,

Cube Avenue Case Study: Dynamic Traffic Assignment to solve traffic chokepoints

User: The Richmond Regional Transportation Planning Organization or PlanRVA

Location: Richmond, Virginia, United States

Challenge: To efficiently analyze major traffic chokepoints in the greater Richmond region and to identify impacts from implementing various mitigation strategies.

Solution: Use Citilabs Cube Avenue to prepare a meso-scale Dynamic Traffic Assignment (DTA) process with the help of big data, providing more realism than a static assignment and less complexity and faster run times than microsimulation.

The Richmond Regional Transportation Planning Organization, PlanRVA, is a collaboration of nine jurisdictions in and around Richmond, the capital of the Commonwealth of Virginia. PlanRVA is the Metropolitan Planning Organization for the Richmond region and is responsible for a variety of transportation and related planning activities.

Richmond is a medium-sized, historic urban area bisected by Interstate 95, the nation’s “Main Street.” Recent growth, especially in suburban areas, has led to similar traffic problems that affect many successful cities. As in most metropolitan areas, the congestion is focused on a handful of “chokepoints” that habitually confound drivers and are a daily source of local frustration.

 

CHALLENGE

In order to solve a problem efficiently, you must be able to analyze it in a realistic manner. The proper tool for the task at hand must be sensitive to the impacts of the strategies that are being considered. PlanRVA has used Cube Voyager to model the metropolitan area, including a static assignment step that is suitable for regional analysis, such as estimating vehicle-miles of travel and mobile source pollutant evaluation. Although static assignment is useful in the proper context, it is limited in its ability to represent many specific traffic congestion situations as well as the potential solutions to such problems.

The alternatives to static assignment include meso-scale assignment and true microsimulation. The latter is extremely data intensive, requires significant development resources, and is not practical for a large area. Meso-scale methods, often referred to as DTA, are more suitable for the size of Richmond’s study area but network and trip table development can still be a daunting task.

Map showing Richmond area traffic bottlenecks | Source: PlanRVA

 

APPROACH

Cube Avenue is the kind of tool that is best suited to a more thorough evaluation of detailed traffic phenomena. Avenue offers a meso-scale process of traffic assignment that tracks the progress of trips through the network over short time increments and accounts for the buildup of queues due to congestion or incidents. Users also have the ability to deal with the stream of traffic flow, in which a slowdown at one location affects the flow of traffic at downstream locations. DTA solutions such as Avenue are becoming the preferred tool for this kind of work because they are easier to use and run faster than traffic microsimulation programs.  PlanRVA’s familiarity with other Cube products enabled a simple, fast, and efficient path for implementing DTA with the Cube Avenue module by easily importing data from the regional Cube Voyager model.

Part of Cube Application Manager flowchart for model application | Source: PlanRVA

 

For each bottleneck area, a subarea network was created, with enhanced characteristics that are known to affect traffic flow at a detailed level, such as turn lanes. PlanRVA used big data based on cell phone Global Positioning System signals to provide “observed” traffic flows and origin/destination (O/D) data. This was input to Cube Analyst Drive to adjust the O/D data from the regional model to have greater fidelity within each of the project study areas and to slice the trip data into 15-minute time segments. The trips were assigned using Cube Avenue and numerous measures of effectiveness were derived, such as congested speeds and queue lengths. The process was calibrated by comparing the model volumes and speeds to link volumes and speeds derived from big data analysis. The model results validated well against the observed data.

 

Volume vs. count validation results for one area | Source: PlanRVA

 

Speed validation results for I-95 | Source: PlanRVA

 

RESULTS

The validated model was used to evaluate a variety of strategies, including both traffic operations and land use solutions, as well as the effects of lane closures. This analysis provided data such as volume, speed, delay, and queuing that PlanRVA needed in order to evaluate these strategies properly. PlanRVA gained important insights from this work, including:

 

  • Big data can be used effectively to develop the subarea demand, with careful O/D expansion methods.
  • DTA can replicate bottleneck conditions at key locations such as merges of major roadways and movements, short ramp segments, and heavy AM/PM loads, with suitable accuracy.
  • The DTA process provides the ability to analyze bottlenecks and evaluate mitigation measures. The use of available observed volume, O/D and speed data from big data sources is very efficient and minimized the need for expensive new data collection.
  • It can be challenging to match volume counts and congested speeds simultaneously.  Fortunately, Avenue provides tools to help the analyst with these tasks.

 

The results of this work demonstrated that Cube Avenue is an excellent tool for analyzing freeway bottlenecks and associated mitigating improvements. Combined with big data, it is a powerful platform for evaluating both traffic operations improvements and the effects of land development.

”Cube Avenue is a good handy platform for conducting mesoscopic Dynamic Traffic Assignment (DTA). Coding and scripting the DTA model has been simpler than I have imagined,” said Srin Varanasai, vice president of transportation systems planning for The Corradino Group. “It allowed us to spend more time on the much-needed calibration.”

Sulabh Arycal, AICP, transportation planning manager with PlanRVA added “Cube Avenue is a great add-on to Cube. A must have for all Cube users.”

 

For more information on this project, contact:

Sulabh Aryal, AICP
Transportation Planning Manager
PlanRVA
saryal@planrva.org

 

 

 

Srin Varanasi
Vice President, Transportation Systems Planning
The Corradino Group
svaranasi@corradino.com

 

 

 

For more information on Cube, contact:

Katie Brinson
Director of North American Sales
Citilabs, Inc.
kbrinson@citilabs.com