Availability and Use of Pedestrian Infrastructure Data to Support Active Transportation Planning (2020)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

37 Approach Following the survey’s conclusion, representatives from five states participated in case example interviews to provide additional detail on pedestrian infrastructure data collection, maintenance and use in their states. The interviews generally addressed the following themes: • Assessment of data collection – Are data currently collected? If so, how? How are data currently used, and what are the desired future uses of pedestrian infrastructure data? • Facility and pedestrian network definition – For states collecting pedestrian infrastructure data, how are they defined? What attributes are collected, and to what extent? • Data collection methods, management, sharing and responsible parties – Who collects and maintains pedestrian infrastructure data? This section also focuses on the mechanics of how data are stored and managed, concerns with data sharing, and methods for sharing data with the public. • Data consistency – How are data kept consistent, including quality control and how data review is handled? • Data maintenance and update strategy – Are plans in place to maintain pedestrian infra- structure data over time to keep them relevant and useful? • Program funding – Are plans are in place to fund data collection and maintenance over time? If so, what is the funding source? The five state transportation agencies selected to participate were Kentucky, Louisiana, New Hampshire, Utah and Washington. In addition to expressing participation interest via the survey instrument, these states represent diverse collection frameworks, different geographies and various development contexts. All five states identified current pedestrian infrastructure data collection and/or use. Each interview included a review of the state’s survey responses and provided opportuni- ties to expand on the original survey responses. Confirmation of the survey results occurred for two primary reasons. First, several case example interviews included a range of state staff who may or may not have been involved in providing survey responses. By asking partici- pants to confirm the survey response, case example participants had the opportunity to pro- vide clarity on the response indicated or identify information not originally known. Second, this provided the opportunity to explore specific answers or a combination of responses in greater depth. This chapter first presents brief summaries of each state interview, with key findings highlighted for each state within the context of the larger department goals and efforts related to pedestrian infrastructure data collection and maintenance. Following each case example, a summary of the interviews is provided based on the resulting themes across all five interviews. C H A P T E R 4 Case Examples

38 Availability and Use of Pedestrian Infrastructure Data to Support Active Transportation Planning Kentucky The Kentucky Transportation Cabinet’s pedestrian and bicycle program coordinator par- ticipated in the case example interview. Part of the planning department, this position is one of the only staff dedicated to pedestrian and bicycle planning for the state, with part-time support from other staff when needed. The program has worked to move the agency forward in better providing for consideration of bicycle and pedestrian needs, in part through improved data collection efforts. The agency indicated that an increasing number of projects seeks state fund- ing by including bicycle and pedestrian components. Many of the state’s pedestrian and bicycle projects are not included in the main transportation improvement plan. Nearly 90% of local planning agencies (LPA/local government) seeking state or federal funding include a pedes- trian or bicycle facility or treatment of some type. Overall, the agency understands that there is increasing demand for more and better facilities throughout the state. To support increasing demand for safe walking and bicycling facilities, the state also has updated its required forms for project approval to include greater consideration for bicycle and pedestrian accommodation. The new process now requires a narrative explanation for how bicycle and pedestrian travel is considered in project design, as opposed to a checkbox confirming that it was considered, as previously mandated. Assessment of Data Collection Over the last 5 years, pedestrian infrastructure data were collected for all state roadways and have expanded to include local roadways in some incorporated areas. Initial pedestrian infra- structure data collection was completed in various ways: using Google Earth and vans equipped with 360-degree cameras and, in some cases, physical verification via site visit. These collection tools enabled verification of existing sidewalks and bicycle facilities along state routes. In some cases, local inventories were collected in urban boundaries. Following this data collection, area development district (ADD) offices were contracted to confirm and perform a preliminary data quality check. At this time, the agency estimates the data to be approximately 80% complete for all roadways collected. The state’s transportation cabinet recently hired a consultant to provide a final data quality check and update to serve as the foundational pedestrian infrastructure database for the agency. The motivation for these efforts is due in part to cost and federal requirements. Not only is the state working toward finalizing an ADA transition plan that the collected data will supple- ment, there also is a benefit for local highway districts in knowing the location and quality of pedestrian facilities in advance when seeking/preparing project bids. FHWA and the U.S. Access Board require facilities to comply with current ADA standards when roadways are improved, reconstructed or newly constructed. In some cases, project bids did not account for the all- needed improvements to pedestrian facilities. In fact, they found that many projects had been underbid by nearly 20%. Currently, Kentucky Transportation Cabinet (KYTC) pedestrian and bicycle travel data are used at the project level. However, there is a desire to better incorporate bicycle and pedestrian facility data into project evaluation to assess connectivity improvements. The agency noted that with improved data, there also is potential to improve safety analyses further. The outcome is to have a better consideration in bicycle and pedestrian facility selection, so that more multimodal roadway corridors are created over time. Facility and Pedestrian Network Definition Main data collection efforts include an inventory and data for paved roadway shoulders, bike lanes, shared-use paths, marked crosswalks and sidewalks. Some trail information (off-roadway

