Influential factors in customer satisfaction of transit services: Using crowdsourced data to capture the heterogeneity across individuals, space and time

In a rapidly evolving and highly competitive transportation market, increasing customer satisfaction and ridership retention are essential for transit agencies. Understanding how different market segments perceive transit services can help service providers to identify potential priority areas and develop specialised strategies to address their varying travel concerns. However, traditional studies have predominantly investigated the varia-tions across socioeconomic cohorts or travel characteristics but ignored the complex effect of spatial, temporal, and user heterogeneity. Recently social media data has attracted growing interest from academia as an alter-native way to compensate public attitude surveys with a high volume of semantic, spatial, and temporal in-formation. This study collected 177,807 microblogs from Sina Weibo to understand how customer satisfaction varies among different market segments characterised by social, temporal, and spatial heterogeneity. Method-ologically, this study applies sentiment analysis as a real-time measurement of customers' satisfaction towards transit services, covering safety, crowdedness, reliability, personnel behaviour, and comfort. A beta regression model is then applied to identify the most important explanatory factors and the extent to which explanatory factors affect customers' satisfaction, respectively. The result indicates that age, gender, travel mode, time, and space significantly contribute to customers' satisfaction with the transit system. Their varying impacts on different service attributes are also identified in this study. This study also reveals the highly polarised nature of online sentiment, explaining gendered attitudes. Our research framework could be used as a benchmark for other service industries to conduct similar market segment analysis and integrate it into the policy decision process.