A Hybrid ML Model for Proactive Occupational Health Surveillance in Nigeria's Informal Sector

Abstract

Occupational health surveillance in Nigeria remains limited, particularly within the informal sector, which employs over 80 % of the workforce, yet lacks structured monitoring systems. This study proposes an AI-enhanced occupational health surveillance framework designed to detect and predict work-related illnesses among informal workers in Lagos, Enugu, and Kano States. The framework integrates Internet-of-Things (IoT) sensors, mobile data collection, and hybrid machine learning (Random Forest + Convolutional Neural Network) to capture and analyse environmental, physiological, and behavioural indicators in real time. The research gaps justify the present study’s focus on: Developing a predictive AI model for the informal sector, creating context-sensitive occupational risk profiles, and designing an ethical, policy-integrated deployment framework. A mixed-methods socio-technical approach was adopted, combining quantitative sensor analytics with qualitative stakeholder input from 300 participants across five occupational groups (welding, tailoring, food vending, construction, and driving). The hybrid model achieved 87.2 % accuracy and an AUC of 0.91, outperforming baseline algorithms in predicting fatigue, heat stress, and exposure-related risks. Correlation analysis revealed that temperature (r = 0.81) and literacy level (r = –0.74) significantly influenced risk levels, underscoring the role of socio-environmental determinants in occupational health outcomes. An ethical audit confirmed compliance levels above 85 % across domains of privacy, consent, and fairness, while a policy workshop validated the feasibility of integrating AI-driven surveillance into Nigeria’s National Occupational Safety and Health Policy (2020). The findings demonstrate that artificial intelligence can transform occupational health monitoring from reactive reporting to proactive prevention, enhance data-driven policymaking, and promote equity in health protection for informal workers. The framework offers a scalable model for digital public-health transformation across sub-Saharan Africa. In conclusion, this research offers a replicable blueprint for sub-Saharan Africa’s transition toward data-driven, inclusive, and ethically guided occupational health governance, positioning AI as a catalyst for sustainable public-health transformation.