Machine Learning for Construction Delay Prediction in Saudi Arabian Mega-Projects: A Structured Review and Research Agenda Aligned with Vision 2030
Saudi Arabia's Vision 2030 has created one of the largest construction programmes in the world, and delivering it on time is a national priority. Schedule delay has nevertheless been a persistent feature of the Kingdom's construction sector for four decades. Machine learning (ML) offers a way to move from explaining delays after they occur to predicting them while there is still time to act. This paper presents a structured review of published work on ML-based construction delay prediction and of the evidence on delay causes in Saudi Arabia, and derives a research agenda for Saudi mega-projects. The review finds that the causes of delay in the Kingdom are well documented and stable over time, with payment, change-order, planning and workforce factors recurring across studies. ML studies report classification accuracies above 90%, but most are trained on small, questionnaire-derived datasets, are validated internally only, and predict a single static outcome for conventional projects. Only a small number use Saudi project records, and none of the reviewed studies addresses giga-project scale. Seven research priorities are proposed, covering shared project-outcome data, time-dependent prediction from progress records, external validation, explainable models, programme-level interdependence, text-based data sources and the link from prediction to mitigation. Each priority is mapped to the objectives of Vision 2030.
Authors
- Ahmed Babu Shahas V P
Publication Details
- Journal
- Iconic Research and Engineering Journals
- Published
- 2026-11-12
- DOI
- https://doi.org/10.64388/ire1725104
- Primary Topic
- Construction Project Management and Performance
- Type
- article
- Field-Weighted Citation Impact
- 0.00