An Overview of CUBE Avenue

Oliver Charlesworth — Fri, 28 Aug 2020 08:01:21 GMT

Current Revision posted to CUBE Wiki by Oliver Charlesworth on 8/28/2020 8:01:21 AM

CUBE AVENUE

CUBE Avenue is an additional module completely integrated within the CUBE Voyager modelling steps. Avenue uses standard CUBE Voyager commands and scripts, with high flexibility for the user, allowing tailoring the assignment for many purposes.

Image 1 - Module Structure

Avenue brings the Mesoscopic approach and Dynamic Traffic Assignment Equilibrium to CUBE, loading and tracking the movement of vehicles throughout the network.

As CUBE Avenue processes the various kinds of information used in traffic assignment, it moves between different levels of aggregation. On a disaggregate level, vehicle packets move, stop, and queue through upstream roads and intersections:

explicitly simulating traffic flow, computing interactions among vehicle units as they travel through the network; and,
explicitly simulating traffic control systems, modelling traffic signals, roundabouts, stop-controlled intersections, and ramp merges.

Then, Avenue estimates an aggregate measure of congested travel time based on vehicle density on road segments during each time increment. The total journey time then consists of:

Link travel time
Junction delay
Queue time

Image 2 - Isolate queues by length and time

CUBE Avenue computes the lowest-cost path for each vehicle unit, based on its departure time, considering the link travel time corresponding to the time segment of the estimated entering time of the packet to that link.

Through an iterative process, CUBE Avenue calculates optimal network conditions at equilibrium.

Typically, the model converges in several iterations, with an extremely fast process for smaller systems and feasible for large urban areas. Indeed, modelers have successfully applied CUBE Avenue on areas with around 3,000 zones and 25,000 links.

The combined use of both static and dynamic processes in the same run can be accomplished through a Hybrid Static and Dynamic Traffic Assignment, including both dynamic and static elements.

CUBE Avenue – Applications

Other than traditional types of analysis, considering aggregate outputs from the model, CUBE Avenue can provide two main additional layers for analysis:

modeling time explicitly - time-specific policies can be studied more realistically; and,
modeling the interaction among vehicles and propagation of queues - more detail in the traffic simulation analysis is provided.

Some of the potential applications are listed here below:

The technologies in CUBE Avenue provide techniques for the analysis of a variety of traffic management and traffic operations improvements including vehicle information systems and benefits from other ITS technologies.
CUBE Avenue may be used to study the impact of improvements to interchanges, the effects of adding or removing signals and other changes to traffic control strategies.
It is also an excellent tool for planning alternatives during roadway construction in terms of optimization of roadworks management.
Time-specific policies, such as variable road pricing, dynamic lane opening/closures, or other policies for alleviating peak period congestion can be more realistically analysed.
Quantify impacts of upstream traffic congestion, measure queuing at intersections and merge points in a network, and isolate secondary impacts from one intersection through another.
Simulate alternative infrastructure, operational, and policy changes to optimize emergency evacuation plan strategies as part of emergency management analysis.
Test strategies to improve arrival and departure from stadiums and other special event facilities, etc.

Image 3 - Hampton Roads Hurricane Evacuation Study (source: http://www.pilotonline.com)

CUBE Avenue – Input Data

Most urban areas can use existing data in their regional transportation model to implement CUBE Avenue. When collecting traffic data for DTA there are two additional components that needs to be considered:

the distribution over time of the demand and traffic conditions, consistent with the simulated time increments; and,
a higher level of detail required for the description of the roadway network.

Basic CUBE Avenue inputs include:

Roadway network in CUBE Voyager format
Roadway distance, capacity, lanes and vehicle storage area (vehicles per distance)
Peak period trip tables and time segmentation
Traffic intersections characteristics

CUBE Avenue – Outputs

The outputs from Avenue are both numerical and visualization aids, with the ability to create 2D animation for bandwidths and vehicle circulation. When combined with CUBE Base, CUBE Avenue can help to analyze and visualize future operating conditions under various demands and road management strategies.

Image 4 - Volume bandwidths and vehicles animation

In terms of numerical results, CUBE Avenue produces performance measures for every specified time segment and for the entire modelled period. Output performance indicators include:

Traffic volumes on a road link
Total traffic in queue
Link operating speed and travel time
Link occupancy/utilization
Intersection LOS and operating conditions

Moreover, data can be summarized with the charting and table functions in CUBE Reports.

Image 5 - Example of output report: queue length by time segment

Avenue provides detailed information on the movement of each packet through the network in a so-called “packet-log file”. By post-processing this output file in CUBE Voyager, many advanced analyses can be implemented:

Build origin-destination tables from log file
Extract average queue for specific packets
Select node/link analysis
Temporal disaggregation
Peak spreading
Other applications: ITS/VMS, parking, sub-area extraction, ME, etc.

Several visualization outputs and animations are also available:

Node/link posting
Multi-bandwidth display
Intersection data by time segment
Bandwidth animation
Packet-log animation
Scenarios comparisons

Image 6 – Intersection detail by time and vehicles animation

Tags: DTA, Mesoscopic, simulation, Avenue, Cube