CHAPTER 4

Case Examples

This chapter presents case examples for the use of operational traffic simulation models by six DOTs—those of Colorado, Indiana, South Carolina, Texas, Virginia, and Washington (Figure 31). In consultation with the topic panel, the following criteria were considered as a basis for choosing the DOTs for the case examples:

• Diversity with respect to geography, level of experience, extent of use, types of applications, multimodal considerations, and implementation challenges
• State representation on the synthesis topic panel (some preference given)
• Willingness to participate in a case example as indicated in the survey

Table 23 lists the DOTs selected for the case examples and the basis for their selection, along with the annual number of projects that use operational traffic simulation models (based on survey response).

The case examples were developed through phone interviews with personnel from the six selected DOTs. Typical case example interview questions are provided in Appendix C. Some of the topics for operational traffic simulation models during the interviews included the following:

• Applications and experience
• Modeling processes
• Staffing and training
• Development and use of resources (e.g., guidelines) to support operational traffic simulation models
• Future direction
• Challenges

The case examples are described in the following sections of this chapter.

Colorado DOT

Overview of CDOT’s Use of Operational Traffic Simulation Modeling

CDOT uses operational traffic simulation modeling for over 100 projects per year. It most frequently uses operational traffic simulation modeling in the environmental and planning stage, typically when developing environmental impact statements. CDOT also uses operational traffic simulation modeling with congestion on some smaller projects. The decision on whether to use operational traffic simulation modeling is driven by project needs and by the consultant.

CDOT is highly decentralized, comprising five regions. Staff members in these regions coordinate the DOT’s efforts for operational traffic simulation modeling for each project and may also perform reviews of models developed by consultants. CDOT’s headquarters is also available to

perform reviews of models developed by consultants and to serve as a resource on operational traffic simulation modeling for the regions, but it may not be involved in every operational traffic simulation project in the state. The consultants who perform the modeling are typically selected on a project basis, not specifically for operational traffic simulation modeling. In some rare cases, peer consultant reviews are performed.

CDOT’s Practices for Operational Traffic Simulation Modeling

CDOT updated its guidance on traffic analysis in 2023 (CDOT 2023). This updated guidance suggests the use of the 2019 TAT guidelines on traffic analysis (Wunderlich et al. 2019); however, it also notes that the use of cluster analysis should be determined on a project-specific basis depending on data availability and project complexity. CDOT’s process for microsimulation analysis is shown in Figure 32. Model scoping is performed on a project-by-project basis, and the models that are developed are project-specific. Model performance is typically assessed through travel time, speed data, and other MOEs. Animation is sometimes used to present results at public hearings and to CDOT staff members.

Training on operational traffic simulation modeling is offered on an ad hoc basis and is typically delivered by software vendors. Software updates are usually installed soon after they are received from the vendor.

Table 23. Overview of case examples.

* As provided in the survey response

Other CDOT Considerations for Operational Traffic Simulation Modeling

CDOT is interested in learning more about how other state DOTs approach operational traffic simulation modeling programmatically. It is working toward improving consistency in modeling practices between regions and toward including more direction on modeling implementation in its existing guidance. CDOT also sponsored the development of an implementation plan for regional DTA in Colorado (Slavin and Morgan 2022). The DTA approach would be based on regional mesoscopic simulation and hybrid mesoscopic-microscopic simulation and could be used to assess operational improvements, project alternatives, management strategies (e.g., dynamic tolling), and innovative technologies. The DTA approach could also leverage new data sources from smartphone apps and CVs. A timeline for implementation of the DTA approach is shown in Figure 33.

Project Highlight: I-270 in Denver, Colorado

Microsimulation modeling is being used by CDOT to provide alternative analyses for an environmental impact study pertaining to a 7-mile section of I-270 in Denver, Colorado (see Figure 34). Alternatives being considered for the corridor include ramp modifications and the

use of express lanes (versus general purpose lanes). Microsimulation was chosen as an operational assessment tool due to the complex geometric and operational characteristics of the corridor, including interactions between express lanes and general purpose lanes; ramp metering with truck bypass lanes; a proposed partial cloverleaf intersection with turbo T intersections at the ramp terminals; afternoon sun glare that influences drivers’ behavior; and the presence of bottlenecks at various locations and times of the day—caused primarily by settlement, deteriorating pavement (because sections of the highway were built on a landfill), and bridge deck condition.

