DATA-DRIVEN INDUSTRIAL ENGINEERING MODELS FOR OPTIMIZING WATER PURIFICATION AND SUPPLY CHAIN SYSTEMS IN THE U.S.

Abstract

This study addresses the problem that U.S. water purification enterprises are rapidly adopting cloud enabled analytics and enterprise information systems, yet purification stability and supply readiness remain inconsistent when analytics capability, data governance, industrial engineering discipline, and cross functional coordination are not integrated into one operational optimization approach. The purpose was to test a data driven industrial engineering capability framework by estimating how five predictors, Analytics Capability (AC), Data Quality and Governance (DQG), IE Process Optimization Maturity (IEM), Supply Chain Visibility and Coordination (SCV), and Operational Flexibility and Responsiveness (OFR), relate to three outcomes, Purification System Performance (PSP), Supply Chain Performance (SCP), and Overall Optimization Outcome (OOO). Using a quantitative cross sectional, case-based design, a structured five-point Likert survey was administered in cloud and enterprise system contexts across U.S. water purification operations, yielding 210 usable responses from operations, maintenance, quality, procurement, and analytics roles. Descriptive results showed moderate to high capability levels (IEM M=3.90, SD=0.58; AC M=3.84, SD=0.62), while SCV and OFR were comparatively lower (SCV M=3.65, SD=0.70; OFR M=3.59, SD=0.73); outcomes were above the midpoint (PSP M=3.88, SD=0.60; SCP M=3.62, SD=0.71; OOO M=3.79, SD=0.63). The analysis plan included data screening, reliability testing, Pearson correlation analysis, and three multiple regression models with multicollinearity checks (all VIF<2.0), and internal consistency was strong (Cronbach’s α=.82 to .91). Correlations supported the proposed relationships, including AC–PSP r=.58, IEM–PSP r=.61, SCV–SCP r=.63, DQG–OOO r=.52, PSP–OOO r=.66, and SCP–OOO r=.62 (all p<.001). Regression findings showed that PSP was explained at R²=.469 (F(4,205)=46.20, p<.001) with IEM (β=.33, p<.001) and AC (β=.24, p<.001) as the strongest predictors; SCP was explained at R²=.430 (F(4,205)=39.10, p<.001) dominated by SCV (β=.46, p<.001); and OOO was explained at R²=.520 (F(3,206)=73.40, p<.001) by PSP (β=.39, p<.001), SCP (β=.31, p<.001), and DQG (β=.17, p=.002). Overall, the model indicates that targeted improvements in process optimization maturity and embedded analytics can lift purification performance, investments in supply chain visibility can stabilize supply outcomes, and stronger data governance can amplify integrated optimization results across enterprise and cloud enabled environments.