CHAPTER 3. FINDINGS AND APPLICATIONS

This chapter contains the key findings by each of the major efforts within the project.

LITERATURE REVIEW

The concept of speed and the many ways it can be expressed and measured, as well as the variety of factors that can influence it, have been the subject of much research in recent decades. An emphasis in NCHRP 15-76 was to review the literature related to these topics and summarize the findings most relevant to the concept of target speed and its influences.

Results from the review found that the use of the term “target speed” is increasing in the profession, but it is often either not defined, or it is described in the context of another speed concept. The 2010 ITE Recommended Practice (RP) titled Designing Walkable Urban Thoroughfares: A Context Sensitive Approach (ITE 2010) provided a definition for target speed that has been used in whole or in part by other references, including the 2018 AASHTO Green Book (AASHTO 2018) and the 2020 Florida DOT Design Manual (Florida DOT 2020b), but other variations of target speed definitions are also used within the profession. Based on the common uses of the term and its application within existing literature and policy, as well as the need to place the term in the context of the design process, the definition for target speed developed by the research team for this project is as follows: a speed selected for a roadway during the design process that guides planning, design, and operational decisions and results in a roadway with a driving environment that encourages drivers to operate at or below that selected speed. Target speed can be used in the design process in conjunction with more established speed concepts such as design speed, operating speed, and posted speed limit.

Speed is used in different ways within the broader project development process that includes planning and traffic engineering components as well as design. Similarly, multiple factors can influence a driver’s speed choice; however, the exact relationships may not be clear or may not been conclusively proven. Adding to the challenge of quantifying the relationship is the interaction between these factors and the overall visual scene for the driver. The contextual expectations of the road can communicate an appropriate speed to a driver, though the driver must be willing to accept that message in a contextual response. More details on the concepts and factors identified in the literature review can be found in Appendix A.

CURRENT PRACTICES

In addition to research that describes findings related to operating speed and factors that potentially influence it, a number of jurisdictions include target speed and related concepts in their guidance. The research team initially identified three states (Florida, Washington, and Oregon) whose departments of transportation (DOTs) had documented procedures or recommended practices for including target speed in the design process. Researchers also identified three cities (Austin, TX; Charlotte, NC; and Chicago, IL) with design guides or policies that included target speed. Each of these agencies that included target speed in their design processes also emphasized the use of a context classification to guide the selection of a target speed value. Each agency considered different factors in establishing their context classification criteria, but those factors typically included roadway cross-sectional elements (e.g.,

number and width of lanes, type and width of median, on-street parking), multimodal elements (e.g., bicycle lanes, bus lanes, sidewalk presence), and roadside or non-roadway elements (e.g., number of driveways, adjacent land use). The policies from state agencies also often included suggestions or recommendations for potential speed management treatments to consider for a given roadway to help achieve the selected target speed. Over the course of the project, researchers identified additional agencies that had either developed new policies that included target speed (e.g., Massachusetts DOT, Texas DOT) or had made updates to their existing policies that were reviewed earlier (e.g., Florida DOT). The details of current practices reviewed during this project can be found in Appendix B.

DATA FRAMEWORK AND SOURCES

Project 15-76 began with three categories of elements that can affect the choice of operating speed: roadway, roadside, and non-roadway. This provided a simple and straightforward method for broadly categorizing the variables of interest, but the variety and quantity of material to be considered in this project required that the information be thoughtfully organized so that it could best inform the direction of the data collection and analysis. Thus, the research team developed a conceptual data structure to use as a framework for structuring information gathered for review and analysis. This structure used factors related to volumes, census data, speeds, and location-related characteristics (e.g., roadway corridor and cross-section characteristics, roadside characteristics, non-roadway features, and traffic control devices). Appendix C provides the format and workflow for the data framework developed in Phase I as a basis for the Phase II database.

A key to this NCHRP project was the ability to connect operating speed with site-specific elements to identify relationships for target speed selection. Identifying those relationships required a large and comprehensive database that included many factors from multiple sources. In addition to the operating speed measure, this database needed to include roadway, roadside, surrounding area, land use, census, traffic control devices, and traffic volume characteristics for the roadway segments being studied. A database that included all the needed components did not exist and needed to be developed within the project. Therefore, this Phase I effort sought to identify available data sources that could provide large quantities of data of interest with the intent to merge these datasets to form the database that would be used in the analysis.

The features and capabilities of available big data sources indicated that speed data from many more sites could be obtained for analysis more efficiently than with traditional data collection techniques. With the availability of big data (vendor) sources for speed data, specifically the INRIX XD database, the project was able to consider a wider range of values for the factors of interest. Rather than being limited to a specific city or metropolitan area within a region to keep travel to a minimum, the research team was able to consider more sites in more locations to provide not only a geographic distribution but also a more robust sample of sites representing a range of values for select factors. The magnitude of the additional data provided a dataset that was sufficient for a robust statistical analysis to produce meaningful results. Appendix C describes the effort by the research team to identify sources for study sites, location data, speed and traffic data, census data, and other site characteristics.

