Leveraging Machine Learning for Resilient Urban Transport Planning under Deep Uncertainty: A Case Study of Abuja, Nigeria

Urban transportation systems in developing cities like Abuja, Nigeria, face growing uncertainties driven by rapid urbanization, fluctuating economic conditions, climate variability, and evolving technologies. Conventional planning models often rely on static assumptions that fail to capture these unpredictable dynamics. This study introduces a machine learning (ML)–based framework designed to enhance resilience in urban transport planning by identifying, modeling, and forecasting uncertain factors that affect mobility and accessibility in Abuja. The research integrates multiple data sources—traffic flow, satellite imagery, weather data, and socioeconomic indicators—to train predictive models using supervised and unsupervised ML techniques such as Random Forests and Gradient Boosting. These models are employed to detect complex, non-linear relationships between transport demand, infrastructure performance, and external stressors. Scenario-based simulations are then used to evaluate the robustness of policy interventions under diverse conditions, including population surges, fuel price volatility, and climate-related disruptions. Preliminary results indicate that ML algorithms significantly improve the accuracy of transport demand forecasting and enable adaptive, data-driven decision-making. The framework empowers planners to simulate policy alternatives and assess their long-term sustainability before implementation. In the context of Abuja, where transport challenges are compounded by rapid population growth and informal transit systems, this approach offers a pragmatic solution for developing responsive mobility strategies that can withstand deep uncertainty. The study’s contributions are twofold: it demonstrates the practical applicability of ML techniques in resource-constrained environments, and it provides actionable insights for policymakers seeking to design resilient and sustainable transport systems. Ultimately, the research advocates for the integration of intelligent decision-support tools into Nigeria’s urban planning processes to promote equitable, efficient, and environmentally sound mobility for all.