Household-based E-commerce demand modeling for an agent-based urban transportation simulation platform

The e-commerce market has grown rapidly in the past two decades. The need for predicting e-commerce demand and evaluating relevant policies and solutions is increasing. However, the existing simulation models for e-commerce demand are still limited and do not consider the impacts of delivery options and their attributes that shoppers face on multiple dimensions of e-commerce demand. We propose a novel framework involving disaggregate behavioral models that jointly predict e-commerce expenditure, purchase amount per transaction, delivery mode, and option choices. The proposed framework can simulate the changes in e-commerce demand and be used to evaluate the impacts of a range of policies and solutions. We specify the model parameters based on various sources of relevant information, integrate the model into an urban freight simulator, and conduct a demonstrative simulation for a prototypical North American city. The results of the analysis highlight the capability and applicability of the proposed modeling framework.