Parking preferences of delivery drivers in the Paris Region: Understanding the role of anticipation using hybrid choice models

This study explores the determinants of parking choices for commercial vehicles in the Paris Region (France). The analysis is based on data from the 2010 Paris Region Urban Goods Movement Survey (UGMS), which offers insights into the parking preferences of delivery drivers. By examining real-world decision-making, the dataset allows us to consider spatial and temporal characteristics as well as the role of parking decision within the delivery process. An integrated choice and latent variable model is employed, whereby drivers select parking locations based on urban environmental attributes, service type, and a latent variable reflecting anticipated delivery difficulty. This difficulty is inferred from observed delivery times and service characteristics; furthermore, temporal variations are incorporated to assess driver behavior, including fluctuations in parking preferences throughout the day. The model also accounts for parking space availability by the means of latent classes. Our findings contribute to a nuanced understanding of delivery drivers' behavior, providing valuable insights for policy-making and operational strategies. These results, as well as our modeling approach, can also be incorporated into broader frameworks such as agent-based models.