1 Introduction
2 Methodology and configuration
2.1 Incident scenarios
Scenario | TIM techniques modelled | |||
---|---|---|---|---|
S1: Collision | 1.1 Close all 3 lanes | 1.2 Incident screen | 1.3 Close 2 of 3 lanes | 1.4 Tow in off-peak |
S2: Weather event | 2.1 Close both 2 lanes | 2.2 Contraflow | 2.3 VMS + Speed limit | |
S3: LGV breakdown | 3.1 Close 1 lane of 3 | 3.2 Repair on-site | 3.3 Tow in off-peak | (closing 2 lanes) |
S4: Obstruction | 4.1 Close both 2 lanes | 4.2 Contraflow | 4.3 Close blocked lane |
2.2 Modelling traffic demand
3 Modelling of dynamic traffic flow and queuing
3.1 Basic model description
Symbol | Definition | Units | Notes |
---|---|---|---|
m,n
| Number of lanes | Available or normal number of lanes according to context | |
μ | Lane capacity | PCU/ha
| Normal capacity of one lane for flowing traffic |
q
| Flow | veh/h | May be subscripted corresponding to sections in Fig. 2
|
ρ | Density | veh/km |
k is also commonly used |
v
| Speed | km/h | Strictly space-mean, though time-mean is often measuredb
|
V
f
| Free-flow speed | km/h | Maximum speed with no other traffic |
ρ
c
| Critical density | veh/km | Point at which flow breaks down and congestion starts |
v
c
| Critical speed | km/h | Corresponding to critical density |
ρ
max
| Jam density | veh/km | Notional density at which speed is zero (k
j
sometimes used) |
α | Speed/flow slope | km/veh | Determines how speed decreases with flow if uncongested |
λ | Jam spacing | m | Effective minimum road space required by vehicle |
τ | Response time | sec | Effective reaction and response time of driver |
w,w
f
| Wave speed | km/h | Speed at which a disturbance propagates upstream |
β,γ,χ | Capacity factors | Parameters used to adjust lane capacities for calibration | |
δ
i
| Diversion factors | Diversion could vary with time (not used in modelling) | |
v
ij
| Wave speeds | km/h | Speeds of propagation of state boundaries in queue |
r
i
| Queue reach | km | Upstream position of tail of queue at various times |
t
i
| Event times | h | Times at which traffic or queue state changes |
X
| Max. queue extent | km | Maximum length of queue, head to tail |
R
| Max. queue reach | km | Maximum distance of queue tail from incident site |
S
i
| Time-space ‘area’ | km-h | Measure of amount of queuing in a given period |
D
| Total delay | veh-h | Allows for time that would have been taken at free-flow |
3.2 Capacity adjustments in and around the queue
3.3 Propagation speed of queue head-wave and other disturbances
3.4 Modelling queue and delay across several incident phases
3.5 Implementation of CTM-v simulation
3.6 A simple method of allowing for diversion
3.7 Summary of differences between queue model and CTM-v
Context | Queue model | CTM-v simulation |
---|---|---|
Time dynamics | Resolves only to the level of TIM phases, allowing different but constant arriving and capacity flows within each phase, with arrivals obtained by averaging 15-min profile values. | Discrete time model with 6 s time steps enabling not only detailed modelling of variations in capacity, demand and arriving flow over time, but also detailed modelling of changes in traffic dynamics. Demand is profiled with 15-min resolution. |
Space dynamics | Considers one homogeneous road section, or more precisely one incident site of unspecified size. The only spatial modelling is of queue extent. Off- and on-ramps are not modelled. | Divides the road section into cells in which the road characteristics may vary, enabling handling of exiting and entering traffic on off- and on-ramps. |
Traffic dynamics | Assumes that vehicle density changes instantly at each queue segment boundary (see Fig. 5).Neglects any delay due to reduced speed in traffic passing the incident or dispersing downstream. | Assumes that vehicle density is constant in each cell during each time step but can vary between cells and time steps. |
4 Calibration and implementation
Attribute | Network 1 | Network 2 |
---|---|---|
Test scenarios using network | S1: Collision | S2: Weather incident |
S3: LGV breakdown | S4: Obstruction | |
Number of lanes | 3 | 2 |
Length of simulation network | 100 km | 100 km |
Speed limit/free speed (v
max
,V
f
) | 80 km/h | 120 km/h |
Normal lane capacity (μ) | 1867 veh/h/lane | 2080 veh/h/lane |
Effective vehicle spacing (λ) | 7.43 m | 7.43 m |
Effective response time (τ) | 1.49 s | 1.45 s |
Time step in CTM (Δt) | 6 s | 6 s |
Length of cells in CTM (Δx) | 243 m | 243 m |
4.1 Capacity reduction in the vicinity of an incident
Lanes on freeway | No incident | Rubbernecking small | Rubbernecking large | Shoulder disable | Shoulder accident | 1 lane blocked | 2 lanes blocked | 3 lanes blocked |
---|---|---|---|---|---|---|---|---|
2 | 100% | 95% | 75% | 95% | 81% | 70% | 0% | N/A |
3 | 100% | 95% | 75% | 99% | 83% | 74% | 51% | 0% |
4.2 Modelling of VMS with variable speed limit
4.3 Handling of time-varying traffic profiles
4.4 Including the impact of novel technology
4.5 Running multiple scenario combinations in queue model
5 Verification of methods and comparison of results
-
Section 5.1 considers incidents and techniques that include one scene management period (no respite period) and constant traffic demand.