The microsimulation modeling being performed for the project is based on CDOT’s 2023 traffic analysis guidelines and on FHWA 2019 traffic analysis guidelines (CDOT 2023; Wunderlich et al. 2019). Some rescoping of the project was required as the project was initially scoped prior to the release of CDOT’s updated traffic analysis guidelines. Data availability has been a significant challenge encountered in using the TAT 2019 guidelines. Data sources include one continuous traffic counter on the corridor, vendor traffic data, data from a weather station on a nearby corridor, and incident data. In addition, queuing data were collected using dashboard video from floating car runs and 15-minute drone videos. Ten hours of the day (four hours in the AM, six hours in the PM) are being modeled. Figure 35 shows the modeled microsimulation network in the area of Vasquez Boulevard.

As of spring 2024, the analysis of the “no action” alternative is nearly complete, and the analysis of other alternatives is in progress. A technical memorandum summarizing the results of the cluster analysis was prepared and submitted to CDOT (Ackermann et al. 2023). MOEs reported include average speed, average delay, average number of vehicles processed, average number of queued trips, vehicle miles traveled, and vehicle hours traveled (Ackermann and Buck 2023).

Indiana DOT

Overview of Indiana DOT’s Use of Operational Traffic Simulation Modeling

The Indiana DOT (INDOT) has been using operational traffic simulation models since 2014. INDOT uses operational traffic simulation modeling for 10–15 projects annually that involve freeway simulation activities; 26%–50% of its operational traffic simulation models are developed by consultants. INDOT most frequently uses operational traffic simulation modeling for freeways, especially for complex service interchanges and interchanges with high arterial flow.

Indiana DOT’s Practices for Operational Traffic Simulation Modeling

INDOT primarily relies on resources from FHWA and other states (e.g., Oregon DOT 2023 and Wisconsin DOT 2019) for operational traffic simulation models. The FHWA 2019 guidelines (Wunderlich et al. 2019) were followed for a study on the I-80/I-94 Borman Expressway; the corridor has been frequently studied and has diverse datasets available. Other projects have used the FHWA 2004 guidelines (Dowling et al. 2004) due to limitations in data availability. The use of simulation on a specific project is identified through an initial project scoping meeting with the consultant that includes discussion of tools, study area, data, and MOEs. Representatives from INDOT, FHWA, and local agencies participate in this meeting.

Consultants are selected through a request for proposal (RFP) process; there are a limited number of consultants that perform simulation modeling for INDOT. In-house staff for simulation modeling work in the INDOT Traffic Mobility Office. Regarding software, INDOT has a license for Vissim for in-house staff but allows consultants to use other simulation software packages. If another software package is used, INDOT allows for additional review time and conducts in-depth review meetings with the consultant. These meetings allow the consultant to present details of the modeling process to INDOT.

Most of INDOT’s operational traffic simulation models are project-specific. For count data, traffic counts are performed for interstates on a continual basis. Model performance is assessed through flow at specific points, speed, bottlenecks, and visual audit of queuing. Animation results are sometimes used for public coordination, especially for innovative treatments.

Regarding training, INDOT makes frequent use of online training. In-person training is sometimes offered by the software vendor.

Other Indiana DOT Considerations for Operational Traffic Simulation Modeling

INDOT finds that data availability is the primary challenge to its use of operational traffic simulation models, especially finding high-quality data and obtaining data for large developments. For the future, INDOT is further formalizing its simulation processes and plans to investigate increased use of simulation for multimodal applications (e.g., bicycle, pedestrian, transit, and rideshare). INDOT is interested in learning more about the successes and struggles of other states in using operational traffic simulation models.

