INTEGRATION OF LEAN SIX SIGMA AND ARTIFICIAL INTELLIGENCE-ENABLED DIGITAL TWIN TECHNOLOGIES FOR SMART MANUFACTURING SYSTEMS

Abstract

This quantitative study investigated the integration of Lean Six Sigma (LSS), artificial intelligence (AI), and digital twin (DT) technologies as a unified framework for achieving measurable performance improvement in smart manufacturing systems. The research aimed to evaluate the extent to which AI-enabled digital twins could enhance Lean Six Sigma’s analytical and process control capabilities and to determine the quantitative impact of this integration on operational efficiency, defect reduction, and production reliability. Data were collected from 150 participants across 20 manufacturing organizations that had implemented digital transformation initiatives involving LSS, AI, and DT frameworks. Using descriptive, correlational, and multiple regression analyses, the study examined how these independent variables jointly influenced key performance indicators, including mean time between failures (MTBF), overall equipment effectiveness (OEE), and defect rate. The results indicated that the integration model was statistically significant, with an adjusted R² of 0.719, confirming that approximately 72% of the variance in performance outcomes could be explained by the combined influence of LSS, AI, and DT. Correlation analysis revealed strong positive associations between AI integration and OEE (r = 0.816) and between DT utilization and MTBF (r = 0.802), while defect rate demonstrated significant negative correlations with all three predictors. Reliability testing produced Cronbach’s alpha values exceeding 0.85 for all constructs, confirming instrument consistency, while validity testing established clear construct alignment through factor analysis. Regression coefficients demonstrated that AI integration had the highest predictive strength (β = 0.447, p < 0.001), followed by digital twin synchronization (β = 0.389, p < 0.001) and Lean Six Sigma implementation (β = 0.312, p < 0.001). These findings provided empirical evidence that combining process improvement methodologies with intelligent simulation and predictive analytics produced significant, quantifiable enhancements in manufacturing performance.