{ "background": "Industrial policy in Kenya has increasingly emphasised advanced manufacturing, yet rigorous evaluation of the engineering process-control systems central to this ambition is lacking. Existing assessments often rely on descriptive case studies, failing to isolate the causal impact of control interventions on production yield.", "purpose and objectives": "This policy analysis article aims to demonstrate the application of a quasi-experimental design for the causal diagnostic of industrial process-control systems. Its objective is to provide a methodological framework for quantifying yield improvements attributable to specific control enhancements, thereby informing evidence-based industrial policy.", "methodology": "We employ a difference-in-differences (DiD) design, analysing panel data from manufacturing firms implementing upgraded proportional–integral–derivative (PID) control loops. The core statistical model is $Y{it} = \\beta0 + \\beta1 (\\text{Treat}i \\times \\text{Post}t) + \\gammai + \\deltat + \\epsilon{it}$, where $Y_{it}$ is yield for firm $i$ at time $t$. Inference is based on cluster-robust standard errors at the firm level.", "findings": "The analysis indicates a statistically significant positive treatment effect. Implementation of tuned PID control systems was associated with an average yield increase of 7.3 percentage points (95% CI: 5.1, 9.5). The principal mechanism identified was a reduction in output variance, leading to fewer off-specification production batches.", "conclusion": "Quasi-experimental methods offer a robust, evidence-based tool for evaluating industrial engineering interventions, moving policy assessment beyond anecdotal reporting. This approach can directly link specific process-control investments to measurable productivity gains.", "recommendations": "Kenyan industrial policy should mandate quasi-experimental or randomised evaluation frameworks for publicly supported process-optimisation programmes. Investment incentives should be contingent on the collection of high-frequency production data suitable for causal analysis.", "key words": "process control, quasi-experimental design, industrial policy