Project Highlight: Clear Path 465

Microsimulation (Vissim) was used for evaluating alternatives and for maintenance of traffic analysis for the Clear Path 465 project, which involved modifications to the I-465/I-69 interchange

Project Highlight: I-80/I-94 Borman Expressway

Microsimulation modeling was used to evaluate various TSMO strategies (e.g., dynamic shoulder lanes, variable speed limits, ramp metering, quicker recovery time for incidents) to improve operations and safety on the I-80/I-94 Borman Expressway in northwestern Indiana (Parsons 2021b, 2022). The project is believed to be one of the first projects nationwide to use the FHWA 2019 guidelines (Wunderlich et al. 2019). As shown in Figure 38, the study area encompassed multiple interchanges and extended into Illinois. Aimsun was selected by the consultant as the most suitable microsimulation tool for this specific application. Data sources included the National Performance Management Research Data Set for speeds, the Northwestern Indiana Regional Planning Commission Travel Demand Model, and StreetLight origin–destination data. A model screenshot is shown in Figure 39.

Cluster analysis was performed to identify representative days for morning peak, afternoon peak, incidents, weather events, Friday afternoon, and summer Sunday afternoon. A wide range of available datasets were used for the cluster analysis, including weather data, vehicle incident data, corridor travel time data, highway peak period volumes, and bottleneck data. The calibration required a significant amount of effort, and there were some gaps in the data. Historic traffic count data from different years and time periods were aggregated and balanced for the corridor.

Once the representative days were identified, the scenarios were modeled in Aimsun, and travel times and bottleneck speeds in each direction were extracted for each 15-minute period. An example plot of travel time and the associated envelopes is provided in Figure 40. Based on the results of the analysis, several packages of strategies were identified for further analysis, and design aspects of various TSMO strategies were underway in 2024.

South Carolina DOT

Overview of South Carolina DOT’s Use of Operational Traffic Simulation Modeling

The South Carolina DOT (SCDOT) started using operational traffic simulation models in the mid-1990s. These models were introduced to enable greater input into the state’s projects and to perform more detailed traffic analysis for intersection layout, signal operations, and capacity

improvements. The majority of SCDOT’s projects are analyzed using Synchro and SimTraffic. If a project involves a roundabout, SCDOT uses SIDRA; if a project is really complex, it will use TransModeler. Freeway projects are typically done in TransModeler but are also occasionally done in Vissim.

SCDOT uses operational traffic simulation models for: (1) evaluating alternatives for intersection layout, lane configuration, storage lengths, and signal operations; (2) analyzing interstate widening and interchange modification; (3) performing traffic analysis for Interstate Access Reports (IARs); and (4) conducting maintenance of traffic analysis for work zone traffic control scenarios. The DOT does not incorporate bicycles, pedestrians, transit, rideshare, or street parking into operational traffic simulation models very often.

SCDOT determines whether operational traffic simulation models should be used for a specific project during that project’s scoping meeting. This determination is based on the type and complexity of the project, the availability of data, and the need for alternative analysis. SCDOT typically uses simulation modeling for projects that involve capacity improvements, operational enhancements, lane configuration changes, interchange revisions or additions, or work zone traffic control scenarios.

SCDOT uses operational traffic simulation models for approximately 11–25 projects a year (approximately 35% of all projects). Of those projects, the DOT typically outsources 51%–75% to consultants. SCDOT obtains the data used for operational traffic simulation models from various sources, such as (1) count data from field observations or automatic traffic recorder systems, (2) INRIX data for speeds and travel times, (3) queue lengths estimated by the count representatives, (4) origin–destination data from Bluetooth devices or StreetLight, (5) projection data for forecasting from the travel demand model, and (6) historical data for comparison and validation.

SCDOT does not have state-specific model design criteria or guidelines for model calibration. It primarily uses the 2004 release of the TAT (Dowling et al. 2004); however, when data are available to leverage protocols from the 2019 version (Wunderlich et al. 2019), the DOT plans to switch to that version. SCDOT mainly focuses on LOS, delay, density, queuing, and speeds as MOEs for reporting model outputs, but these metrics are not used as quantitative measures of calibration or validation. Rather, the DOT relies on engineering judgment and visual cues when reviewing model outputs and calibration in order to validate findings.

