Developing urban parking supply and pricing policies in response to automated vehicle market penetration

The ability of Automated Vehicles (AVs) to drive autonomously enables them to drop off passengers at their destinations and park in another location. However, these ghost trips to parking spaces can exacerbate congestion. Parking supply and its pricing can be used as two levers to impact the parking choices of the AVs and the resulting traffic. In this study, we deploy a mesoscopic simulator combined with a genetic algorithm to find the optimal supply and pricing of parking throughout the city to minimize the total network travel time. We also consider the “parking at home” option, which enhances the realism of the problem and expands the feasible solution space. This, in turn, enables the optimization of the objective function with a lower budget in each scenario. Our analysis indicates that pricing alone is less effective in reducing congestion compared to an integrated approach that optimizes both pricing and parking supply. Moreover, as the AV penetration rate increases, the parking supply should gradually shift from suburbs and midtown to the Central Business District (CBD). We demonstrated that, up to a certain budget threshold, it is advisable to concentrate parking supply more in midtown; however, as the budget increases further, the supply should gradually shift towards the CBD. The results also showed that with the increase in penetration rate, the sensitivity of the objective function–network in-vehicle total travel time–to the budget also increases, which means that at higher penetration rates, increasing the budget leads to a more significant improvement in the objective function.