Customer service and riding experience are crucial for the success of public transportation systems. This study utilises operational data from a public bike program to develop a quality of service (QoS) model, which involves constructing a dataset of available bikes and docks at each station recorded every five minutes over 55 days across 1,379 rental stations. We developed performance indices and created spatiotemporal visualisations for operational assistance. Additionally, we investigated Google Maps reviews posted by bike users using natural language processing and deep learning techniques to develop a quality of experience (QoE) model. The QoE model analysed 4,256 text reviews and 4,164 image reviews, categorised into intent sentiment, text content and image content. Classification models were developed for detailed opinion analysis. A case study focusing on New Taipei’s YouBike system highlights bike shortages as the most significant challenge, particularly at smaller stations. The QoS model identified bike shortages correlated with negative user perspectives in the QoE model, indicating a connection between objective operational data and subjective cyclist opinions. This QoS-QoE joint model provides an integrated approach to assessing service quality and riding experience for public bike operators and city transportation authorities.
