Y.M.W.H.M.R.R.L.J.B. Kiridana’s scientific contributions

What is this page?

This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (1)

Figure 1: Distribution of NLP models across application areas (Source: Developed by authors) 2.2 STATE OF THE ART OF NLP MODELS IN CONSTRUCTION Recent advancements in NLP have revolutionised the construction industry's approach to contract analysis and management. Notably, Padhy et al. (2021) demonstrated an 80% increase in efficiency with an NLP model designed to detect exculpatory sentences. As validated by Hand et al. (2021), F1 scores exceeding 70% indicate a reliable and effective model. Lee et al. (2019) and Lee et al. (2020) further demonstrated the competence of NLP in automatically detecting problematic clauses with impressive F1 scores of 81.8% and 80%, respectively. These findings highlight the robustness of NLP in scrutinising contractual documents, flagging critical clauses, and enhancing decision-making processes in the construction industry.
Figure 2: Blueprint for the NLP-PCMN (Source: Developed by authors)
Figure 3: Synthesising user input data with pre-existing databases (Source: Developed by authors)
Sources of textual data in a construction project
Profile of the experts
AI models for predicting construction disputes in Sri Lanka
  • Conference Paper
  • Full-text available

August 2024

·

214 Reads

Y.M.W.H.M.R.R.L.J.B. Kiridana

·

M.D.T.E. Abeynayake

·

Construction disputes pose persistent challenges in Sri Lanka's construction industry, leading to project delays, cost overruns, and strained professional relations. This research seeks to alleviate these issues by introducing an AI-powered predictive model designed to identify and analyse dispute risks at the project's outset. By offering proactive insights, the AI model aims to enhance decision-making and facilitate the implementation of dispute prevention strategies, thereby improving overall project outcomes. Employing a mixed-methods approach, the study comprehensively examined project features contributing to disputes within the Sri Lankan context. Quantitative data on project characteristics and their correlation with dispute occurrence were gathered through structured questionnaires, while qualitative insights into dispute causes and stakeholder challenges were obtained via in-depth interviews with industry experts. Through meticulous analysis of this combined data, key predictors of construction disputes were identified, including contract ambiguities, unrealistic timelines, payment delays, poor communication, and unforeseen site conditions. These findings drove the development of a machine learning-based predictive model trained to recognise patterns, predict dispute likelihoods, and suggest their nature based on identified risk factors. This innovative AI tool has the potential to revolutionise dispute management practices in Sri Lanka's construction industry. By providing stakeholders with early warnings of potential disputes, the model enables proactive mitigation strategies, such as enhanced contract drafting, optimised communication, and timely alternative dispute resolution. The long-term impact of this research extends to fostering a more collaborative and sustainable construction industry, ultimately contributing to the successful delivery of projects across Sri Lanka.

Download