An Integrated Approach to Process Optimization and Quality Monitoring Manufacturing Industry

Abstract

This study applied multiple linear regression analysis in conjunction with
statistical process control (SPC) to monitor and improve the quality of plastic
bottle production. Process inputs such as additive level, melt temperature,
injection speed, mold temperature, cooling time, and ambient temperature were
analyzed against three key quality outputs: tensile strength, surface quality score,
and dimensional precision. X̄ control charts were used to detect variations in each
quality characteristic, while regression models identified which process inputs
significantly influenced these outcomes. Results revealed that additive level and
melt temperature were most impactful on tensile strength, mold temperature and
cooling time influenced surface quality, and injection speed and mold temperature
strongly affected dimensional precision. Sensitivity analysis on the surface quality
model showed that optimized input values could align output performance with
control chart expectations, confirming the utility of regression for process
optimization. The study concludes that integrating regression analysis with SPC
provides a statistically grounded approach for identifying critical variables and
improving product quality in manufacturing environments.