INTEGRATING BUSINESS INTELLIGENCE AND IOT TO ENHANCE CIRCULAR REVERSE LOGISTICS OF HIGH-VALUE COMPONENTS

High-value industrial components, such as gears, turbine blades, and microprocessors, account for most production costs and environmental impact when discarded at end-of-life. Gears are particularly critical because of their design complexity, load-bearing role, and direct impact on gearbox efficiency and reliability. Extending their lifespan is therefore essential for both economic and environmental sustainability. The Internet of Things (IoT) enables this extension through continuous monitoring and predictive analytics. However, IoT deployments face a persistent bottleneck in the Business Intelligence (BI) & Analytics layer, where interoperability and scalability issues limit right-time insight generation. This study develops a fivelayer IoT architecture (Perception, Edge Computing, Microservice, BI & Analytics, Business) and focuses on a vertically scalable BI & Analytics layer that supports closed-loop monitoring and proactive maintenance of gears. The research follows a structured methodology: literature review, industrial consultation, conceptual modelling, two rounds of industrial validation, laboratory implementation on a gearbox rig, vertical scalability experiments with simulated sensors, and academic validation. The gearbox rig is instrumented with MPU6050 accelerometers and DHT11 temperature-humidity sensors on Arduino UNO and Raspberry Pi 4, exercising the full stack from ingestion to actuation. Findings show stable ingestion and analytics at five-second sampling intervals, with no missing data and consistent medians across sensors. Threshold-based anomaly detection triggers alerts and control feedback, while vertical scalability demonstrates near-linear throughput growth from 100–800 sensors, sustained performance at 1,000, and buffered burst delivery without data loss at 2,000 sensors under CPU saturation. The proposed architecture, validated academically and operationally, provides an industry-ready blueprint for scalable IoT deployments. By demonstrating vertical scalability in the BI & Analytics layer, this work shows how IoT extends gear life, reduces waste, and enhances operational efficiency. Future research explores horizontal scaling and advanced ML-based anomaly detection.