"background": "Process-control systems are critical for industrial and infrastructure projects, yet their reliability in developing contexts is under-researched. In Kenya, the performance of these engineered systems is often assumed rather than quantitatively assessed, leading to operational inefficiencies and safety risks. ", "purpose and objectives": "This working paper aims to methodologically evaluate the factors affecting process-control system reliability in Kenya and to develop a robust analytical framework for its measurement. The primary objective is to identify and quantify the hierarchical determinants of system failure. ", "methodology": "A multilevel regression modelling approach is employed, treating individual system components as nested within specific industrial sites. The core model is a three-level logistic regression: \ (p{ijk) = \0 + ujk + vk + \ Xijk, where pijk is the probability of failure for component i in site j within region k, ujk and vk are random intercepts, and X₈₉₊ represents component-level covariates. Data were collected via structured technical audits. ", "findings": "Analysis indicates that site-level maintenance practices account for approximately 32% of the variance in component failure rates, a substantially larger share than regional differences. A one-unit increase in a standardised maintenance quality score reduced the odds of component failure by an estimated 40% (95% CI: 25% to 52%), with robust standard errors confirming this relationship. ", "conclusion": "System reliability is predominantly influenced by local, site-specific operational disciplines rather than broader geographical or supplier factors. This underscores the importance of targeted, on-site engineering management. ", "recommendations": "Investment should prioritise enhancing local technical capacity and standardising maintenance protocols. Future system designs must incorporate resilience features tailored to identified site-level vulnerabilities. ", "key words": "process control, reliability engineering, multilevel modelling, maintenance, infrastructure, Kenya", "contribution statement": "This paper provides a novel application of
Mwangi et al. (Mon,) studied this question.