Volume 2 - Issue 3, May - June 2026

Nigerian polytechnics face significant infrastructure management challenges, including deteriorating facilities, reactive maintenance practices, and rising energy costs. Smart infrastructure systems that integrate Internet of Things (IoT) sensors, cloud computing, and machine learning offer a promising approach for transforming facilities management from reactive to predictive. However, empirical evidence of such systems in sub-Saharan African polytechnics remains limited. This study presents the design, deployment, and twelve-month evaluation of a smart infrastructure and predictive maintenance pilot system implemented across five academic buildings at Auchi Polytechnic, Edo State, Nigeria. A pilot deployment approach was adopted, involving the installation of 143 IoT sensor nodes for monitoring temperature, humidity, energy consumption, structural vibration, and occupancy. The sensors were deployed across the Faculty of Engineering, School of Management, Library Complex, ICT Centre, and Science Block. Data were collected through a cloud-based MQTT platform and analysed using Random Forest, Long Short-Term Memory (LSTM), and Support Vector Machine (SVM) machine learning models trained on twelve months of baseline data. Pre- and postdeployment performance was assessed using descriptive statistics and paired Wilcoxon signed-rank tests. The results revealed substantial improvements following deployment. Average energy consumption across the pilot buildings decreased by 26.9%, while emergency repairs and reactive maintenance work orders declined by 74.5% and 61.3%, respectively. Among the predictive models, LSTM achieved the highest fault prediction accuracy of 96%, followed by Random Forest (95%) and SVM (86%). The system generated an estimated annual net financial benefit of ₦9.93 million and achieved a projected payback period of 2.8 years on the ₦27.8 million investment. The findings demonstrate that IoT-enabled predictive maintenance systems are both technically feasible and economically viable for improving energy efficiency, maintenance performance, and asset management in Nigerian polytechnic institutions.

Bldr. Bamidele Osamudiamen,"Smart Infrastructure and Predictive Maintenance Systems for Academic Buildings in Nigerian Polytechnics: Evidence from a Pilot Deployment at Auchi Polytechnic" International Journal of Advanced Multidisciplinary Research and Educational Development, V2(3): Page(1030-1038) May-June 2026. ISSN: 3107-6513. www.ijamred.com. Published by Scientific and Academic Research Publishing.