Cargo Revenue Prediction in Nigeria Using Machine Learning: An Ensemble Regression Approach

The cargo business forms the backbone of global trade and national economies, yet revenue forecasting in emerging markets like Nigeria remains highly challenging due to infrastructural deficiencies, regulatory inconsistencies, fuel volatility, port congestion, and seasonal disruptions. The global freight transport market was valued at approximately $19.34 trillion in 2023 and is projected to reach $28.65 trillion by 2030 (International Transport Forum, 2024). In Nigeria, the freight and logistics market stood at USD 60.22 billion in 2022 and continues to grow rapidly (Verified Market Research, 2022; Mordor Intelligence, 2023). Traditional forecasting approaches, reliant on historical benchmarking and linear models, achieve only 65–75% accuracy under stable conditions and deteriorate sharply during volatility (Christopher, 2022; McKinsey & Company, 2023). This study develops and prototypes a localized machine learning framework for cargo revenue prediction tailored to Nigeria’s unique operational realities. Drawing on transactional data from selected Nigerian logistics operators, the research (i) identifies context-specific predictors, (ii) designs a conceptual model integrating Nigerian-specific variables (road conditions, fuel scarcity, port dwell times, and regulatory factors), (iii) evaluates model performance, and (iv) implements a functional prototype decision-support system. By bridging the gap between imported global models and local conditions, the framework delivers significantly higher predictive accuracy than conventional methods. The findings offer practical tools for revenue optimization, resource allocation, and risk reduction, while contributing to the literature on data-driven logistics in developing economies. Limitations include regional data scope and handling of missing values. Future extensions will incorporate macroeconomic indicators and multi-firm datasets.