-
Section 5.2 considers scenarios and techniques that include one scene management period but for which the traffic demand varies over time.
-
Section 5.3 considers techniques that involve two periods of scene management or clearance separated by a respite period.
-
Section 5.4 compares results of applying the macroscopic simulation model and the queue model for all scenarios and techniques with one level of demand profile.
5.1 One scene management period with constant demand
Simulation tool | TIM technique | Total delay [veh-h] | Max queue extent [km] | Total duration [h] |
---|---|---|---|---|
CTM | 4.1 (close all lanes) | 4912 | 20 | 2.6 |
Queue model | 4.1 (close all lanes) | 3764 | 20 | 2.6 |
CTM | 4.2 (contraflow) | 2323 | 11 | 2.1 |
Queue model | 4.2 (contraflow) | 1909 | 11 | 2.1 |
CTM | 4.3 (close blocked lane) | 1070 | 6 | 2.0 |
Queue model | 4.3 (close blocked lane) | 1024 | 7 | 2.0 |
Simulation tool | TIM technique | Total delay [veh-h] | Max queue extent [km] | Total duration [h] |
---|---|---|---|---|
CTM | 4.1 (close all lanes) | 3764 | 20 | 2.6 |
Queue model | 4.1 (close all lanes) | 3764 | 20 | 2.6 |
CTM | 4.2 (contraflow) | 1974 | 11 | 2.1 |
Queue model | 4.2 (contraflow) | 1908 | 11 | 2.1 |
CTM | 4.3 (close blocked lane) | 999 | 6 | 2.0 |
Queue model | 4.3 (close blocked lane) | 1024 | 7 | 2.0 |
5.2 One scene management period with time-varying demand
Simulation tool | TIM technique | Total delay [veh-h] | Max queue extent [km] | Total duration [h] |
---|---|---|---|---|
CTM | 1.1 (close all lanes) | 16,612 | 58 | 5.0 |
Queue model | 1.1 (close all lanes) | 20,785 | 63 | 5.0 |
CTM | 1.2 (incident screen) | 23,878 | 63 | 5.6 |
Queue model | 1.2 (incident screen) | 28,503 | 69 | 5.7 |
CTM | 1.3 (close some lanes) | 11,018 | 50 | 4.5 |
Queue model | 1.3 (close some lanes) | 15,015 | 55 | 4.5 |
CTM | 1.4 (Tow in off-peak) | 4101 | 33 | 4.9 |
Queue model | 1.4 (Tow in off-peak) | 7243 | 45 | 5.0 |
Simulation tool | TIM technique | Total delay [veh-h] | Max queue extent [km] | Total duration [h] |
---|---|---|---|---|
CTM | 1.1 (close all lanes) | 16,612 | 58 | 5.0 |
Queue model | 1.1 (close all lanes) | 17,265 | 67 | 5.2 |
CTM | 1.2 (incident screen) | 23,878 | 63 | 5.6 |
Queue model | 1.2 (incident screen) | 23,941 | 73 | 5.7 |
CTM | 1.3 (close some lanes) | 11,018 | 50 | 4.5 |
Queue model | 1.3 (close some lanes) | 11,413 | 55 | 4.6 |
CTM | 1.4 (Tow in off-peak) | 4101 | 33 | 4.9 |
Queue model | 1.4 (Tow in off-peak) | 4066 | 33 | 5.0 |
5.3 Two scene management periods with time varying demand
5.4 Comparing the methods in all test cases
6 Conclusions and future research needs
Data set | 1 | 2 | 3 | 4 | 5 | 6 | Coverage (1–5) |
Number of months | 7 | 12 | 10 | 2 | 15 | 29 | 27 |
Number of incidents | 1018 | 942 | 5277 | 128 | 8322 | 103,359 | 15,687 |
Of which accidents | 621 | 238 | 1059 | 112 | 1355 | 21,374 | 3385 |
Attributes desired | Attributes recorded | Sets 1–5 only | |||||
Road identifier | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 100% |
Date | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 100% |
Start time | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 100% |
Response time | ⊠ | ⊠ | ✓ | ⊠ | ⊠ | ⊠ | 20% |
Cause | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 100% |
Duration | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 100% |
Severity | ⊠ | ✓ | ✓ | ✓ | ✓ | ⊠ | 60% |
Number of lanes available | ⊠ | ✓ | ⊠ | ⊠ | ⊠ | ⊠ | 20% |
Number of lanes closed | ✓ | ⊠ | ⊠ | ✓ | ⊠ | ⊠ | 40% |
Heavy vehicle involved | ✓ | ✓ | ✓ | ✓ | ⊠ | ⊠ | 80% |
Number of vehicles involved | ⊠ | ✓ | ⊠ | ✓ | ⊠ | ⊠ | 40% |
Number of persons/injuries | ⊠ | ✓ | ⊠ | ✓ | ⊠ | ⊠ | 40% |
Delay/Effect on traffic | ⊠ | ✓ | ⊠ | ✓ | ⊠ | ✓ | 40% |
Numerical delay estimate | ⊠ | ⊠ | ⊠ | ⊠ | ⊠ | ✓ | 0% |
Directions affected | ✓ | ✓ | ⊠ | ✓ | ⊠ | ⊠ | 60% |
Information about diversions | ⊠ | ✓ | ⊠ | ✓ | ⊠ | ⊠ | 40% |
Keywords or Text | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | 100% |