Почетна Преглед Помоћ
Пријавите се
ardo
/
Obsidian
1
0
Креирај огранак 0
Датотеке Дискусије 0 Захтеви за спајање 0 Вики
Дрво: a2f8402268
Гране Ознаке
Obsidian
/
Zotero
/
001_artiklid
/
zhanginterchangeconfigurationssafetycomparison2025.md

zhanginterchangeconfigurationssafetycomparison2025.md 32 KB
Историја Датотека

category: literaturenote citekey: zhanginterchangeconfigurationssafetycomparison2025 title: Interchange configurations safety comparison tool authors: "Zhang, Wei; Himes, Scott" year: 2025 date: 2025-00-00 2025 doi: 10.1016/j.aap.2025.108074 publication: "Accident Analysis & Prevention" url: "https://doi.org/10.1016/j.aap.2025.108074" zotero_key: DRAF77EG zotero_storage: HXVEDJF9 collections: transpordi_planeerimine folder: 001_artiklid firstAuthor: "Zhang, Wei"

status: converted

Contents lists available at ScienceDirect

Accident Analysis and Prevention

journal homepage: www.elsevier.com/locate/aap

Interchange configurations safety comparison tool☆

Wei Zhang a,* , Scott Himes b

b Vanasse Hangen Brustlin, Inc, Raleigh, NC, USA

ARTICLE INFO

Keywords: Service interchange Interchange configurations safety comparison tool ISCT

Interchange justification report IJR Safety performance prediction

ABSTRACT

The Interstate System is a fully access controlled surface transportation network connecting diverse geographical areas for the movement of people and goods. It is of national interest to preserve and enhance this system. Service interchanges are critical facilities linking State and Local roads to the Interstate System. When the need arises to insert a new interchange or modify an existing interchange, the request is presented in the Interchange Justification Report (IJR). Although many IJRs are submitted annually, the types of interchanges proposed are limited. Improvements in traffic operation due to proposed interchanges are usually well-described based on traffic analysis results. However, the safety performance of proposed interchanges is often not well substantiated due to a lack of safety modeling tools. Historic data indicates that high-traffic facilities like interchanges exhibit stable traffic crash patterns over time, meaning they should be predictable. Although many factors are known to induce traffic crashes, the safety community hasn't fully grasped how to properly quantify those factors. Decomposing an interchange into simple components and adding up the components' annual crash predictions generally doesn't yield results that match field data. The Federal Highway Administration (FHWA) developed the Interchange Safety Analysis Tool (ISAT) and supported the development of its enhanced version, ISATe. Both of these tools decompose interchanges into simple components for analysis. The FHWA recently completed a study treating the service interchange as a whole and using input data typically available in the planning phase to predict the safety performance of planned service interchanges. It includes the safety predictions of eight interchange configurations representing 78 percent of all interchanges considered in IJRs reviewed by FHWA. This paper presents the functionalities of this tool and explains how the tool may be used. This tool gives the IJR reviewers a consistent methodology for assessing the safety performance of proposed service interchange projects.

1. Introduction

Launched by the Federal-Aid Highway Act of 1956, the Interstate System of the United States is a full access controlled surface transportation system that connects diverse geographic areas to reduce travel distance and time and provides access to employment, shopping, entertainment, and many other destinations. It is of national interest to continually preserve and enhance this system by assuring it provides the highest level of service in terms of safety and mobility. Although primarily planned for civilian purposes to stimulate economic development and facilitate the movement of people and goods, its military value in evacuating cities and redeploying military equipment and personnel during nuclear attack was well understood by the then Congress, thus

the Interstate System was also known as the National System of Interstate and Defense Highways (The National Archives, 1956). Among all the roadway classifications, the Interstate System has the highest lane capacity.

Service interchanges are facilities that connect state and local roads to the Interstate or Expressway System. They serve as gateways for different modes of users to enter or exit the Interstate System. Given the functional nature of service interchanges, they are often hot spots of traffic congestion and crashes. Full control of access along the Interstate mainlines and ramps, as well as control of access on the crossroad at interchanges, is critical to providing the highest level of service. The FHWA helps maintain the nation's Interstate System and provides the requirements for the justification and documentation necessary for any

E-mail address: wei.zhang@dot.gov (W. Zhang).

☆ This article is part of a special issue entitled: 'RSS 2024' published in Accident Analysis and Prevention.

