Digital Work Sampling Using Internet of Things and Artificial Intelligence-Based Monitoring System in Nigeria’s Industrial Hubs

Abstract

Introduction: With the introduction of the Industry 4.0 technologies, there are new possible ways to optimize labour productivity and to make operations in manufacturing easier and more efficient.

Objectives:This study presents and justifies a new digital work sample system by combining the Internet of Things (IoT) sensors with the Artificial Intelligence (AI) enabled analytics to track, categorize, and analyze the workforce activities on real-time basis.

Methods: Over a period of approximately 6 months, the data of 150 workers in manufacturing plants located in Ogun State, Nigeria, was empirically gathered leading to a datasets gathered via wearable IoT devices implanted into the different sections of the manufacturing facility, such as the assembly and packaging teams and quality control departments. Random Forest and Neural Networks were used as AI algorithms: to do the classification of activities and detect anomalies.

Results: The results showed that the system obtained a classification accuracy greater than 96%, showed 96.7% decrease in observational demand, up to 50% and 67% decreased anomalies in idle time and reports of incorrect tasks respectively. The overall productivity in the respective departments was improved by up to 29% whereas the latency in decision making decreased by 80%. The analysis of the return on investment revealed 6-month payback period with N200million savings on the operations.

Conclusions: This study drew this conclusion, that integration of IoT-AI in workforce monitoring will make the activity much more productive, easier to view performing and saving much more costs. The suggested system addresses scalable model in case of digital transformation in labour-intensive manufacturing industries, especially, in the realm of emerging economies.