AI-Driven Project Portfolio and Business Analytics for Forecasting Cost Overruns, Schedule Delays, Resource Bottlenecks, and Enterprise Performance Across U.S. Industries

Abstract

This study addresses the continuing difficulty experienced by project-intensive enterprises in identifying cost overruns, schedule delays, and resource bottlenecks early enough to prevent broader portfolio and organizational performance deterioration. The purpose was to quantitatively examine whether AI-driven project portfolio and business analytics capabilities improve forecasting and management effectiveness across diverse U.S. enterprise cases. A quantitative, explanatory, cross-sectional, embedded multiple-case design was employed across construction and engineering, information technology and software, manufacturing, healthcare, energy and utilities, and transportation and logistics organizations. Purposive sampling produced 312 valid responses from project managers, program and portfolio managers, PMO professionals, business and data analysts, finance and project-controls personnel, operations managers, and resource-planning professionals, representing an 89.1% usable response rate. The principal variables were AI-driven data integration and analytics, predictive cost analytics, predictive schedule analytics, AI-enabled resource analytics, portfolio decision intelligence, cost-overrun forecasting effectiveness, schedule-delay forecasting effectiveness, resource-bottleneck management effectiveness, and enterprise performance. Data were collected using a structured five-point Likert questionnaire and analyzed through descriptive statistics, reliability testing, KMO and Bartlett's assessment, Pearson correlation, and multiple regression in SPSS. Respondents reported strong AI-driven data integration (M = 4.08, SD = 0.61), portfolio decision intelligence (M = 4.02, SD = 0.64), and enterprise performance (M = 4.01, SD = 0.63). Predictive cost analytics was the strongest predictor of cost-overrun forecasting effectiveness (β = .36, p < .001; R² = .482), predictive schedule analytics most strongly predicted schedule-delay forecasting effectiveness (β = .34, p < .001; R² = .447), and AI-enabled resource analytics was the dominant predictor of resource-bottleneck management effectiveness (β = .39, p < .001; R² = .415). Enterprise performance produced the strongest overall model, explaining 52.8% of variance, with portfolio decision intelligence as the strongest predictor (β = .38, p < .001). These findings imply that integrated AI analytics can strengthen early warning, resource coordination, portfolio prioritization, and enterprise-level decision quality when analytical outputs are embedded in project governance and managerial action.