Data-Driven Operational Optimization: A Longitudinal Case Study on the Smart Transition in the Metal Manufacturing Factory

Transitioning conventional brownfield metalworking facilities into smart factories often faces practical execution barriers, including legacy equipment constraints, high capital expenditures, and operational downtime risks. This study presents an empirical, longitudinal framework for Industry 4.0 transformation by integrating a 5C Cyber–Physical System (CPS) architecture with Lean Philosophy and Total Productive Maintenance (TPM) tools in a precision metal manufacturing factory in Taichung, Taiwan. Developed over a four-year lifecycle, the deployment utilizes a phased pilot strategy featuring Industrial Internet of Things (IIoT) data acquisition, Ultra-Wideband (UWB) indoor positioning, visual dashboards, and human-in-the-loop dispatching. Analysis of a two-year longitudinal operational dataset demonstrates marked operational enhancements. Value Stream Mapping (VSM) of the specific product on the pilot shop floor reveals a 3.7% decrease in processing time, a 40% reduction in waiting time, and a 20% cut in changeover time, while the operator-to-machine ratio improved from 1:4 to 1:6. Paired-sample t-tests performed using data from the whole pilot shop floor confirm statistically significant improvements in the production capacity achievement rate (PCAR, p (probability) < 0.05) and product defect rate (p < 0.05), with defect rates dropping to 0.19% alongside the decoupling of defect generation from machine failures. Furthermore, a SWOT analysis contextualizes the empirical outcomes, practical limitations, and future predictive maintenance pathways. The proposed methodology provides an actionable roadmap for small and medium-sized manufacturing enterprises (SMEs) pursuing smart manufacturing transitions.