ANALYTICAL APPROACH

The review results demonstrated that there is a wide variety of factors that potentially influence operating speed, and subsequently target speed, that could be included in a thorough analysis, and those factors have different methods available to model them and investigate their relationships to speed. To investigate the effects of those various factors on operating speeds and to explore potential analyses approaches, the research team considered multiple modeling and analysis strategies to identify the most appropriate modeling approach as well as the factors most suitable for investigation. Researchers used Analysis of Covariance (ANCOVA) models that included both categorical/discrete factors and continuous covariates as predictors to explore the relationships among factors based on a dataset consisting of speed variables and dozens of candidate predictor variables. In all, seven preliminary models were developed and evaluated for suitability, and a model was chosen that included 50 factors for data collection and analysis in Phase II. Appendix D provides the details of the various models and factors that the research team considered in this effort.

SUITABILITY OF INRIX XD SPEED DATA

Researchers needed a suitable source of speed data for each of the chosen study segments. Traditional methods of collecting speed data (such as spot speed data measured at the site) provide detailed speed readings for a limited number of vehicles at a limited number of sites, while speed data measured using probe vehicles (e.g., the INRIX XD™ database) provide binned speed readings for potentially thousands of vehicles at tens of thousands of sites. The research team investigated available sources of data and their relative features, with an emphasis on determining suitability of the INRIX XD™ database for use in the project. Results of the investigation (see Appendix E for additional details) indicated that the INRIX XD™ database was appropriate for use in the project and that the ffspd measure provided by INRIX, representing the 67^th-percentile speed of the available data for a given study segment, would provide a suitable speed measure for use in the analysis. Details of this investigation and its findings are provided in Appendix F and Appendix G.

DATA COLLECTION AND ANALYSIS

An objective of NCHRP Project 15-76 was to develop a comprehensive list of potential study locations and categorization of the analysis elements. Using the states and sites identified in Phase I as a starting point, the research team reviewed available sources of site data from state DOTs and online databases and ultimately identified several potential segments in California, Florida, Massachusetts, Oregon, Texas, Utah, Virginia, and Washington to use for data collection and analysis. The research team further reviewed the list of potential factors provided in Phase I based on direction from the panel at the Interim Meeting to determine what factors to emphasize in Phase II. The team decided to proceed with a similar list of 56 factors for analysis in Phase II, and they grouped the factors list based on similarities in data collection method.

The research team conducted data collection and analysis as part of Phase II. Data collection was organized into Rounds 1 through 6 and Rounds A through D, to develop a database of characteristics and speeds for the segments chosen for analysis. Each round included a review of segments to identify any segments that needed to be removed from consideration due

to a lack of data or a characteristic that necessitated its removal. This review in each round allowed the team to avoid data collection in future rounds for segments already identified to be removed. Ultimately, researchers included 6,327 segments from the aforementioned eight states in the final database for analysis.

The research team used ffspd as a dependent variable (corresponding to a measure of operating speed) to explore the relationship between operating speed and various roadway, roadside, and non-roadway elements. The team employed ANCOVA models that included both categorical/discrete factors and continuous covariates as independent variables to explore the relationships between ffspd and more than 50 independent variables obtained from the 6,327 segments. As researchers ran multiple models and reviewed the results, they eliminated and/or combined some variables that were minimally influential or that were correlated with other variables. After reviewing multiple combinations of factors in a variety of models, the research team decided to focus on variables within the following five efforts:

• Evaluation using posted speed limit variable, to provide insight into the influence of the number on the speed limit sign.
• Evaluation using variables associated with transportation professionals, which included variables that could be under the influence of a transportation professional.
• Evaluation using demographics variables, which included any available variable that could help to identify conditions that are associated with different operating speeds.
• Evaluations focused on center line and edge line markings, which considered the presence or absence of center line or edge line markings in the subset of the database containing two-lane roads with no median, an average lane width of no more than 13 ft, and a PSL of 25 to 35 mph.
• Evaluation focused on average lane width, which explored the relationship between lane width and operating speed for segments with average lane width between 8 and 13 ft.

Details of the data collection and analyses are found in Appendix F. The results of the analyses are summarized in Chapter 4 and described in detail in Appendix F and Appendix G.

PRACTITIONER’S GUIDE

The research team used the findings from the analyses and the review of literature and current practices to develop an implementable procedure for practitioners to incorporate target speed into the design process. This procedure is contained in the NCHRP Web-Only Document 460, Volume 2: Implementation Guide (Brewer et al. 2026).

TECHNOLOGY TRANSFER

Preparing a webinar presentation and developing a draft article suitable for publication in TR News were requirements of the research. Those items are provided as separate files supplemental to this research report.

DOCUMENTATION OF RESEARCH EFFORTS

The final effort within the project was to document the methodology and findings from the research, which is contained within this report.