* Corresponding author at: Office of Safety and Operations Research and Development, Federal Highway Administration, 6300 Georgetown Pike, McLean, VA 22101, USA.

proposed changes in access to the Interstate System. The FHWA's decision to approve new or revised access points to the Interstate System—under Title 23, United States Code, Section 111—must be supported by information justifying and documenting that decision. To substantiate any proposed changes in access to the Interstate System, an analysis of potential safety performance impacts is required (GovInfo, 2011). Typically, substantiated information is presented in the IJRs, often written early in the project planning process, with details generally consistent with conceptual design and descriptions of potential safety performance impacts (FHWA, 2010; FHWA, 2017). Changes in access to the Interstate System may involve proposing a new interchange or modifications to an existing interchange. Historic crash data shows that high-traffic facilities, such as service interchanges, exhibit stable crash patterns over time, meaning their annual safety performance should be predictable. To this date, no simple slam-dunk predictive methodology has been developed that can accurately predict an interchange's safety performance in terms of total annual crash frequency and severity distribution.

With the various safety performance functions (SPF) for different types of (simple) roadway elements presented in the Highway Safety Manual (HSM)(AASHTO, 2014), it is tempting to think that decomposing an interchange into simple components and adding up the components' annual crash predictions would yield a reliable estimate of the safety performance of a service interchange. Unfortunately, this approach doesn't work well currently because the input parameters used in the SPFs are few (for intersections, the inputs include AADTs on major and minor roads, and may be a few site specific parameters, and the boundary conditions between roadway elements modeled in the HSM are very different from those that exist in an interchange). For the SPFs to predict crash frequency reliably, there is an implicit requirement that the input parameters used in the SPFs are the controlling factors of the safety outcome, and all other factors of the facility of interest are similar to the facility modeled in the HSM, including vehicle mix, driver behaviors, pavement conditions, and local annual weather pattern, etc.

1.1. Evolution of safety predictive tools for interchanges

In 2007, FHWA launched the Interchange Safety Analysis Tool (ISAT), a spreadsheet-based tool for performing safety assessments of interchanges and adjacent roadway segments and intersections. This original tool used a building block approach and focused on traffic volumes, the number of lanes on each interchange roadway segment, ramp configurations, and ramp terminal control. ISAT also provided limited opportunity for the consideration of design features and the differences in interchange configurations. ISAT was developed as an interim tool to meet immediate needs, providing crash estimates of only three typical interchange configurations (diamond, partial cloverleaf, and full cloverleaf) (Harwood, 2007).

In 2009, FHWA began collaborating with the Transportation Research Board to develop the Enhanced Interchange Safety Analysis Tool (ISATe), an enhanced prediction methodology and safety analysis tool for corridor and site-specific analysis. ISATe also provides information about the relationship between roadway geometric design features and safety. This version of the tool automates a safety prediction method consisting of extensive algorithms and equations. The lead researchers of ISAT and ISATe also worked as the lead researchers in the development of the HSM. Hence, many of the methodologies used in ISAT and ISATe are adopted in the HSM (Bonneson et al., 2012).

In 2014, ISATe became a supplemental component of the HSM, which provides detailed descriptions of the underlying data needs and predictive models implemented and serves as the primary means for evaluating safety performance for freeway segments, ramp segments, and ramp terminals. To predict crashes for these facilities, the ISATe and HSM predictive models require detailed sets of inputs—down to specific design characteristics—for features such as lane width and shoulder width as well as specific locations and design of barriers and curves for freeway facilities. Ramp connection locations and ramp terminal design details are also required. These details are not commonly known during the planning phase, leading agencies to struggle with determining how to estimate the required information—which requires an intensive data collection effort when details become available. Without these design details, the HSM predictive models cannot generate predictions of crash frequency and severity reliably. Some interchange types and components, such as diverging diamond interchanges or speed change lanes on crossroads, are not considered in the HSM predictive method. As an alternative to using ISATe, for interchange types not considered in the existing tool, analysts may use a crash modification factor (CMF) to assess the difference in expected safety performance from one interchange configuration to another. However, a single CMF fails to capture complex interactions taking place between the causation factors. While ISATe filled a critical need, analysts identified several difficulties in utilizing this tool.

2. Method

2.1. Development of interchange configurations safety Comparison tool (ICSCT)

In 2020, FHWA began developing planning-level models and tools for predicting crash frequency and severity for an existing or proposed interchange. The vision outlined for this tool is to allow analysts to compare the potential safety performance of freeway access and interchange design decisions earlier in the project development process without knowing the geometric details. To maximize the applicability of the new tool, FHWA began with a survey to all division offices to identify interchange configurations that make up at least 75 percent of those considered in the IJRs for inclusion in this study. Responses from 45

Table 1 Responses to questions 6 through 11 (Himes et al., 2023).

