LWR Model

The Lighthill-Whitham-Richards Model
Emanuel P. Fontelles
T
opicos de Mec
anica Estatstica
June, 2016
Introduction
Continuity Equation:
The continuity equation is a partial differential or difference equation for
the macroscopic quantities (density) and V (speed) or Q (flow). Due to
the hydrodynamic relation flow equals density times speed these two
options are equivalent. While the parameterless continuity equation is
always valid, we need an additional equation for the flow or speed to
complete the model.
Since the continuity equation is completely determined by the geometry of the road infrastructure, the macroscopic models differ in their
modeling of speed or flow, only.
Continuity Equation:
tot
t
(tot V )
x
=0
8.1 Model Equations
Lighthill-Whitham-Richards (LWR) Model:

In 1955 and 1956, Lighthill and Whitham, and independently also
Richards, proposed the following static relation to complement the
continuity equation:
Q(x, t) = Qe ((x, t))
or V (x, t) = Ve ((x, t))
8.1 Model Equation
Inserting the assumption LWR on continuity equation and applying the

dQ()
Qe
=
, yield the simplest form of a Lighthill
chain rule
x
d x
Whitham Richards model
dQ()
+
=0
t
d x

e
+ Ve + dV
=0
d
t
x
LWR Model
LWR Model
8.1 Model Equations

The schematic example of a static speed-density relation to LWR
models.
Figure: Schematic example of a static speed-density relation for LWR models.
Lighthill-Whitham-Richards (LWR) Model

Proposed by Lighthill and Whitham (1955) and Richards (1956), the LWR
model describes traffic flow on a single one-way road without entrances or
exits:

t + V () = 0
x
= density
V () = preferred velocity, a given nonincreasing function of ,
nonnegative for between 0 and m (thejam density)
Predicts piece-wise smooth density, with transitions between regions
approximated by shocks
Problem: Doesnt adequately describe the motion of cars passing
through shocks (cars change velocity instantaneously)
8.2 Propagation of Density Variations

One of important parameters is the propagation velocity c(), this
parameter is related with density variation, and using the traveling-wave
ansatz
(x, t) = 0 (x ct)
0 (x) = (x, 0)
We can derive the propagation velocity c.

The partial differential equations are nonlinear wave equations
Describes the propagation of kinematic waves
Propagation velocity c
c() =
dQe
d
d(Ve ())
d
8.2 Propagation of Density Variations
Figure: Propagation velocity c = Qe0 () of density and speed variations in the

LWR model in comparison with the local vehicle speed V (). In the fundamental
diagram (top), c is given by the slope of the tangent while V is given by the
slope of the secant through the origin.
8.3 Shock Waves
8.3.1 Formation
Continuous LWR models describes density variations of constant
amplitudes but with varying local propagation velocities
How to density variations affects the traffic?
How to behave the vehicles on a stop-go-wave? (Density Profile)
8.3 Shock Waves

8.3.1 Formation
Figure: Emergence of shock waves due to the density-dependent local propagation

velocities in the LWR model.
It is noted a discontinuous transition indicated by the vertical line

which is the defining feature of a shock wave or shock front.
8.3 Shock Waves

8.3.2 Derivation of the Propagation Velocity
The transitions free congested and congested free in the LWR
models are unrealistic
The propagation of the wave positions, the motion of transition zones
to and from extended congested traffic, are described realistically
Figure: A shock front at location x12 (t) with constant flow and density within
small road sections on either side.
8.3 Shock Waves

8.3.2 Derivation of the Propagation Velocity
To derive the propagation velocities, to find the velocity c12 = dxdt12 , we
will express the rate of change in the number of vehicles, dn
dt , in two
different ways:
From the conservation of vehicles
And with the definition of the density
dn
= Q1 Q2
dt
dx12
n = 1 x12 + 2 (L x12 ) dn
dt = (2 1 ) dt = (2 1 )c12
Comparing both expressions for
c12 =
dn
dt
gives us
Q2 Q1
Qe (1 ) Qe (2 )
=
2 1
2 1
Propagation of Shock Waves
8.3 Shock Waves

8.3.3 Vehicle Speed Versus Propagation Velocities
The LWR model allows us to extract all relevant velocities directly
from the fundamental diagram
Figure: Visualization of how to obtain vehicle speeds and propagation velocities

from the fundamental diagram.
8.3 Shock Waves
8.3.3 Vehicle Speed Versus Propagation Velocities

