Traffic Flow (LWR Model)
Also known as: LWR model, Traffic wave, Kinematic wave theory
The Lighthill-Whitham-Richards (LWR) model is a macroscopic traffic flow model that treats traffic as a compressible fluid, applying conservation of vehicles and a flow-density relationship. Introduced independently by Lighthill and Whitham (1955) and Richards (1956), the model predicts traffic wave propagation, congestion formation, and bottleneck behavior on highways.
Key highlights
- Simple mathematical foundation based on conservation laws, making it computationally efficient and transparent
- Captures qualitative traffic phenomena: shock waves, congestion formation, and hysteresis in flow-density relationships
- Suitable for macroscopic planning studies and quick assessment of lane closure impacts or demand changes
- Can be calibrated from traffic counting data (flow, occupancy) without requiring detailed vehicle trajectories
- Enables analytical solutions for idealized cases, providing insight into traffic dynamics
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
LWR modeling is appropriate for highway corridor analysis, incident management, and capacity planning. It is efficient for long stretches of freeway with homogeneous traffic. However, it is less suitable for complex intersections with signal control, arterial streets with frequent stops, or situations requiring detailed driver behavior (lane changing, acceleration profiles). For network-scale analysis, link-based models combining LWR with routing are effective.
Strengths & limitations
- Simple mathematical foundation based on conservation laws, making it computationally efficient and transparent
- Captures qualitative traffic phenomena: shock waves, congestion formation, and hysteresis in flow-density relationships
- Suitable for macroscopic planning studies and quick assessment of lane closure impacts or demand changes
- Can be calibrated from traffic counting data (flow, occupancy) without requiring detailed vehicle trajectories
- Enables analytical solutions for idealized cases, providing insight into traffic dynamics
- Assumes homogeneous traffic (uniform driver behavior and vehicle types); does not capture variability or heterogeneity
- Oversimplifies driver behavior: assumes vehicles instantaneously adjust speed to match density, missing acceleration/deceleration
- Single fundamental diagram may not accurately represent all flow conditions; real traffic shows significant scatter
- Cannot model individual lanes or lane changing; requires multi-lane extensions
- Accuracy decreases during highly congested conditions where vehicle queuing dynamics dominate
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
What is the fundamental diagram, and how do I measure it from field data?
The fundamental diagram plots flow (vehicles/hour) vs. density (vehicles/mile). Measure using inductive loop detectors or cameras: record flow and occupancy (surrogate for density) in 5-15 minute intervals. Plot flow vs. occupancy; the resulting scatter is the fundamental diagram. Fit a linear or triangular model for the LWR solver.
How do I model lane closures or incidents using LWR?
Reduce the capacity (maximum flow) of the affected segment for the duration of the closure. The model automatically generates backup upstream and a shock wave propagating backward. The severity depends on demand relative to reduced capacity.
Can LWR model merging traffic at on-ramps?
Basic LWR assumes homogeneous flow. Merging introduces discontinuities (shock waves). Extended models include on-ramp flow as a boundary condition and use merge rules (first-in-first-out, proportional allocation) to split downstream flow. This requires discretizing on-ramps as separate links.
When does LWR lose accuracy, and should I use a microscopic simulator?
LWR becomes inaccurate in highly congested conditions with stop-and-go traffic, at intersections with signals, or when lane changing and acceleration profiles matter. Use microscopic simulation (agent-based, driver-behavior models) for these cases. LWR is best for preliminary analysis and macroscopic planning.
Sources
- 1.Lighthill, M. J., & Whitham, G. B. (1955). On kinematic waves I. Flow movement in long rivers. Proceedings of the Royal Society A, 229(1178), 281-316.
- 2.Richards, P. I. (1956). Shock waves on the highway. Operations Research, 4(1), 42-51.
- 3.Daganzo, C. F. (1994). The cell transmission model: A dynamic representation of highway traffic consistent with the hydrodynamic theory. Transportation Research Part B, 28(4), 269-287.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 3). Traffic Flow (LWR Model). ScholarGate. https://scholargate.app/civil-engineering/traffic-flow