Development and Validation of an AI-Enabled Predictive Quality and Reliability Framework for U.S. Manufacturing Systems

Abstract

The increasing adoption of artificial intelligence in U.S. manufacturing has created substantial opportunities for predictive quality management and manufacturing reliability, yet limited empirical evidence has explained how predictive analytics, intelligent quality monitoring, predictive maintenance, and real-time data integration operate collectively within an integrated performance framework. This study aimed to develop and quantitatively validate an AI-enabled predictive quality and reliability framework for U.S. manufacturing systems. A quantitative, cross-sectional, case-study-based research design was employed using a structured five-point Likert-scale questionnaire administered to professionals representing diverse U.S. manufacturing organizational cases and sectors. Of 330 questionnaires distributed, 312 were returned and 304 usable responses were retained, producing a 92.1% usable response rate. The principal variables included AI-Based Predictive Analytics, Intelligent Quality Monitoring and Defect Detection, AI-Enabled Predictive Maintenance, Real-Time Manufacturing Data Integration, Predictive Quality Performance, and Manufacturing-System Reliability. Data analysis incorporated descriptive statistics, Cronbach’s alpha reliability testing, Kaiser-Meyer-Olkin and Bartlett’s tests, Pearson correlation analysis, and multiple regression modeling. The findings demonstrated high levels across all constructs, including AI-Based Predictive Analytics, M = 4.12, SD = 0.56; Intelligent Quality Monitoring and Defect Detection, M = 4.18, SD = 0.53; AI-Enabled Predictive Maintenance, M = 4.07, SD = 0.60; Real-Time Manufacturing Data Integration, M = 3.94, SD = 0.64; Predictive Quality Performance, M = 4.16, SD = 0.55; and Manufacturing-System Reliability, M = 4.21, SD = 0.52. The Predictive Quality Performance regression model explained 65.2% of variance, R² = .652, with Intelligent Quality Monitoring emerging as the strongest predictor, β = .35, p < .001. The Manufacturing-System Reliability model explained 69.1% of variance, R² = .691, with Predictive Quality Performance producing the strongest effect, β = .39, p < .001. All six hypotheses were supported. The results indicate that coordinated AI-enabled sensing, monitoring, data integration, quality prediction, and predictive maintenance capabilities can strengthen manufacturing quality and reliability, providing practical guidance for manufacturers seeking integrated, data-driven operational improvement.