Title: Identifying Trip Purposes on Trip Level for Vehicle Sensor Data
Abstract: The understanding of car usage patterns and the reason for a trip is important for policymakers to derive measures to influence car usage as well as for manufacturers and service providers to create target-oriented products and offers. There are different types of data to describe car usage. Survey data provide various information that explain behavior of individuals for a short time period. In contrast, sensor data from cars contain detailed usage data over a longer time period, but do not allow conclusions to be drawn about the purpose of the trip. In existing literature is a lack of research on how to combine the best of both data types without explicit validation by participants. The aim of this study is to identify and analyze the purposes of trips in sensor data by using a car use model that is based on survey data from a national household travel survey. The characterization of trips with different purposes is used to train a model. This is applied to sensor data of about 51,000 cars from nine European countries with 7,489,686 trips in the course of half a year of a German premium Original Equipment Manufacturer. The results show that the chosen approach is useful for the identification of trip purposes in sensor data. All in all, 73% of the trip purposes in the model could be correctly predicted at trip level. In an in-depth analysis, the authors compare car usage across the nine countries considered and evaluate the trips differentiated by fuel type.
Publication Year: 2021
Publication Date: 2021-01-01
Language: en
Type: article
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot