Fare Evasion

Fare evasion in public transportation represents a significant financial loss for operators and hinders accurate ridership analysis. Current research focuses on leveraging data from automated fare collection (AFC) and automatic passenger counting (APC) systems, often employing machine learning techniques like deep neural networks (including LSTM models) and geostatistical methods to predict evasion patterns and estimate overall occupancy. These advancements aim to improve resource allocation, enhance operational efficiency, and provide a more comprehensive understanding of passenger behavior.