Question Response Number of
Responses*
6. How many IJRs does your State typically 0–5 35
submit per year? 6–10 8
11–15 2
16–20 0
More than 1
20
7. Of those IJRs, how many involve system 0–2 39
interchanges (interchanges providing access 3–5 6
from one freeway to another)? 6–8 1
More than 1
8
8. Approximately what percentage of service 0–25 29
interchange IJRs (interchanges that provide 26–50 11
access between a freeway and a nonfreeway 51–75 2
road network) are for rural interchanges (as More than 5
opposed to urban/suburban interchanges)? 75
9. Approximately how many service interchange 0–5 46
IJRs involve a new interchange? 6–10 1
11–15 0
More than 0
15
10. Approximately how many service interchange 0–5 40
IJRs involve reconfiguring an existing 6–10 6
interchange? 11–15 0
More than 1
15
11. Approximately how many service interchange 0–5 40
IJRs per year involve minor changes (such as 6–10 3
modifying ramp terminal access control, adding 11–15 2
lanes to an existing ramp, or relocating a ramp 16–20 1
terminal to a different roadway)? More than 0
20
No 1
response

*Categories with the highest number of responses are italicized.

states, Puerto Rico and Washington, DC were received. Table 1 shows the summary of responses to survey questions 6 through 11. Survey questions 1 to 5 collected contact information of survey responders and are not reported in Table 1.

The results indicate that most divisions receive fewer than five IJRs per year, with most of those focused on service interchange projects. However, some larger States receive more IJRs per year, with one agency reporting more than 20 IJRs annually. Question 8 responses indicate that the majority of IJRs focus on urban applications. The majority of respondents indicated reviewing 0 to 5 new interchange IJRs per year in question 9 and provided similar responses for reconfiguration of existing interchanges in question 10; the respondents provided similar responses for question 11 for minor interchange modifications.

For question 12, the research team provided respondents with 15 potential interchange configurations and asked how often each configuration was considered by each agency annually. Fig. 1 provides illustrations of the interchange configurations included in the survey.

After analyzing the survey results, the research team determined that the following eight interchange configurations represent 78 percent of the interchanges considered in the IJRs and 87 percent of the interchanges constructed (in order of most to least responses, and they become the interchange types included in this study):

  • Standard diamond interchange.
  • Diverging diamond interchange.
  • Single-point diamond interchange.
  • Compressed diamond interchange.
  • Tight diamond interchange.
  • Partial cloverleaf (Parclo) type A.
  • Parclo type B.
  • Parclo type AB.

2.2. Interchange area for input consideration

After determining the interchange types to be included in this study, the next step was to define the analysis area for inputs to the predictive models. Due to the variability of an interchange's influence area, a planning-level predictive method based on influence area was impractical. This level of analysis would have included the freeway sections outside of the interchange area— all the way to adjacent interchanges and crossroad sections— to at least the nearest intersections. Therefore, the predictive method focused on the interchange area. After discussions between the research team and the Technical Advisory Panel, it was decided to define an interchange area shown in Fig. 2, as the boundary for the predictive models, which consists of the freeway mainline, crossroad, and ramp terminals. The freeway mainline includes 457 m (about 1,500 ft) upstream and downstream of the entrance and exit ramp gores and 30 m (about 100 ft) beyond the curb return on the crossroad. The distance on the freeway mainline is consistent with planning-level analysis methods, such as those included in FHWA's Safety Analyst and ISAT for identifying interchange versus base freeway segments. This distance provides consistency with ISATe's research, which shows a decreasing influence from ramp entrances and exits, with no influence beyond 805 m (2,641 ft). The research team included all crashes observed in the interchange area in the scope for this tool; however, the tool does not predict crash frequency by location within the interchange area, only for the interchange area as a whole.

The method underlying this tool separately predicts fatal, injury, and property damage-only crash frequencies. When combined, it provides a prediction for the total crash frequency for each interchange configuration considered. The tool includes inputs to evaluate the relationship between geometric and operational features on the predicted severity level of injury crashes. Analysts can use the tool to evaluate the predicted crash frequency for varying interchange configurations,

Fig. 1. Interchange configurations considered in the survey (Himes et al, 2022).

Fig. 2. Definition of interchange area for the predictive models (Himes et al., 2023).

considering a standard diamond interchange or compressed diamond interchange as the baseline.

2.3. Data collection for development of predictive models

Based on inputs from survey responders, the research team compiled a list of 263 interchanges from five states (Arizona, Missouri, North Carolina, Ohio, and Utah) covering all interchange types selected, and went through a laborious process to extract the geometric measurements of each interchange using on-line mapping and GIS tools, and then gathered traffic and crash data from the respective states. Table 2 shows the range of interchange types, traffic and safety data from the data collection sites.

