Data-driven mobility permits allocation policy in congested highways

Abstract

Highway congestion during peak periods poses a significant challenge for transportation authorities, necessitating the effective management of highway access across origin–destination (OD) pairs. This paper presents a data-driven mobility permit allocation policy designed to optimize highway access under stochastic demand. We model the problem using stochastic nonlinear programming with the dual targets of minimizing total travel time and ensuring a minimum proportion of highway access for each OD pair. The model is reformulated into a target-based framework that applies quadratic penalties to deviations from both efficiency and fairness targets. We develop an efficient online algorithm that allocates permits in response to each demand scenario and theoretically establish its convergence properties. Numerical experiments using both synthetic and real-world data from the Beijing Capital Airport Highway demonstrate that our proposed method achieves near-optimal system performance, with total travel times only 1% higher than System Optimal solution while reducing travel times by up to 22% compared to User Equilibrium solution. These results highlight our approach’s ability to achieve system-optimal efficiency while maintaining fair access across different OD pairs.