Ontology-based modelling of lifecycle underground utility information to support operation and maintenance

Abstract

Operation and maintenance (O&M) of underground utilities is crucial to ensure normal functionality of utilities by performing different inspection and maintenance activities. A variety of data are required by maintenance staff to evaluate utility condition and make decisions in planning maintenance works. Nevertheless, data required for utility O&M are usually originated from different sources, with various formats or representations. Interpreting and integrating such heterogenous data for utility O&M remains a significant challenge. Therefore, this paper proposes an ontology-based framework for modelling lifecycle utility information such as to facilitate activities and support decision making during utility O&M. The framework consists of (1) a data layer to collect various types of utility lifecycle data, (2) a knowledge layer to extract common mechanisms and methods for utility O&M, (3) an ontology layer where a domain ontology, Utility Operation and Maintenance Ontology (UOMO), is developed with entities and semantic rules to represent different categories of information related to utility projects, (4) an application layer to automate manual tasks and support decision making by leveraging the query and reasoning capability of ontology. To validate the framework, a case study is performed using real-world inspection data where a user interface is developed to allow users to implement different functions without technical knowledge. The case study demonstrates that the framework can, efficiently and automatically, extract and integrate various information to assist different activities, e.g. defect rating, condition assessment and validation, as well as to support decision making in prioritizing maintenance works, and selecting appropriate maintenance methods. This study contributes by proposing the first ontology-based framework for improving efficiency in utility O&M, where the query and semantic reasoning capabilities of the ontology are fully utilized. The developed ontology with lifecycle utility information is expected to facilitate data interoperability in the utility domain and can be extended with other knowledge for large-scale infrastructure management.