SCDOT’s traffic design review group (in its central office) does the majority of in-house modeling and project review. District traffic staff members also use simulation models for signal operations and maintenance projects independently. SCDOT provides training on operational traffic simulation software to its internal staff when new versions of the software become available. These trainings are typically conducted by the software vendor.

SCDOT uses videos of operational traffic simulation models to convey design and performance of alternatives to various audiences, including at public information meetings and in workshops. Model visualizations have also been used to illustrate work zone traffic control scenarios, such as to show the effects of a directional split or a bypass lane on the interstate.

A few factors that hinder SCDOT’s use of operational traffic simulation models include: (1) staffing limitations—both in terms of the number of people and the level of expertise; (2) the lack of state-specific guidelines or criteria for model development, calibration, and review; (3) the difficulty of obtaining and validating data from different sources—especially during the COVID-19 pandemic; (4) the challenge of maintaining and updating large network-wide models that can be used for multiple projects; and (5) the diversity of software tools and versions that are used for different types of projects and analyses.

Project Highlight: Feasibility Report for US 29 (West Shockley Ferry Road) and SC 28 Bus (South Main Street) in Anderson

This project was identified as an issue due to the asphalt width and number of lanes—both encouraging speeding—that makes turning movements dangerous in this area. Due to the amount of growth in this area, right-of-way impacts were carefully considered and needed to be minimized. Additional concerns were queue lengths, lane configurations, and signal upgrades and spacing, as well as pavement markings.

Operational traffic simulation modeling was used to model the existing conditions and then to evaluate the effectiveness of five identified alternative solutions. The network was created with

experience with simulation. TxDOT staff provide consultant oversight and review for simulation modeling. In addition, TxDOT has incorporated peer consultant review for larger projects, with peer review used for three projects between 2022 and 2024. Training is usually given by the software vendor.

Other Texas DOT Considerations for Operational Traffic Simulation Modeling

TxDOT finds that data availability and cost are the primary challenges to its use of operational traffic simulation models. The agency strives to manage the costs of operational traffic simulation models, as it finds that simulation analysis costs 5 to 20 times more than HCM analysis. Looking to the future, TxDOT is working toward possible use of simulation for safety analysis, increased use of simulation modeling for multimodal applications, and continuing the transition to the 2019 TAT guidelines. TxDOT is interested in learning more about the real-world experiences of other states in using operational traffic simulation models.

Project Highlight: Capital Express North (I-35)

Microsimulation (Vissim) was used to evaluate alternatives for the Capital Express North project on I-35 from US 290E to SH 45N in Austin, Texas (HDR 2022). A map of the study area is shown in Figure 42. The 11-mile corridor experiences heavy congestion and delays during

most of the day. Proposed improvements include, but are not limited to, construction of one managed lane in each direction, addition of auxiliary lanes, bridge reconstruction, construction of a diverging diamond interchange at Wells Branch Parkway, ramp modifications, and improvements for pedestrians and cyclists on frontage roads and at crossings.

The analysis was performed based on the 2019 TAT (Wunderlich et al. 2019) using AM and PM peak-hour periods that included two hours of seeding time, three hours of time for evaluation, and one hour of cool-down time. A cluster analysis was conducted using various data sources, including seven-day traffic volume counts, mainline speeds provided by INRIX, TxDOT traffic incident reports, weather data, 72-hour ramp counts, and eight-hour peak period counts for turning movements. Three clusters were analyzed for the AM and PM peak periods (see example cluster plot for AM peak in Figure 43). Based on the cluster analysis, Thursday and Tuesday were chosen as the representative days for the AM and PM peak periods, respectively. A plot of the variation envelope for AM peak travel times is shown in Figure 44, and a heat map for the AM peak is provided in Figure 45.

LOS for freeways, merge/diverge, and weaving segments for intersections were the primary MOEs used for the analysis. Example results for freeway, merge/diverge, and weaving segment LOS are shown in Figure 46.

