[Paper Review] A kinematic wave theory of capacity drop

Wen‐Long Jin, Qijian Gan+1 more|arXiv (Cornell University)|Oct 9, 2013

Traffic control and management1 references4 citations

TL;DR

This paper proposes a kinematic wave model of capacity drop at active bottlenecks using a discontinuous boundary flux function within a continuous fundamental diagram framework. By modeling capacity drop as a result of upstream queue formation and enforcing an entropy condition via a discontinuous flux, the model uniquely solves the Riemann problem and reproduces observed capacity drop without requiring discontinuous flow-density relations.

ABSTRACT

Capacity drop at active bottlenecks is one of the most puzzling traffic phenomena, but a thorough understanding is practically important for designing variable speed limit and ramp metering strategies. In this study, we attempt to develop a simple model of capacity drop within the framework of kinematic wave theory based on the observation that capacity drop occurs when an upstream queue forms at an active bottleneck. In addition, we assume that the fundamental diagrams are continuous in steady states. This assumption is consistent with observations and can avoid unrealistic infinite characteristic wave speeds in discontinuous fundamental diagrams. A core component of the new model is an entropy condition defined by a discontinuous boundary flux function. For a lane-drop area, we demonstrate that the model is well-defined, and its Riemann problem can be uniquely solved. We theoretically discuss traffic stability with this model subject to perturbations in density, upstream demand, and downstream supply. We clarify that discontinuous flow-density relations, or so-called "discontinuous" fundamental diagrams, are caused by incomplete observations of traffic states. Theoretical results are consistent with observations in the literature and are verified by numerical simulations and empirical observations. We finally discuss potential applications and future studies.

Motivation & Objective

• To resolve the theoretical and empirical conflict between observed capacity drop and discontinuous fundamental diagrams in traffic flow theory.
• To develop a physically consistent model of capacity drop that avoids infinite wave speeds associated with discontinuous flow-density relations.
• To provide a mathematically well-defined framework for modeling capacity drop in kinematic wave theory using a discontinuous entropy condition.
• To enable application of the model in traffic control strategies such as variable speed limits and ramp metering by ensuring unique, stable solutions.
• To reconcile observed capacity drop with continuous fundamental diagrams by attributing apparent discontinuities to incomplete state observations rather than intrinsic flow discontinuities.

Proposed method

• Introduces a discontinuous boundary flux function based on upstream demand and downstream supply to model capacity drop, replacing traditional continuous flux functions.
• Applies an entropy condition defined by the discontinuous flux to select unique, physically meaningful solutions to the Riemann problem at bottlenecks.
• Uses continuous fundamental diagrams for upstream and downstream links, avoiding non-differentiable points and infinite wave speeds.
• Derives the solution of the Riemann problem for a lane-drop bottleneck, proving uniqueness under the proposed entropy condition.
• Incorporates the model into the Cell Transmission Model (CTM) for simulating network-wide traffic dynamics and control strategies.
• Validates the model through theoretical analysis, numerical simulations, and consistency with empirical observations of capacity drop.

Experimental results

Research questions

• RQ1 How can capacity drop be modeled within kinematic wave theory without relying on discontinuous fundamental diagrams?
• RQ2 What is the role of upstream queue formation in triggering capacity drop, and how can it be captured via a boundary flux function?
• RQ3 Can a discontinuous flux function at the bottleneck yield a unique and physically consistent solution to the Riemann problem?
• RQ4 How does the proposed model reconcile observed capacity drop with continuous flow-density relations in empirical data?
• RQ5 To what extent can this model be used to design effective traffic control strategies like variable speed limits and ramp metering?

Key findings

• The model uniquely solves the Riemann problem for a lane-drop bottleneck by using a discontinuous boundary flux function as an entropy condition.
• Theoretical analysis confirms that discontinuous flow-density relations are not inherent but result from incomplete observation of traffic states, not physical discontinuities.
• The model avoids infinite characteristic wave speeds by maintaining continuous fundamental diagrams, resolving a key theoretical limitation of prior models.
• Numerical simulations and empirical observations confirm that the model reproduces the key features of capacity drop: reduced discharge flow when queues form.
• The model is compatible with the Cell Transmission Model (CTM), enabling simulation of network-wide traffic dynamics and control strategy evaluation.
• The solution is uniquely determined by the flux function $ q = \begin{cases} d_1, & d_1 \leq s_2 \\ \min\{s_2, C_*\}, & d_1 > s_2 \end{cases} $, which captures capacity drop when upstream demand exceeds downstream supply.