, leading to suboptimal maintenance decisions. This thesis presents a comprehensive and integrated framework for intelligent maintenance system that leverages advanced natural language processing techniques, particularly large language models (LLMs), and domain-specific knowledge to provide decision support on predictive maintenance, fault diagnosis, and maintenance task planning. The proposed framework consists of three key components: 1) a multi-modal fusion methodology for integrating sensor data and event logs to improve the accuracy of predictive maintenance and fault diagnosis models; 2) a task-centric knowledge graph (TCKG) schema and extraction framework for capturing maintenance knowledge from unstructured documents; and 3) an intelligent question-answering system that integrates LLMs and TCKGs to provide accurate, context-aware answers to maintenance queries and support task planning. The effectiveness and potential impact of the proposed framework are demonstrated through extensive experiments and case studies using real-world maintenance datasets. The results show substantial improvements in fault prediction accuracy, maintenance knowledge extraction, and question-answering performance compared to state-of-the-art baselines. The thesis also discusses the limitations of the current work and identifies future research directions, such as the need for explainable AI techniques, industry-specific ontologies, and the integration with other Industry 4.0 technologies. This thesis contributes to the field of intelligent maintenance system by proposing new methods for data integration, knowledge representation, and intelligent question answering for maintenance task planning. The framework's modular architecture allows for seamless integration with existing systems and adaptation to evolving industry needs. The research outcomes have the potential to drive improvements in operational efficiency, cost reduction, and asset performance, aligning with Industry 4.0 objectives.","abstract_has_math":false,"creators":["Liu, Zengkun"],"institution":"ResearchSpace@Auckland","degree_name":"PhD","degree_level":"Doctoral","degree_discipline":"Engineering","degree_department":null,"school":null,"contributors":[],"advisors":["Aw, Kean","Hui, Justine"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-07-24T01:05:30Z","subjects":["Intelligent maintenance automation"],"languages":[],"rights":["Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated."],"rights_urls":["https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2292/71212","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Aw, Kean","Hui, Justine"]},{"key":"dc:creator","label":"Author","values":["Liu, Zengkun"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-02-06T19:28:07Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-02-06T19:28:07Z"]},{"key":"dc:date.issued","label":"Date","values":["2025"]},{"key":"dc:publisher","label":"Institution","values":["ResearchSpace@Auckland"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["PhD"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["The University of Auckland"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Intelligent maintenance automation"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated."]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/2292/71212"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Intelligent maintenance system (IMS) is crucial for optimizing maintenance strategies, reducing downtime, and improving overall equipment effectiveness in modern manufacturing systems. However, existing maintenance systems often struggle to effectively integrate and analyse multi-modal and heterogeneous data sources, leading to suboptimal maintenance decisions. This thesis presents a comprehensive and integrated framework for intelligent maintenance system that leverages advanced natural language processing techniques, particularly large language models (LLMs), and domain-specific knowledge to provide decision support on predictive maintenance, fault diagnosis, and maintenance task planning. The proposed framework consists of three key components: 1) a multi-modal fusion methodology for integrating sensor data and event logs to improve the accuracy of predictive maintenance and fault diagnosis models; 2) a task-centric knowledge graph (TCKG) schema and extraction framework for capturing maintenance knowledge from unstructured documents; and 3) an intelligent question-answering system that integrates LLMs and TCKGs to provide accurate, context-aware answers to maintenance queries and support task planning. The effectiveness and potential impact of the proposed framework are demonstrated through extensive experiments and case studies using real-world maintenance datasets. The results show substantial improvements in fault prediction accuracy, maintenance knowledge extraction, and question-answering performance compared to state-of-the-art baselines. The thesis also discusses the limitations of the current work and identifies future research directions, such as the need for explainable AI techniques, industry-specific ontologies, and the integration with other Industry 4.0 technologies. This thesis contributes to the field of intelligent maintenance system by proposing new methods for data integration, knowledge representation, and intelligent question answering for maintenance task planning. The framework's modular architecture allows for seamless integration with existing systems and adaptation to evolving industry needs. The research outcomes have the potential to drive improvements in operational efficiency, cost reduction, and asset performance, aligning with Industry 4.0 objectives."]},{"key":"dc:title","label":"Title","values":["Intelligent Industrial Maintenance: Using Natural Language Processing Technology and Domain Knowledge for Decision Support"]}]}],"canonical_facts":{"dc:contributor.advisor":["Aw, Kean","Hui, Justine"],"dc:creator":["Liu, Zengkun"],"dc:date.accessioned":["2025-02-06T19:28:07Z"],"dc:date.available":["2025-02-06T19:28:07Z"],"dc:date.issued":["2025"],"dc:description.abstract":["Intelligent maintenance system (IMS) is crucial for optimizing maintenance strategies, reducing downtime, and improving overall equipment effectiveness in modern manufacturing systems. However, existing maintenance systems often struggle to effectively integrate and analyse multi-modal and heterogeneous data sources, leading to suboptimal maintenance decisions. This thesis presents a comprehensive and integrated framework for intelligent maintenance system that leverages advanced natural language processing techniques, particularly large language models (LLMs), and domain-specific knowledge to provide decision support on predictive maintenance, fault diagnosis, and maintenance task planning. The proposed framework consists of three key components: 1) a multi-modal fusion methodology for integrating sensor data and event logs to improve the accuracy of predictive maintenance and fault diagnosis models; 2) a task-centric knowledge graph (TCKG) schema and extraction framework for capturing maintenance knowledge from unstructured documents; and 3) an intelligent question-answering system that integrates LLMs and TCKGs to provide accurate, context-aware answers to maintenance queries and support task planning. The effectiveness and potential impact of the proposed framework are demonstrated through extensive experiments and case studies using real-world maintenance datasets. The results show substantial improvements in fault prediction accuracy, maintenance knowledge extraction, and question-answering performance compared to state-of-the-art baselines. The thesis also discusses the limitations of the current work and identifies future research directions, such as the need for explainable AI techniques, industry-specific ontologies, and the integration with other Industry 4.0 technologies. This thesis contributes to the field of intelligent maintenance system by proposing new methods for data integration, knowledge representation, and intelligent question answering for maintenance task planning. The framework's modular architecture allows for seamless integration with existing systems and adaptation to evolving industry needs. The research outcomes have the potential to drive improvements in operational efficiency, cost reduction, and asset performance, aligning with Industry 4.0 objectives."],"dc:identifier.uri":["https://hdl.handle.net/2292/71212"],"dc:publisher":["ResearchSpace@Auckland"],"dc:rights":["Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated."],"dc:rights.uri":["https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm"],"dc:subject":["Intelligent maintenance automation"],"dc:title":["Intelligent Industrial Maintenance: Using Natural Language Processing Technology and Domain Knowledge for Decision Support"],"dc:type":["Thesis"],"thesis:degree_discipline":["Engineering"],"thesis:degree_level":["Doctoral"],"thesis:degree_name":["PhD"],"thesis:institution_name":["The University of Auckland"]},"updated_at":"2026-07-24T01:05:30Z"}