The Research Department of Economics and Management of Tadbir Nikan Business, Marketing, and Finance Open 3 Serial Number 17 2026 09 01 Investigating the Impact of Machine Learning and Artificial Intelligence on Activity-Based Project Costing Investigating the Impact of Machine Learning and Artificial Intelligence on Activity-Based Project Costing 1 19 EN Journal Article 2026 03 10 <p>This study aimed to investigate the impact of artificial intelligence and machine learning capabilities on the effectiveness of activity-based project costing among professionals working in project-oriented organizations in Tehran. This applied, quantitative, cross-sectional study was conducted in 2026 among professionals involved in project management, project control, cost engineering, accounting, financial management, information systems, and data analysis in project-oriented organizations in Tehran. Using purposive and stratified sampling, 286 valid responses were obtained from 320 distributed questionnaires. Data were collected using a researcher-developed questionnaire assessing artificial intelligence utilization, machine learning capability, automated activity classification, intelligent cost-driver identification, predictive cost forecasting, and activity-based project costing effectiveness. Content validity was evaluated by experts, while construct validity and reliability were assessed using confirmatory factor analysis, Cronbach’s alpha, composite reliability, and average variance extracted. Data were analyzed using Pearson correlation coefficients and structural equation modeling. Structural equation modeling showed that artificial intelligence utilization had a significant positive effect on activity-based project costing effectiveness (β = .24, p &lt; .001). Machine learning capability also significantly predicted costing effectiveness (β = .29, p &lt; .001). Automated activity classification (β = .13, p = .009), intelligent cost-driver identification (β = .21, p &lt; .001), and predictive cost forecasting (β = .31, p &lt; .001) each demonstrated significant positive effects. Predictive cost forecasting produced the strongest direct effect. Collectively, the predictor variables explained 68% of the variance in activity-based project costing effectiveness (R² = .68). The structural model demonstrated satisfactory fit indices (χ²/df = 2.24, CFI = .946, TLI = .937, RMSEA = .066, SRMR = .054). Artificial intelligence and machine learning can substantially enhance activity-based project costing by improving predictive forecasting, cost-driver identification, activity classification, cost allocation accuracy, and managerial decision support, thereby strengthening cost management in complex project environments.</p> https://www.bmfopen.com/index.php/bmfopen/article/download/577/517