1
The propagation velocity of density variations c() = Qe0 () is given

by the slope of the fundamental diagram.
The propagation velocity of shock fronts c12 is given by the slope of

the secant connecting points of the fundamental diagram
corresponding to traffic on either side of the front.
The vehicle speed Ve = Qe ()/ is given by the slope of the secant

connecting the origin with the corresponding point on the
fundamental diagram.
8.4 Numerical Solution

This is generally done by finite-difference methods: Space is divided
into cells of generally constant length x (although this is not
required), and time in the index k increasing in the downstream
direction.
The equations for the LWR models have the form of a so-called
conservation law for which many specialized explicit solution
methods are available.
Figure: Cells of the CTM for a simple straight road and definition of the relevant
quantities for the supply demand method.
8.4 Numerical Solution
The most common integration method for LWR models is the

Godunov scheme
Courant-Friedrichs-Levy condition (CFL condition) for LWR models
These discretization errors lead to the phenomenon of numerical
diffusion which increases with the cell size
8.5 LWR Models with Triangular Fundamental Diagram
The simplest of the Lighthill-Whitham-Richards models uses a

triangular fundamental diagram
The continuous version is called section-based model.

The discrete version is formulated as an iterated coupled map with
time and space discretized into time steps and cells, respectively, and
supplemented by a special supply-demand update rule. This model
is known as cell-transmission model
Figure: Triangular fundamental diagram, as used in the cell transmission model

and the section-based model.

8.5.1 Model Parameters
For vehicular traffic, the maximum density corresponds to the inverse of

the minimum distance headway leff , which is the average vehicle length
plus the average minimum gap s0 in stopped traffic:
leff = s0 + l =
1
max

8.5.1 Model Parameters
For vehicular traffic, the maximum density corresponds to the inverse of

the minimum distance headway leff , which is the average vehicle length
plus the average minimum gap s0 in stopped traffic:
leff = s0 + l =
1
max
8.5.9 Examples
8.6 Diffusion and Burgerss Equation
Shock waves are not very realistic in describing traffic flow. Furthermore,
the associated discontinuities turn out to be problematic for a numerical
solution - at least for non-triangular fundamental diagrams. As a simple
phenomenological solution, one may introduce diffusion to the continuity
equation by adding a diffusion term D/x 2 with the diffusion constant
D > 0:
Questions?

LWR Model

Uploaded by

Copyright:

Available Formats

LWR Model

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

LWR Model

Uploaded by

Copyright:

Available Formats

The Lighthill-Whitham-Richards Model

8.1 Model Equations

Lighthill-Whitham-Richards (LWR) Model:

8.1 Model Equation

Inserting the assumption LWR on continuity equation and applying the

8.1 Model Equations

Figure: Schematic example of a static speed-density relation for LWR models.

Lighthill-Whitham-Richards (LWR) Model

8.2 Propagation of Density Variations

We can derive the propagation velocity c.

8.2 Propagation of Density Variations

Figure: Propagation velocity c = Qe0 () of density and speed variations in the

8.3 Shock Waves

8.3 Shock Waves

Figure: Emergence of shock waves due to the density-dependent local propagation

It is noted a discontinuous transition indicated by the vertical line

8.3 Shock Waves

8.3 Shock Waves

Propagation of Shock Waves

8.3 Shock Waves

Figure: Visualization of how to obtain vehicle speeds and propagation velocities

8.3 Shock Waves

8.3.3 Vehicle Speed Versus Propagation Velocities

The propagation velocity of density variations c() = Qe0 () is given

The propagation velocity of shock fronts c12 is given by the slope of

The vehicle speed Ve = Qe ()/ is given by the slope of the secant

8.4 Numerical Solution

8.4 Numerical Solution

The most common integration method for LWR models is the

8.5 LWR Models with Triangular Fundamental Diagram

The simplest of the Lighthill-Whitham-Richards models uses a

The continuous version is called section-based model.

8.5 LWR Models with Triangular Fundamental Diagram

Figure: Triangular fundamental diagram, as used in the cell transmission model

8.5 LWR Models with Triangular Fundamental Diagram

For vehicular traffic, the maximum density corresponds to the inverse of

8.5 LWR Models with Triangular Fundamental Diagram

For vehicular traffic, the maximum density corresponds to the inverse of

8.5 LWR Models with Triangular Fundamental Diagram

8.6 Diffusion and Burgerss Equation

You might also like