Dynamic Resequencing for High-Throughput Autonomous Traffic Coordination

This direction is merged into another direction: mixed-autonomy coordination, FIFO bottleneck. Its last version stays readable and citable here.

Version 1 of 147 papers69 conceptsfrozen 2026-10-04

Traditional traffic management—from the humble stoplight to early autonomous intersection protocols—relies on a simple principle: First-In-First-Out (FIFO). While chronologically "fair," this rigid ordering is a primary driver of urban congestion. If a slow-moving truck arrives at an intersection a split-second before a fast-moving car, FIFO forces the car to wait, creating a "void" in the intersection that could have been filled.

A growing body of research suggests that the key to unlocking the full capacity of our infrastructure lies in dynamic resequencing: treating traffic not as a queue, but as a scheduling problem where the order of passage is optimized in real-time.

The Inefficiency of Chronological Ordering

In human-driven traffic, FIFO is a necessity because drivers lack the coordination to safely "jump the line." However, as Yue Zhang argues in A Decentralized Optimal Control Framework for Connected Automated Vehicles at Urban Intersections with Dynamic Resequencing, strict arrival-based sequencing fails to account for the varying speeds and distances of approaching vehicles. This rigidity prevents the system from exploiting available gaps, particularly when vehicles on shorter paths are forced to wait for distant arrivals on longer paths.

The cost of this rigidity is high. Tsz-Chiu Au and Peter Stone point out in Enforcing Liveness in Autonomous Traffic Management that processing requests chronologically often forces large streams of high-speed traffic to slow down for a single vehicle, failing to optimize the "crossing windows" for groups. By relaxing these constraints, researchers like Andreas A. Malikopoulos have demonstrated travel time reductions of up to 34% in simulated urban environments.

From Queues to Combinatorial Optimization

If we move beyond FIFO, how do we determine the "correct" order? The research identifies several mechanisms for real-time reordering:

  1. Mixed-Integer Linear Programming (MILP): This is the most common mathematical tool for resolving spatial conflicts. By treating the crossing order as a set of binary variables, Andreas A. Malikopoulos shows in Time-Optimal Coordination for Connected and Automated Vehicles at Adjacent Intersections how to optimize crossing sequences based on arrival windows and safety headways rather than just arrival order.
  2. Tree Search and Heuristics: For high-density scenarios where MILP becomes computationally heavy, Huile Xu and Christos G. Cassandras employ Monte Carlo Tree Search (MCTS) to evaluate passing orders in complex conflict areas like work zones. Similarly, Adittyo Paul and Jiaoyang Li developed Switchable-Edge Search (SES), which allows agents to bypass delayed peers by dynamically re-optimizing the edges of a coordination graph.
  3. Flexible Order (FO) Policies: Changliu Liu proposes an event-driven stochastic framework where vehicles reorder themselves based on "arrival gaps." In Analyzing Traffic Delay at Unmanaged Intersections, Liu demonstrates that FO policies prevent the exponential delay "explosions" typical of high-density FIFO traffic.

Batching and Streamlining

A recurring intuition across these papers is that throughput is maximized when we group similar movements together. In rail logistics, Sándor P. Fekete notes that clustering trips in the same direction reduces the overhead of switching directions. This same logic applies to intersections.

Pavankumar Tallapragada introduces the concept of "bubbles"—dynamic clusters of vehicles that cross the intersection as a single unit. By optimizing the crossing order of these bubbles rather than individual vehicles, the system reduces the energy-intensive "stop-and-go" waves. Similarly, Can Zhou and Wenbin Xiao argue that the specific grouping of traffic streamlines within a phase has a greater impact on efficiency than the chronological sequence of the phases themselves.

Beyond the Intersection: Merging and Logistics

The benefits of dynamic resequencing extend to highway merging and logistics. Traditional ramp merging often uses distance-based FIFO, which Yang Zhou 0019 and Meng Wang argue is suboptimal. By using MILP to balance density between the mainline and the ramp, they reduce the time required for vehicles to converge to safe distances.

In the realm of energy efficiency, Zhouqiao Zhao and Matthew J. Barth demonstrate in Optimal Control-Based Ramp Merging System for Connected and Automated Electric Vehicles that evaluating all possible merging permutations can identify sequences that minimize aggregate fuel consumption, achieving up to 47% fuel savings.

Even in container terminals, the shift from rigid to flexible service is transformative. Managing Relocation and Delay in Container Terminals with Flexible Service Policies shows that allowing out-of-order retrievals—delivering a "blocking" item early rather than moving it to temporary storage—can reduce physical movements and truck wait times by over 20%.

Conclusion

The transition from FIFO to dynamic resequencing represents a fundamental shift in how we conceptualize movement through shared resources. By prioritizing global flow over individual arrival times, these frameworks transform bottlenecks into optimized conduits, proving that the most efficient path through a conflict is rarely the one that arrived first.

Papers in this version

History & citationVersion 1 · 2026-10-04

Merged into another direction · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_dynamic_resequencing_for_high_throughput_d7287e2c,
  title = {Dynamic Resequencing for High-Throughput Autonomous Traffic Coordination},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/dynamic-resequencing-for-high-throughput-autonomous-traffic-coordination/rdr_d7287e2c8af24b88aa133a541ce04040},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 1, frozen text):

@misc{lacuna_dynamic_resequencing_for_high_throughput_d7287e2c_v1,
  title = {Dynamic Resequencing for High-Throughput Autonomous Traffic Coordination},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/dynamic-resequencing-for-high-throughput-autonomous-traffic-coordination/rdr_d7287e2c8af24b88aa133a541ce04040/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}