Curating a domain ontology for rework in construction: challenges and learnings from practice.

Rework remains an ever-present reality in construction projects. How rework data is defined, its format, location, and quantification contribute to the difficulty in managing its risks. This paper examines the challenges and acquired learnings encountered while attempting to create and curate a domain rework ontology in construction. An explanatory case study approach utilizing the lens of pragmatism is used to develop a rework ontology in a real-life transport infrastructure mega-project procured using a program alliancing delivery strategy. As part of an alliance's continuous improvement strategy, it aims to redress its rework problem by assessing its risk pathways and managing them using a semantic model through an ontology. We use a hybrid approach to develop an ontology. The Correlation Explanation (CorEx) topic modelling approach, a machine learning method, is used to help define our ontology's scope and generate a rework taxonomy representative of practice. The theoretical and practical implications of developing a domain rework are also discussed. Our paper's main contribution is the propagation of a knowledge repository that others can learn from when developing an ontology and addressing the problem of rework, which is often embedded in the practice of construction. [ABSTRACT FROM AUTHOR]