Speaker
Description
The public transport systems in most Sub-Saharan African cities (SSA) lack a standardized fare structure, resulting in arbitrary pricing that disproportionately affects both commuters and operators of various socio-demographic groups. This study addresses the need for a fair, transparent, equitable and adaptive fare system by developing a kilometre-based pricing model using AI. By analyzing economic, environmental and traffic attributes such as distance, vehicle type, fuel price, day type, weather condition, traffic level and traffic period, a predictive fare algorithm was developed and integrated into a prototype web-based application called Fare Estimation App (FEsApp). With its localization feature, FEsApp can estimate the transport fare for various public transport modes, and can provide a breakdown of what constitutes the estimated fare. It can simulate the expected change in transport fare over time, and can instantly compare the transport fare for five public transport modes for the same trip length. This estimation framework can help stabilize pricing, improve equity, and promote fairness and transparency in the public transport pricing system of SSA cities. With its potential to integrate various online fare payment systems, the application can support multimodal and integrated public transport systems of both formal and informal services.
Keywords: Public Transport, Transport Fare, Transport Fairness, Transport Equity, Machine Learning, Freetown