The predictive models developed identify that the relative safety performance among interchange types differs as a function of the freeway volume per lane, crossroad volume per lane, or interaction of freeway volume to ramp volume changes. In addition to traffic volumes, this tool includes crash frequency adjustment factors (like the HSM predictive method) for planning-level geometric and operational considerations, including:

  • The number of through lanes on the freeway and crossroad.
  • The interchange location (within an urban, suburban, or rural area).
  • Closely spaced adjacent interchanges on the freeway mainline.

Table 2 Summary data for geometric, operational, and crash data (Himes et al., 2023).

Int. Config. Freeway AADT Range Crossroad AADT Range Entrance Ramp AADT Exit Ramp AADT KABC Crashes PDO Crashes
Diamond 5,000–210,000 350–40,500 100–33,400 125–24,500 2,139 7,157
CD 23,100–236,000 11,000–52,900 6,800–25,500 4,250–22,500 1,246 4,278
TDI 17,000–207,300 3,200–55,000 4,000–36,500 4,500–36,700 4,552 13,659
DDI 29,000–191,000 2,000–47,000 2,000–38,500 2,000–45,000 4,037 10,883
SPDI 21,000–261,000 3,700–64,000 3,100–70,000 3,200–75,000 11,964 38,102
Parclo A2 6,400–115,300 1,500–30,615 650–9,400 1,300–21,800 673 1,283
Parclo A4 46,181–135,000 12,000–68,000 10,200–34,400 9,300–39,600 1,032 3,663
Parclo B2 7,298–123,000 150–32,000 35–14,800 35–12,400 744 1,789
Parclo B4 23,900–144,000 1,200–67,500 4,900–32,200 4,300–31,000 555 1,855
Parclo AB2 5,500–300,000 200–51,500 200–29,200 200–24,600 3,144 9,841
Parclo AB4 22,000–132,300 9,000–57,000 5,600–27,600 5,500–27,200 423 1,253

Int. Config. = interchange configuration.

  • The presence of one or more managed lanes on the freeway mainline.
  • Skew between the freeway mainline and crossroad.
  • The number of left-turn lanes present on the crossroad.
  • Consistency in traffic volumes among interchange ramps.
  • This tool also includes adjustment factors for planning-level geometric and operational characteristics that will impact injury severity, given that a crash has occurred. Considerations impacting crash severity include:
  • Traffic volume on the freeway mainline or crossroad.
  • Closely spaced adjacent interchanges on the freeway mainline.
  • The presence of an adjacent intersection in close proximity to the ramp terminal on the crossroad.
  • The posted speed limit on the freeway mainline or crossroad.
  • The number of through lanes on the freeway mainline or crossroad.
  • The number of pedestrian crossings conflicting with right turns at ramp terminals.

The above inputs—both for crash frequency and severity—provide analysts the flexibility to understand the impacts of geometric design decisions on interchange-level safety without needing detailed design information for each of the freeway mainline, ramps, and ramp

terminals. Further, these planning-level models incorporate the interactions among interchange elements by considering features together rather than independently.

3. Result

To facilitate using this new tool, the research team implemented the predictive models into a spreadsheet tool [9], organizing it in five sheets (or pages), namely: Instructions, PDO, KABC, SDF, and Charts. The Instructions page provides high level explanations of how to use this tool, what cells are for user inputs, and what cells are for results display. The PDO page is the main data input page with input data values shared with prediction models for KABC crash severities. The KABC sheet displays linked input data, and prediction results for fatal and injury crashes. The SDF page displays results of severity distribution function. The Chart page displays the confidence levels of crash predictions for different interchange configurations. Fig. 3 shows the user interface of this new tool, it allows users to enter data for any or all applicable interchange configurations for simultaneous analysis. Users can enter the exact data for each alternative or enter specific features as needed. The tool predicts property damage only, fatal and injury, total crash frequency, and a

Fig. 3. User interface and results display of the new Interchange Safety Comparison Tool (FHWA, 2023).

95 percent confidence interval for each interchange configuration entered. Additionally, it provides a graphic representation of the outputs for visual analysis. The companion User Guide [10] gives users details on input data requirements and examples of data elements required for applying the predictive model. Analysts can use the tool to compare the predicted safety performance for alternative interchange configurations, focusing on the planning-level inputs required to use the model. Analysts can evaluate the impacts of interchange spacing, proximity to adjacent intersections, number of through lanes on the freeway or crossroad, presence of managed lanes, posted speed limit, and number of right turns conflicting with pedestrian crossing movements. The results of the analyses may be used along with other considerations—ones that are typically considered in a particular region but not explicitly considered in the tool—to evaluate alternatives early in the planning and conceptual design phases as well as provide supporting documentation for decisions made.