Virginia DOT

Overview of Virginia DOT’s Use of Operational Traffic Simulation Modeling

VDOT uses operational traffic simulation modeling for 51–100 projects per year (approximately 75% of projects). VDOT first started using operational traffic simulation models more than 20 years ago in an effort to do the kind of analysis needed to obtain accurate answers that would point to the best potential solutions and enable VDOT to make informed decisions. VDOT uses traffic simulation to more accurately model complex geometries and oversaturated conditions, and for other considerations that simpler tools do not handle appropriately.

To reduce the ambiguity around when to use different traffic analysis software tools, VDOT developed its Traffic Operations and Safety Analysis Manual (TOSAM) (VDOT 2020a). The purpose of this governance document was to provide consistent guidance and prioritization for alternatives across all districts and to address various issues related to operations, safety, and access management. As part of the TOSAM, VDOT developed a software selection tool, which is a macro-based Excel tool that helps users choose the most appropriate analysis tool(s) for their project, based on the project characteristics and the capabilities of different software (see Figure 47).

The majority of projects that leverage operational traffic simulation models are performed by consultants. Many simulation projects are completed by consultants pre-approved through one of VDOT’s on-call contracts. RFPs that include traffic simulation requirements specify the software that will be used, and consultants are expected to show their expertise using that specific software platform. To manage the consistency in traffic analytics across the commonwealth by different consultants, VDOT invests substantial resources into the development of documents such as the TOSAM (VDOT 2020a) and the complementary Vissim User Guide (VDOT 2020b), as well as into training programs.

VDOT has developed and facilitated training sessions on the TOSAM for both staff and consultants. The training covers software selection, scoping, model calibration, and performance metrics for simulation models. VDOT typically conducts new rounds of training when releasing new versions of the governance and guidance documents. For example, VDOT is working on an update to the TOSAM; once it is released, trainings will be offered to familiarize staff and consultants with the document and the updates made. These trainings are done in-house or with consultant support.

VDOT has partnered with vendors to do some training on specific software packages; however, the agency has frequently invested in more thorough training that details how to use those software packages within the context of VDOT projects. For instance, VDOT conducted in-person, multiday training for its staff on Vissim, the microsimulation software that is used for most simulation projects across the commonwealth. During the COVID-19 pandemic, this training program was converted to an e-Learning platform that enables users to take the training on demand. Although in-house staff do not often build and run simulation analyses, they are responsible for reviewing and validating the results. Therefore, these trainings were developed with an emphasis on the knowledge and skills required for review of models submitted by consultants.

Although VDOT has traditionally reviewed most of the operational traffic simulation models that are developed by consultants, the agency has started using peer reviews for large projects or for times when resource constraints limit its ability to do so in-house. These resource constraints can include a lack of work hours available or a lack of sufficient technical experience in the specified software. Before any new software is used, VDOT internally tests and verifies various aspects

of the software. New software programs that have same or similar functionalities as software that is in current use are limited in order to balance the time and resources needed to train staff on a new software’s performance outputs as well as to maintain consistency in analytics across the state. Specialized software programs are used if needed, with help from consultants to perform reviews.

The development of simulation models is mostly done by consultants, who are selected through a robust process that includes RFPs, proposals, résumés, presentations, and interviews. The central office provides resources to districts (e.g., TOSAM, Vissim User Guide, trainings); however, the districts have typically scoped and managed the projects located within their jurisdictions. The central office often gets involved in large or highly complex projects, reviews models for high-profile projects, provides subject matter expertise requested by the districts, and answers questions from the Virginia General Assembly.

A few changes that VDOT is considering in its approach to operational traffic simulation modeling include: (1) using more mesoscopic modeling for TSMO alternatives and regional multimodal mobility programs, (2) incorporating more pedestrian and bicycle considerations in microsimulation modeling (if reliable data and guidance are available), and (3) exploring the use of simulation modeling for safety analyses. VDOT anticipates that the next version of the TOSAM will incorporate simulation guidance for mesoscopic models in addition to that for microscopic models. VDOT indicated that national guidance related to best practices and reliable data for incorporating nonmotorized road users into operational traffic simulation models would be valuable.

