Autonomous cars and activity-based bottleneck model: How do in-vehicle activities determine aggregate travel patterns?

Abstract

When traveling in an autonomous car, the travel time can be used for performing activities other than driving. This paper distinguishes users’ work-related and home-related activities in autonomous cars and proposes an activity-based bottleneck model to investigate travelers’ behavior in the morning commute, shedding light on how the scope to undertake in-vehicle activities affects travelers’ trip-timing preferences and decisions, and therewith social welfare. These welfare effects can be expected to depend on the optimality of both the market for trips, and the market for vehicles. We therefore consider different supply regimes for vehicles, as well as unpriced congestion versus queue-eliminating road pricing. We reveal analytically the relationship between various in-vehicle activities and trip-timing decisions by users of autonomous and normal cars. Three supply regimes for autonomous cars are investigated: welfare-maximizing public supply, competitive marginal cost supply, and profit-maximizing private supply. Pricing rules under different supply regimes are compared analytically, and the relative efficiencies in terms of the welfare gains are compared numerically. The results show that travelers’ in-vehicle activity choices have significant impacts on travel patterns, congestion externality, supply decisions, and the associated welfare effects.