Predicting the Choice of Online or Offline Shopping Trips Using a Deep Neural Network Model and Time Series Data: A Case Study of Tehran, Iran

Abstract

This study examines the determinants of online and offline shopping trip choices and their implications for urban transportation, the environment, and the economy in Tehran, Iran. A questionnaire survey was conducted to collect data from 1000 active e-commerce users who successfully placed orders through both online and offline services in districts 2 and 5 of Tehran during the last 20 days of 2021. A deep neural network model was applied to predict the type of shopping trips based on 10 variables including age, gender, car ownership, delivery cost, and product price. The model’s performance was evaluated against four other algorithms: MLP, decision tree, LSTM, and KNN. The results demonstrated that the deep neural network model achieved the highest accuracy, with a rate of 95.73%. The most important factors affecting the choice of shopping trips were delivery cost, delivery time, and product price. This study offers valuable insights for transportation planners, e-commerce managers, and policymakers. It aims to help them design effective strategies to reduce transportation costs, lower pollutant emissions, alleviate urban traffic congestion, and enhance user satisfaction all while promoting sustainable development.