AI-Enabled Decision Support Systems for Industrial Energy Optimization in U.S. Manufacturing

Abstract

This study addresses the persistent problem of inefficient and reactive energy management in U.S. manufacturing, where fragmented monitoring, delayed reporting, and weak coordination often lead to avoidable energy waste, higher operating costs, and weaker sustainability performance. The purpose of the research was to examine whether AI-enabled decision support systems improve industrial energy optimization by strengthening data-driven decision quality across manufacturing operations. Using a quantitative, cross-sectional, case-based design, the study collected structured questionnaire data from 240 respondents drawn from cloud-enabled and enterprise-oriented manufacturing cases, including plant managers, production supervisors, maintenance engineers, sustainability officers, and data or operations analysts in medium and large firms. The key independent variables were AI-enabled decision support systems capability, predictive analytics capability, and real-time monitoring and automation, while industrial energy optimization served as the main dependent variable, with operational cost reduction and sustainability performance treated as linked outcome variables. Data were analyzed using descriptive statistics, reliability analysis, Pearson correlation, and multiple regression. Findings showed strong internal consistency across constructs, with Cronbach’s alpha values ranging from 0.81 to 0.88. Descriptive results were highly positive, including mean scores of 4.12 for AI-enabled DSS capability, 4.06 for predictive analytics capability, 4.18 for real-time monitoring and automation, and 4.21 for industrial energy optimization. Correlation analysis indicated significant positive relationships with industrial energy optimization for AI-enabled DSS capability (r = 0.61, p < .01), predictive analytics capability (r = 0.58, p < .01), and real-time monitoring and automation (r = 0.66, p < .01). Regression results further revealed that the model explained 54.7% of the variance in industrial energy optimization (R² = 0.547; F = 95.140; p < .001), with real-time monitoring and automation emerging as the strongest predictor (β = 0.370), followed by AI-enabled DSS capability (β = 0.290) and predictive analytics capability (β = 0.240). The study implies that manufacturers can achieve better energy efficiency, cost control, and sustainability outcomes by integrating intelligent decision support, predictive analytics, and real-time automated monitoring into operational routines.