Source-linked AI summary
Platoons of connected vehicles can double throughput in urban roads
Jennie Lioris, Ramtin Pedarsani, Fatma Yildiz Tascikaraoglu, Pravin Varaiya
TL;DR
Urban intersections limit network throughput because their capacity is much lower than that of connecting roads, motivating platoon-based capacity increases. The paper analyzes queuing models and simulates a road network to assess this approach. It concludes that increasing saturation flows can proportionally increase demand without changing signal control, while also offering queue and delay reductions through shorter cycles.
Problem
Intersections are urban-road bottlenecks, and it is unclear how increased saturation flows affect throughput, queues, and travel time in signalized networks.
Method
The paper combines three queuing-model analyses with a mesoscopic simulation of a signalized road network using fixed-time controls and offsets.
Results
A saturation-flow increase by factor Γ supports demand increased by the same factor Γ with no increase in queuing delay or travel time under the same signal control.
Takeaways & Limitations
Γ = 2 or 3 is reported as technically achievable, allowing capacity gains to be shared between increased demand and reduced queuing delay through shorter cycle times.
Takeaways & Limitations
The analysis considers the limiting case of 100 percent connected-vehicle penetration and does not establish how small the gain factor is at lower penetration rates.
Abstract
from arXiv · showhide
Intersections are the bottlenecks of the urban road system because an intersection's capacity is only a fraction of the flows that the roads connecting to the intersection can carry. This capacity can be increased if vehicles can cross the intersections in platoons rather than one by one as they do today. Platoon formation is enabled by connected vehicle technology. This paper assesses the potential mobility benefits of platooning. It argues that saturation flow rates, and hence intersection capacity, can be doubled or tripled by platooning. The argument is supported by the analysis of three queuing models and by the simulation of a road network with 16 intersections and 73 links. The queuing analysis and the simulations reveal that a signalized network with fixed time control will support an increase in demand by a factor of (say) two or three if all saturation flows are increased by the same factor, with no change in the control. Furthermore, despite the increased demand vehicles will experience the same delay and travel time. The same scaling improvement is achieved when the fixed time control is replaced by the max pressure adaptive control. Part of the capacity increase can alternatively be used to reduce queue lengths and the associated queuing delay by decreasing the cycle time. Impediments to the control of connected vehicles to achieve platooning at intersections appear to be small.
1 Introduction
Connected vehicle technology is being investigated for information, safety, and mobility applications. This paper focuses on using connected vehicles to organize intersection-crossing platoons and increase urban road capacity.
- Connected vehicle technology: Connected vehicle technology supports vehicle-to-infrastructure and vehicle-to-vehicle information exchange for transportation management and safety applications.The cited programs target improved real-time traffic data, hazard awareness, and driver or infrastructure warnings.
- Prior research: Research has examined vehicular ad-hoc networks and vehicle control for cooperative cruise control, merging, and safely crossing intersections.These studies indicate potential for automating selected driving functions.
- Research landscape: Connected vehicle research also includes commercial connectivity and driver-assistance technologies, but these efforts are not primarily aimed at increasing mobility through intersection platooning.The paper distinguishes vehicle-resident assistance systems from V2V-based approaches.
- Paper objective: The paper proposes using connected vehicle technology to organize vehicles into platoons and increase intersection capacity by a factor of two to three.It evaluates the mobility implications through queuing models and a network simulation.
2 Intersection capacity
Urban intersections can carry far less flow than their connecting roads because signal phasing restricts simultaneous movements. Platooning reduces vehicle headways, raising saturation flow and potentially multiplying intersection capacity.
- Intersection capacity: 3,800 vph is the intersection capacity for eight movements with 1,900 vph lane capacity and a maximum 0.25 effective green ratio per movement.Only two movements can safely operate simultaneously in the four-approach example, making the intersection capacity one quarter of the connecting-road capacity.
- Platoon flow rates: 4800 vph is the platoon saturation flow at 0.75 s headway and 45 mph, 3.8 times the observed 1272 vph and 2.5 times HCM’s 1900 vph.At 30 mph, a 40-foot space headway yields 3960 vph, twice HCM’s 1900 vph and up to three times observed intersection rates.
- Platoon mechanism: Platooning decreases intersection-crossing headway by a gain factor Γ and increases saturation flow by the same factor Γ.ACC or CACC can maintain short headways when queued vehicles begin moving at a green signal.
- Platoon formation: Connected vehicles can form platoons using sensing, adaptive cruise control, vehicle communication, and signal-phase information.The paper describes ACC for tight headway control and CACC for even shorter headways.