A few factors that hinder VDOT’s use of operational traffic simulation models include: (1) funding and time constraints that limit the scope, data collection, and calibration of the models; (2) lack of guidance and reliable data on future conditions (such as demand, gap acceptance, and changes in driver behavior due to land use changes); (3) challenges of data conflation, data validity, and calibration criteria for existing and future conditions; (4) difficulty of archiving and reusing models due to software version changes and project duration; (5) need to demonstrate the return on investment for using novel software packages that are more resource consuming; (6) newer and more data-intense calibration methods; and (7) need to leverage new data sources. VDOT indicated that national guidance demonstrating the impact of new calibration and data collection recommendations—such as comparing the TAT 2004 and 2019 microsimulation guidance—would be beneficial for demonstrating to its leadership the ways in which additional investment will result in greater accuracy and can better inform decisions.

VDOT was a part of a pooled fund study with FHWA that developed the TSSM. The VDOT representative on the TSSM task force that reviewed the document said that it represents much-needed guidance for simulation modeling; a significant amount of public funding has been invested toward developing the manual. VDOT hopes that this resource will become publicly available in the near future.

Project Highlight: I-495 NEXT

VDOT, in cooperation with FHWA and Fairfax County, evaluated improvement alternatives for an extension of the express lanes along approximately 3 miles of I-495, also referred to as the Capital Beltway, from their current northern terminus in the vicinity of the Old Dominion Drive overpass to the George Washington Memorial Parkway in the McLean area of Fairfax County. The project location is shown in the vicinity map in Figure 48.

Vissim (version 9.0) was used for a comprehensive network traffic analysis for the freeways, interchanges, and adjacent intersections within the traffic operations analysis area limits. Vissim

is able to account for system-wide operations, including upstream and downstream conditions at any roadway segment, as it stochastically simulates traffic operations for individual vehicles on freeway segments and provides traffic operational data—including vehicle delay, density, and travel speeds—on freeway networks. Vissim reports average density as vehicles/mile/lane, and density analysis results are depicted in similar levels of congestion to the HCM density-based LOS thresholds.

Surface street intersection operations were evaluated through a combination of Synchro 10 (to develop preliminary optimization for phasing and signal timing) and Vissim (for microsimulation and analysis). The expanded arterial network beyond intersections that are immediately adjacent to freeway interchanges in the corridor was evaluated solely through Synchro. The Vissim model was calibrated to reflect existing real-world conditions according to VDOT requirements in the TOSAM (VDOT 2015).

Traffic operations analysis consisted of an evaluation of existing conditions (in 2018), No Build conditions (in 2025 and 2045), and Build conditions (in 2025 and 2045). Typical sections from the alternatives are shown in Figure 49.

Existing conditions demonstrated that travel demand exceeds available roadway capacity; therefore, the I-495 corridor has experienced increasing congestion, and these peak periods have an impact on parallel routes. Specific challenges within the study area include heavy volumes at the interchange of I-495 and Route 267, cut-through travel behavior on local parallel arterials, and weaving between high-occupancy toll lanes, I-495 express lanes, and general purpose lanes. The preferred Build alternative analyzed for 2025 and 2045 demonstrated improvement, and in 2045 was estimated to increase total persons moved by 6%–33% in the AM peak period and by 29%–35% in the PM peak period. Figure 50 provides a heat map comparison of average speeds between 2045 No Build and Build conditions for the AM peak period along the I-495 general purpose lanes.

Washington State DOT

Overview of Washington State DOT’s Use of Operational Traffic Simulation Modeling

The Washington State Department of Transportation (WSDOT) has been using operational traffic simulation models for approximately 20 years, after cities and counties had first started to employ them in Washington. WSDOT deploys operational traffic simulation modeling for 25–50 projects per year. WSDOT most frequently uses operational traffic simulation modeling in the planning stage, typically for more complex projects such as interstate access revisions, corridor studies, or for heavily congested areas. DTA is sometimes incorporated into larger projects. WSDOT’s policy requires the use of the least complicated and least data-intensive software available for a specific project (WSDOT 2023). WSDOT usually employs other deterministic tools (e.g., SIDRA) for roundabouts and has developed protocols to guide their use on state highways.