4. Discussion

The development of ICSCT bridged a gap, attempting to strike a balance between the need for assessing the safety implications of designlevel details with those details that are generally known during project concepts and preliminary engineering. This tool focused on considering the interactions between individual project elements, rather than using a building-block approach, although the safety effects of some interchange components were difficult to isolate (such as ramp terminal control). This new tool was developed from data collected from the service interchanges most commonly considered in IJRs and cannot be applied to unique interchange configurations or system interchanges. Future efforts may build on the foundations of this tool to incorporate more locations and more interchange configurations considered by agencies as viable alternatives during project planning.

5. Conclusions

This project aimed to develop a planning-level safety prediction model to assess the safety performance of interchange configurations to support reviewing IJRs. Based on survey responses from 47 states or equivalent agencies, eight interchange types were selected that represent 78 percent of the interchanges considered in the IJRs, and 87 percent of the interchanges constructed. The predictive models were implemented into an easy-to-use spreadsheet tool, the Interchange Configurations Safety Comparison Tool (ICSCT) with instructions, input fields, prediction outputs displays, and distributions of severity functions, and confidences of predictions organized in five well-defined sheets. In its current form, the spreadsheet tool doesn't identify if inputs are valid or within the range of data used to develop the predictive models; it generates predictions for all configurations simultaneously. The user can provide the same inputs for all configurations or vary input parameters for each configuration when conducting a simultaneous test. Practitioners should exercise caution when making decisions based purely on safety prediction models. The estimated safety performances should be considered along with other factors, such as cost, operational performance, and environmental constraints. Professional judgment should be exercised if interchange features are atypical or outside the bounds of the data used to develop the models. Below are the key conclusions:

a. The predictive method found no safety performance difference between a diamond and CD interchange. These two configurations combined can be assumed as the base condition. The predicted crash frequency (for both KABC and PDO crashes) for interchange configurations relative to the base condition depends on the freeway volume, crossroad volume, combined ramp volumes, number of freeway lanes, and number of crossroad lanes.

  • b. The predictive method adjusts for area type, interchange skew, the presence of a nearby interchange, the presence of managed lanes, variability in ramp volumes, and the number of left-turn lanes on the crossroad.
  • c. For interchange crash severity, the base condition continued to be a diamond or CD interchange configuration. Crash severity was associated with interchange configuration; however, the project team found a similar severity distribution for SPDI, DDI, and TDI configurations.
  • d. The SDFs further identified crash severity proportion to be associated with freeway AADT, crossroad AADT, freeway posted speed limit, crossroad posted speed limit, the presence of a nearby adjacent interchange, the presence of a nearby adjacent intersection on the crossroad, the number of pedestrian crossings conflicting with rightturning vehicles, number of freeway through lanes, and the number of crossroad through lanes.

6. Disclosure

We certify that this work is original, it was presented at the 9th Road Safety and Simulation Conference, Oct. 28–31, 2024(RSS2024), Lexington, KY, and was recommended by the review committee for publication on the special issue of Accident Analysis and Prevention entitled "Advances in Road Safety and Simulation".

A Funding Source Declaration

Funding for the research was provided by the Office of Safety and Operations Research and Development, Federal Highway Administration.

CRediT authorship contribution statement

Wei Zhang: Conceptualization, Formal analysis, Methodology, Project administration, Writing – original draft. Scott Himes: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This study was funded by the Office of Safety and Operations Research and Development, Federal Highway Administration under Contract No DTFH6116D00040. We would like to acknowledge the contribution of research assistance: Vikash Gayah, Jeff Gooch, and Ginny O'Connor, all members of the research team for helping collecting and analysing data, and preparing graphics.

Data availability

Data will be made available on request.

References

Aashto, 2014. Highway Safety Manual Supplement. American Association of State Highway and Transportation Officials, Washington, DC.

The National Archives. National Interstate and Defense Highways Act (1956). https:// www.archives.gov/milestone-documents/national-interstate-and-defense-highwaysact.

Bonneson, J., S. Geedipally, M. Pratt, and D. Lord. 2012. Safety Prediction Methodology and Analysis Tool for Freeways and Interchanges. NCHRP Web-Only Document 306. Washington, DC: The National Academies Press.

Fhwa, 2010. Interstate System Access Informational Guide. Federal Highway Administration, Washington, DC.

Fhwa "Policy on Access to the Interstate System" (web 2017 page).