AbdElMottaleb, Mohamed (2025) AI in project management: Machine learning application for construction scheduling. PhD thesis, University of Salford, UK.
Abstract
This study investigates the integration of Artificial Intelligence (AI) and Machine Learning (ML) into construction project scheduling to address inefficiencies in traditional methods. While conventional tools like Primavera and MS Project rely heavily on manual input, existing AI/ML research often neglects critical scheduling components such as activity relationships, critical path analysis, and user-friendly interfaces. This research aims to bridge these gaps by developing an ML-driven scheduling tool that automates activity sequencing, duration prediction, and critical path identification, while incorporating a practical interface for industry adoption.The study employs a multi-methods approach, integrating both quantitative and qualitative data gathered from a survey of construction professionals with a design science framework. Drawing on these insights, a Random Forest Regressor model was developed and trained using real-world power sector project data specifically overhead transmission line (OHTL) and substation construction, with its performance evaluated through R² and MAE metrics. Additionally, a prototype interface was constructed using Streamlit to assess usability and facilitate practical integration. Crucially, external validation by five highly experienced industry experts (>15 years relevant roles) confirmed its significant speed advantage (80% found it 'Much faster'). Experts positively rated its user-friendliness and ease of schedule generation, with 80% considering it for actual projects as a valuable initial planning tool.Preliminary findings, based on historical power sector data showed that the developed model demonstrates high predictive accuracy (R² values of 0.91 for successor, 0.93 for predecessor; MAE < 2.2 days). It autonomously determines activity sequences, predicts durations, and identifies the critical path, significantly enhancing planning efficiency and reducing errors. A user-friendly Streamlit interface enables dynamic schedule generation in under 30 seconds, with options to adjust duration constraints and export to Excel and XER with minimal manual intervention.Theoretically, this study contributes a novel, integrated Machine Learning framework that, unlike prior fragmented research, holistically automates the end-to-end scheduling process. Practically, it delivers a validated software artifact that demonstrates significant efficiency gains reducing schedule generation time from days to seconds and provides new empirical insights into the factors governing AI adoption in construction project management.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Thesis advisor: | Munir, Mustapha and Underwood, Jason |
| Index terms: | machine learning, validation, integration, activity sequence, construction project, accuracy, survey, drawing, usability, prototype, science, project data, artifact, critical path, scheduling, forest, option, project management, artificial intelligence, critical path analysis, duration, substation, machine learning application, construction project management, export, construction professional, successor, construction scheduling, efficiency |
| Subjects: | production management, environmental science, economic analysis, operations research, specialized education, infrastructure and transport systems, organizational analysis, project controls, sociology, decision analysis, professional development, project management theory and practice, performance management, artificial intelligence, modelling and simulation, technical documentation, data collection methods, user-centered design |
| Topics: | Design Practice, Digital Applications, Time Control, Organizational Design, Research Practice, Information Management, Business Strategy, Quality Management, Education, Project Management, Engineering Principles, Sustainability, Risk Management |
| Descriptive scope: | 3 PCE |
N.B. Descriptive scope is a count of how many of the five facets of empirical research are indicated by the words used in title, abstract and keywords. It is not intended as a judgement on the research; merely a count of the kind of word we would expect to indicate Phenomenon, Concepts, Theoretical framing, Empirical techniques, Analytical techniques. If all five are present, then a code of “5 PCTEA” will indicate this. If you feel the coding for this record is questionable, we welcome discussion around the terms we matched or the way we categorized them. The facet you would expect may not be coded, or a facet may be coded inappropriately. This can also bear on a larger question, of which facets should be treated as defining in construction management research. Please get in touch, and we will look at it. More details here