Washington state law requires WSDOT to incorporate complete street features into projects meeting certain conditions; this is considered as a baseline need on most projects (Washington State Legislature 2022). Simulations include nonautomobile modes (as appropriate) to estimate selected measures of effectiveness (MOEs), although nonsimulation methods are most often used to evaluate the effect of alternatives on nonvehicle modes.

Most of the simulation models in Washington are developed by consultants. Of those consultant-developed simulation models, the majority are developed through local agencies that hire their own consultants to perform the simulation modeling; WSDOT provides oversight

throughout the project. WSDOT’s regional traffic or planning offices often serve as the department lead for simulation work. However, on National Highway System (NHS) routes, WSDOT headquarters’ Transportation Operations division is included on the official review team and occasionally may serve as department lead (WSDOT 2014). An excerpt from WSDOT’s review checklist for microsimulation analysis is shown in Figure 51.

Washington State DOT’s Practices for Operational Traffic Simulation Modeling

To help ensure consistency in simulation modeling practices, WSDOT developed its own Vissim protocol in 2014 (WSDOT 2014). The protocol covers various topics, such as project scoping, data collection, model development and validation, and reporting. Multiple deliverables, such as a project methods and assumptions document, an analysis methods and assumptions document, a confidence and calibration report, model files, a summary of results, and other documentation, are required. The models developed are nearly always project-specific. In calibrating simulation models, WSDOT starts with what can be readily measured (e.g., speed, volumes, priority rules). Most data are obtained from permanent traffic recorders and short counts. WSDOT and its consultants also coordinate with local metropolitan planning organizations (MPOs) and regional transportation planning organizations to develop future year volumes. Animation is sometimes used to present project results and to help get project buy-in.

For training, self-guided resources from the software vendor are often used. Occasionally, live training is offered. WSDOT typically uses the most recent software versions from the vendor. Because WSDOT requires that the same version of Vissim is used throughout a project, it will typically delay use of the newest version of the program to allow time for service pack releases.

Other Washington State DOT Considerations for Operational Traffic Simulation Modeling

WSDOT finds various aspects of operational traffic simulation modeling to be challenging, especially needs for high-quality data, training, and transition planning for changes in staffing. For the future, WSDOT is working toward (possibly) adding another software platform, looking more closely at cluster analysis, and potentially increasing the use of MRM. WSDOT is interested in learning more about the practices of other state DOTs, especially regarding calibration tools, ways to streamline the modeling process, review practices, and data aggregation.

Project Highlight: Mukilteo Ferry Terminal

The Mukilteo ferry connects the cities of Mukilteo and Clinton. It is a 15-minute ride, with approximately 34 daily crossings (WSDOT and CH2MHill 2011). Microsimulation (Vissim) was used to help develop a business case for the addition of an elevated structure for pedestrians at the Mukilteo ferry terminal to improve operations and safety when loading pedestrians and vehicles. Two pedestrian loading options were modeled: (1) at-grade loading, in which pedestrians use the vehicle ramp to board the ferry before vehicles, and (2) elevated structure loading, which would allow passengers to board simultaneously with vehicles. The modeling required custom software programming (using Vissim’s vehicle actuated programming language) to incorporate schedules, storage constraints, queue processing, and loading rules. The modeling incorporated conditional routing by vehicle class, signal heads that control loading time, street operations for the entry to the holding area, and the loading of pedestrians and cyclists. Animation (see Figure 52) was used to help visualize the proposed design. A new Mukilteo ferry terminal opened in 2020, and the pedestrian walkway began operation in 2021 (see Figure 53) (WSDOT 2024).

Summary of Case Examples by Topic

The six DOTs described in this chapter have diverse experiences with respect to operational traffic simulation models. Some of the key findings from the DOT interviews, based on the experiences of these six DOTs, are summarized in this section.

