Calming traffic in unsignalized bidirectional streets by introducing one-lane bottlenecks
Abstract
A common strategy to calm traffic in an unsignalized, bidirectional street consists in narrowing the street to a single lane at one or more locations. These locations become bottlenecks. Two policies are commonly used there: first-in-first-out (FIFO), and directional priority (DP), which means one of the directions has priority. This paper examines the performance of these two policies on streets with one or more bottlenecks. Two measures of performance are considered: the expected delay (evaluated for typical situations with low demand) and the system capacity. The latter is defined as the pair of flows (d1, d2) that depart the street in the two opposite directions when the system is oversaturated with queues. For a street with a single bottleneck, the capacity under both FIFO and DP is not unique: In both cases, the values of (d1, d2) form a continuous decreasing curve. As expected, we find that arrival flow pairs (a1, a2) below this curve can get through the system without generating queues. Those above cannot. For a single bottleneck it is shown that FIFO yields lower expected delay than DP under low demand. For high demand, DP's capacity exceeds FIFO's when the flows are balanced, but are lower otherwise. For multiple bottlenecks, the expected total delay on the street under low demand is that of a single bottleneck multiplied by the number of bottlenecks. The multi-bottleneck capacity under FIFO turns out to be the same as for a single bottleneck. For DP, traffic near capacity is complex. Capacity points turn out to be numerous and disjointed. Formulas are derived for their values. Higher capacities are typically obtained for longer streets. Curiously, these capacity points are often above the capacity curve of a street with a single FIFO or DP bottleneck, indicating that an extra bottleneck can increase capacity.
Related
- Scalability bottlenecks in entanglement-based satellite quantum key distribution: downlink asymmetry, orbit dependence and finite-key effects
- Transition states and dynamical bottlenecks in the Li-NC <-> Li-CN isomerizing reaction with the Lagrangian betweenness
- Does Your Self-Consistent Field Program Know How Wrong It Is? Introducing Quantum Chemistry Uncertainty Propagation.
- Unveiling the Scaling Potential of Drain Merge through Active (DMtA) in CFETs: Breaking the Super-Via Bottlenecks and Unlocking New PPA Boosters
- Bottlenecks in Hamiltonian-Adaptive Resolution Simulation Method for Modeling Interfaces
BibTeX
@misc{knoop2026calming,
title = {{Calming traffic in unsignalized bidirectional streets by introducing one-lane bottlenecks}},
author = {Victor L. Knoop and Carlos F. Daganzo},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2607.13650},
eprint = {2607.13650},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.13650},
note = {CC BY-NC-SA 4.0},
}
RIS
TY - UNPB
TI - Calming traffic in unsignalized bidirectional streets by introducing one-lane bottlenecks
AU - Victor L. Knoop
AU - Carlos F. Daganzo
PY - 2026
DA - 2026/09/16
PB - arXiv
DO - 10.48550/arxiv.2607.13650
UR - https://arxiv.org/abs/2607.13650
AB - A common strategy to calm traffic in an unsignalized, bidirectional street consists in narrowing the street to a single lane at one or more locations. These locations become bottlenecks. Two policies are commonly used there: first-in-first-out (FIFO), and directional priority (DP), which means one of the directions has priority. This paper examines the performance of these two policies on streets with one or more bottlenecks. Two measures of performance are considered: the expected delay (evaluated for typical situations with low demand) and the system capacity. The latter is defined as the pair of flows (d1, d2) that depart the street in the two opposite directions when the system is oversaturated with queues. For a street with a single bottleneck, the capacity under both FIFO and DP is not unique: In both cases, the values of (d1, d2) form a continuous decreasing curve. As expected, we find that arrival flow pairs (a1, a2) below this curve can get through the system without generating queues. Those above cannot. For a single bottleneck it is shown that FIFO yields lower expected delay than DP under low demand. For high demand, DP's capacity exceeds FIFO's when the flows are balanced, but are lower otherwise. For multiple bottlenecks, the expected total delay on the street under low demand is that of a single bottleneck multiplied by the number of bottlenecks. The multi-bottleneck capacity under FIFO turns out to be the same as for a single bottleneck. For DP, traffic near capacity is complex. Capacity points turn out to be numerous and disjointed. Formulas are derived for their values. Higher capacities are typically obtained for longer streets. Curiously, these capacity points are often above the capacity curve of a street with a single FIFO or DP bottleneck, indicating that an extra bottleneck can increase capacity.
ER -