The mode share of app-based ride-hailing services has been growing steadily in recent years and this trend is expected to continue. Ride-hailing services generate two types of trips – passenger hauling trips and deadheading trips. Passenger hauling trips are the trips made while transporting passengers between places. Virtually all other trips made by a ride-hailing vehicle when there are no passengers in the vehicle are called deadheading trips or empty trips. Trips between the drop-off location of one passenger and the pick-up location of the next passenger could comprise a substantial share of total travel by ride-hailing vehicles, both in terms of number of trips and miles of travel. This paper aims to model the deadheading trips produced by app-based ride-hailing services at the disaggregate level of individual trips. Passenger trip data published by the app-based ride-hailing company Ride Austin is used to impute deadheading trips. The pick-up locations of passengers are then modeled using a nonlinear-in-parameters multinomial logit framework, essentially capturing the deadheading that takes place from the drop-off of one passenger to the pick-up of the next passenger. The model is sensitive to socio-demographic characteristics, as well as employment opportunities and built environment characteristics of the study area. The model results shed light on the characteristics of deadheading trips at different locations and at different time periods in a day. The paper concludes with a discussion of how transportation planners and app-based ride-hailing companies may utilize knowledge about deadheading to enact policies and pricing schemes that reduce deadheading.

近年来，基于应用程序的乘车服务的模式份额一直在稳步增长，并且这一趋势有望继续。乘车服务产生两种类型的旅行-载客旅行和无头旅行。旅客拖运是指在不同地点之间运送旅客的旅行。实际上，在没有乘客的情况下，乘车车辆进行的所有其他行程称为无头行程或空行程。从乘车次数和行驶里程两方面，一位乘客的下车地点与下一位乘客的上车地点之间的行程可占乘车车辆总行程的很大一部分。本文旨在对各个行程的分类水平上由基于应用程序的打车服务产生的无聊行程进行建模。由基于应用的乘车公司Ride Austin发布的乘客旅行数据用于估算无头旅行。然后使用非线性参数多项式logit框架对乘客的接送位置进行建模，从本质上捕获从一名乘客下车到下一名乘客的接送发生的沉头现象。该模型对研究区域的社会人口特征，就业机会和建筑环境特征敏感。该模型结果揭示了一天中不同位置和不同时间段的无头旅行的特征。本文最后讨论了运输计划人员和基于应用程序的乘车公司如何利用有关空头的知识来制定减少空头的政策和定价方案。