Key Case Example Findings for Applications for Operational Traffic Simulation Modeling

• Operational traffic simulation modeling is used for a wide range of applications, including, but not limited to (1) assisting in the environmental and planning stages of projects (e.g., environmental impact statements); (2) assessing freeways (especially for complex interchanges and interchanges with high arterial flow); (3) evaluating alternatives for intersection layout, lane configuration, storage lengths, and signal operations; (4) performing traffic analysis for studies that are used for IARs; (5) conducting maintenance of traffic analysis for work zone traffic control scenarios; (6) assessing the impacts of truck climbing lanes and roundabouts; (7) considering complex geometries; (8) simulating oversaturated conditions; and (9) other considerations that simpler tools do not handle appropriately.
• The need for the use of operational traffic simulation models is typically determined on a project-specific basis. In some cases, scoping meetings at the beginning of the project provide an opportunity to discuss the use of simulation as an appropriate tool.

Key Case Example Findings for Practices for Operational Traffic Simulation Modeling

• These DOTs use various resources for guidance on operational traffic simulation modeling, including their own DOT-specific guidance, guidance from other DOTs, and FHWA resources.
• There is general movement toward the use of 2019 FHWA guidelines (Wunderlich et al. 2019); however, data availability for the cluster analysis is viewed as a significant challenge to broader implementation.
• Most simulation models are developed by consultants, with review and oversight from DOT staff.
• DOTs use various software packages for operational traffic simulation modeling, and software updates are typically installed on a regular basis. Consultants often play an active role in

• selection of the software tool, and the software deemed most suitable for the specific application is chosen. In some cases, RFPs with traffic simulation requirements specify the software that will be used.
• DTA is sometimes used on larger projects.
• Initial scoping meetings are often conducted at the beginning of a project to discuss the project scope and modeling approach.
• Simulation models are typically project-specific, although in some cases prior models are incorporated.
• Data for simulation modeling are obtained from various sources, such as count stations; field observations; historical data (for comparison and validation); projection data (for forecasting from the travel demand model count data); subscriptions to vendor data (e.g., StreetLight for origin–destination data, INRIX and Wejo for speeds and travel times); local agencies; and consultants.
• Examples of methods and measures used to assess model performance include travel time, speeds, LOS, delay, density, queuing, bottlenecks, throughput visual cues, and engineering judgment.
• Examples of required deliverables include a project methods and assumptions document, analysis methods and assumptions document, confidence and calibration report, model files, summary of results, and other documentation.
• Training is typically provided online and in person from software vendors; in some cases, in-house training is given.
• Animation is sometimes employed for presenting project results and conveying design and performance of alternatives to various audiences, such as in public information meetings, workshops, or presentations.
• Peer reviews are sometimes performed for larger consultant projects, but their use is infrequent.
• Some local agencies conduct operational traffic simulation modeling, often using a consultant.

Key Case Example Findings for Challenges and Future Opportunities for Operational Traffic Simulation Modeling

• Challenges faced by these DOTs in the use of operational traffic simulation models include, but are not limited to, data availability; cost; staffing limitations; guidance and training needs; difficulty in obtaining and validating data from different sources—especially during the COVID-19 pandemic; diversity of software tools and versions that are used for different types of projects and analyses; archiving and reusing models due to software version changes and project duration; and demonstrating the return on investment for new approaches.
• Examples of ways these DOTs are working toward potentially enhancing their use of operational traffic simulation modeling in the future include encouraging consistency in modeling practices; providing more direction on modeling implementation; expanding the use of simulation for multimodal applications (e.g., bicycle, pedestrian, transit, rideshare); applying simulation for safety analysis; continuing the transition to the 2019 TAT guidelines (Wunderlich et al. 2019); using mesoscopic modeling for TSMO alternatives and regional multimodal mobility programs; adding other software platforms; and increasing the use of MRM.
• State DOTs are interested in learning about various aspects of other state DOTs’ experiences with operational traffic simulation models, such as how they approach operational traffic simulation modeling programmatically, their successes, and their struggles.

Summary of Case Examples by DOT

A summary of case example findings by state DOT is shown in Table 24.

Table 24. Summary of case example findings by DOT.