AI model for analyzing construction litigation precedents to support decision-making

Abstract

Litigation among stakeholders in construction projects has a significantly negative impact on successful project completion and overall performance. Prompt decision-making in relation to litigation is crucial, but the manual review of extensive document sets is time-consuming. In this paper, the natural language processing (NLP) technique was applied to litigation data to develop a model for case summarization and winner prediction. By automatically summarizing the data and predicting litigation outcomes, the proposed model aids practitioners in making timely decisions and enhances document management during disputes. This paper contributes to existing knowledge in two ways. Firstly, the model aids practitioners in making timely decisions about proceeding with litigation. Secondly, unlike previous studies that manually processed raw data such as contracts and specifications, this study utilized NLP to process raw litigation case data automatically. As big data becomes increasingly common, the methodology employed in this study holds academic significance.