Factors influencing rideshare satisfaction in a university community

On-demand mobile applications for ridesharing services are a relatively recent development in transportation that promotes peer-to-peer resource reallocation and fosters efficiency and sustainability by optimizing available resources and reducing fuel consumption, traffic, and transportation inequality. Customer satisfaction and retention are key to reaping these benefits, however, and while previous research has examined ride ratings through the lens of customer biases, it fails to evaluate the relationship between customer rideshare satisfaction and other trip features. This study examined the service criteria of rideshare journeys to and from a university in Arlington, Texas between 2021 and 2022. A variety of models were developed to assess the influence of several trip parameters on ride ratings, and the best performing model, the random forest model, was selected for further evaluation. The results indicated that ride distance, duration, month, and day of the week had the greatest impact on the ratings. Partial dependence plots were also created to increase the interpretability of the model, and recommendations were developed for stakeholders. The results have important implications for legislators, rideshare service providers, and transportation professionals.