- Operational scope: The proposal applies CACC only to vehicles stopped at intersections, distinguishing it from using CACC to reduce headways continuously along roads.The paper argues that increasing road capacity alone does not increase urban-network throughput when intersections remain the bottleneck.
- Market penetration: 1.67 is the illustrative saturation-flow gain with two manually driven vehicles in a 12-vehicle queue, versus approximately 2.67 when all vehicles are connected.The models and simulations can represent any average saturation-flow multiplier Γ, though the multiplier depends on connected-vehicle penetration and technology.
3 Three predictions
The three queuing models predict that proportionally increasing saturation flows and demand increases throughput while preserving delay and travel time under idealized conditions. Finite link storage and cycle-time constraints qualify these benefits, while reducing cycle time can reduce queues and delay when capacity gains permit it.
- Model comparison: The M/M/1 analysis is optimistic because it replaces signalized on-off service with a constant saturation rate.The on-off model instead finds throughput gains without a corresponding delay benefit when demand and service rates scale together.
- Throughput and delay: When saturation flow and demand increase together, queue length grows by Γ while average delay remains essentially unchanged if cycle time is fixed.This prediction follows from the on-off queue analysis for Γ increased to 2 or 3.
- Throughput and delay: A factor Γ increase in saturation flow lets the network support throughput increased by Γ while queuing delay and travel time remain unchanged.The on-off queue and fluid network produce this consistent prediction.
- Finite capacity: The factor-Γ scaling predictions rely on idealized infinite-capacity queues or conditions where storage does not block arrivals.With finite capacities, rigorous network analysis is difficult and bottlenecks depend on signal control and offsets.
- Cycle time: Reducing cycle time while preserving effective green ratios reduces queue length and delay, allowing some platooning capacity gains to be used for shorter queues.The fluid model predicts queue reduction under service-rate speedup, and the paper notes that platooning can provide the needed gain.
4 Case study
The case study simulates a 16-intersection, 73-link network to test proportional scaling of demand and saturation flows under fixed-time and max-pressure controls. Queue lengths grow approximately with the scaling factor, while faster phase switching reduces queues.
- Network and experiments: 16 signalized intersections, 73 links, and 106 queues define the simulated Los Angeles study network.Each queue represents a turn movement.
- Network and experiments: Demand and saturation flows are jointly scaled by Γ = 1.0, 1.5, 2.0, 2.5, and 3.0, with fixed-time, MP4, and MP6 controls compared.MP4 allows four phase changes per cycle; MP6 allows six.
- Queue lengths: Γ = 3 produces approximately twice the total queue length of Γ = 1.5 under fixed-time, MP4, and MP6 control.This matches the predicted proportional growth in queue lengths.
- Queue lengths: 1:1.26:1.76:2.31:2.97 are the fixed-time queue-sum ratios as Γ increases in the proportion 1:1.5:2.0:2.5:3.0.The queue sums remain bounded while growing roughly in proportion to the scale factor.
- Queuing delay: Mean queuing delay remains within a narrow range for the examined queue as demand and saturation flows scale proportionally.Under fixed-time control, queue x(138,140) changes from 22.35 to 20.69 seconds across the tested scale factors.
- Faster phase switching: MP6 produces smaller queues than MP4 at each demand level, with MP4-to-MP6 queue ratios near the predicted 1.5.The observed ratios range from 1.45 to 1.6 across Γ = 1 to 3.
5 Conclusion
The paper argues that connected-vehicle-enabled platooning can substantially raise intersection saturation flow and urban road capacity. Queuing analysis and simulation indicate proportional demand scaling without increased delay under unchanged signal control, subject to important penetration and short-link limitations.
- Capacity from platooning: 4800 vph per movement is the proposed platoon saturation flow, enabled by 0.75-second headways at 45 mph or 0.7-second headways at 30 mph.This represents a two- to threefold increase in intersection capacity relative to current conditions described in the paper.
- Capacity from platooning: CACC can achieve shorter platoon headways than ACC through superior car-following performance and can permit lane changes within a platoon.The paper relates these capabilities to freeway platooning experiments using ACC vehicles augmented with V2V communication.
- Network-level effects: A factor Γ increase in saturation flow supports the same factor Γ increase in demand without increasing queuing delay or travel time under the same signal control.The associated queues also grow by Γ, which can make throughput gains sub-linear if links saturate.
- Network-level effects: 200 to 300 percent is the paper’s stated potential urban-road infrastructure productivity increase when Γ = 2 or 3 is technically achievable.Shortening cycle time can instead allocate part of this gain to reducing queues and queuing delay.
- Limitations: 100 percent penetration is assumed, while the penetration-dependent gain σ(p)Γ is unknown in this study.Short urban links can also reduce upstream saturation flow as queues build up, limiting the full productivity benefit.