Managing bottleneck congestion with tradable credit scheme under demand uncertainty

In this paper, we focus on managing a single bottleneck with tradable credit scheme (TCS) when demand uncertainty is considered. Assuming that the demand is stochastic and follows a uniform distribution, we seek to find a universal scheme that can accommodate various realized demand scenarios. To this end, we first discuss about the mean trip cost-based optimal TCS, which minimizes the mean trip cost across all demand realizations. Unfortunately, we find that such a scheme will always be nullified with a zero credit price. To address this problem, we propose two modified schemes, a demand-adaptive scheme (DA-TCS) and a government-intervened scheme (GI-TCS). Then a Monte Carlo simulation is conducted on an example to evaluate the performances of the two proposed schemes. Results suggest that the generalized travel cost under the GI-TCS is statistically lower than that under the DA-TCS, but the latter enjoys a lower variance in its performance.