Case Examples 39 right of way) has been collected but is limited to trails that clearly serve a transportation purpose. These typically are paved shared-use paths alongside roadways or at crossings. For trails that do not follow roadway rights of way, these data are more difficult to include because the roadway data are attached to roadway centerlines. Crossings and marked crosswalks also have been explored, because they are understood to be a vital part of nonmotorized infrastructure. However, Kentucky is considering the best way to classify and maintain these data. Following the FHWA definition of a crosswalk, the agency wants to track those that have a constructed pedestrian facility on both sides of the roadway. Relying on crosswalk markings as a way to gather these data, viable unmarked crossings are omitted, and marked crossings that do not meet the crosswalk definition may be falsely identified. Signal information is based on installation records and maintained only for state roadways outside of urban areas. Local agencies may maintain more detailed inventories, including the location of actuated pedestrian signals (APS), for major urban areas. Existing records for state signals have not been inventoried or integrated into spatial datasets. Data Consistency While the consultant review and quality check of data will address consistency among datasets to some degree, the agency also has developed a simple data scheme that can easily translate to a shapefile. This scheme is maintained as a spreadsheet and seeks to improve on previous collec- tion efforts that, for example, required a more subjective judgment of quality. Data Sharing and Program Funding The data scheme is shared via the KYTC walk/bike website and with local jurisdictions upon request. It also is available for download via the state’s data mart. The agency expressed no con- cerns with sharing the data. Funding for data collection efforts currently is drawn from a combination of sources. The pedestrian and bicycle program budget and general highway transportation funds each sup- port the effort and likely will be utilized for ongoing data maintenance. There is hope that other departments also will be able to integrate the data into their efforts. Data Maintenance and Update Strategy State contracts require ADD offices to collect and review/update these data. Further, every time the walk/bike program provides a consideration review regarding a roadway project, the existing data are referenced and reviewed. Agency GIS staff also frequently review existing data against the publicly available Strava Global Heatmap to determine if there are potentially missing facili- ties based on nonmotorized travel patterns. Louisiana The case example interview for Louisiana Department of Transportation was conducted with an engineer from the highway safety section. With a background in Complete Streets, she has assumed many of the department’s bicycle and pedestrian coordinator duties. Other staff share additional aspects of this role. However, given limited planning department staff with traditional planning background/education, engineers typically address many items that usually fall to a planner. Current data tracking efforts are related to federal requirements and a series of perfor- mance metrics that the department is required to report on each year. The annual report to the

40 Availability and Use of Pedestrian Infrastructure Data to Support Active Transportation Planning legislature covers Complete Streets and includes items such as miles of bike lanes traveled and number of people trained in Complete Streets. Assessment of Data Collection In addition to data items collected annually per HPMS requirements, the state has completed multiple data collection efforts over time. The most recent is through a consultant who was hired to complete a comprehensive data collection for all state routes and local roads. The consultant uses a roadway survey vehicle to capture roadway characteristics. Collected data inform project-level work and aid in safety analyses. In the future, however, staff would like to approach safety analyses more systematically. They also would like to use sidewalk data on local roadways to inform local master-planning and project-level efforts. There also are opportunities to use data in statewide Safe Routes to Public Places project selection and funding. In general, the state is working toward Complete Streets as a guiding principle where bicycle and pedestrian travel are considered in all projects and exploring the use of available data to help achieve this. Facility and Pedestrian Network Definition Data collected on local roads include, but are not limited to, pavement conditions, roadway type, sidewalk presence, posted speed, sidewalk width, number of lanes and lane width. The collection effort is managed through the planning department. Through the vendor-sourced data collection, MPOs were provided the opportunity to collect additional data. Coordination is done to varying degrees with MPO GIS departments, and local technical assistance is avail- able to support efforts such as sharing data schemes, navigating data-sharing questions and exploring how to integrate other data. Attributes collected primarily were guided by the elements specified in FHWA’s Model Inventory of Roadway Elements (MIRE). While the state collects information beyond these elements, the MIRE framework is intended to help guide agencies toward a more robust data management system that allows for performance-measure development and safety assessment. The ADA transition plan includes sidewalk locations, deficiencies, conditions and needs pri- oritization. Because the ADA data were collected before the recent comprehensive inventory, the ADA program and GIS group worked together to integrate data and resolve discrepancies. Data Maintenance and Update Strategy Several vendor-collected data items are updated yearly or biannually for interstates, National Highway System (NHS) routes, on-system routes and HPMS samples, and as projects are completed and inventories updated. The data maintenance plan in place for the agency speci- fies the three-year period, but includes additional annual updates. Items specifically noted for maintenance include roadway volume, roadway class and sidewalks. However, the frequency with which sidewalk data maintenance occurs is unknown. Physical sidewalk maintenance is not completed by the state, so updates rely primarily on projects or additions to the network. Data Sharing and Management The Louisiana Department of Transportation and Development (DOTD) has made all col- lected raw data available for download and use via an FTP site. All collected data items also are available via REST Services, with state-owned roadway information loaded into an enterprise

Case Examples 41 system. Staff are in the process of having the local data transitioned into the system, with completion expected by fall 2020. DOTD currently is involved in a pilot project with an MPO to determine best practices for sharing large amounts of data. The ADA transition plan is publicly available via the DOT website. There are some concerns regarding data sharing, specifically related to liability. Examples include failure to update deficiencies as projects are completed (ramps are always fixed, but sidewalks might not be). The liability question is more general in nature, however, and an explicit concern has not been defined. New Hampshire A traffic engineer from the New Hampshire Bureau of Traffic participated in the case example interview. This individual’s bureau role includes work on signals, signage and other project- oriented tasks. Projects have included a range of ADA-focused efforts. In addition to the traffic engineer interviewed, several others throughout the agency contributed to the survey responses. As such, it is important to note that there might be other ongoing efforts not fully reflected here. Assessment of Data Collection Data collection primarily is driven by ADA compliance efforts. The ADA office is associ- ated with the Bureau of Highway Design, and with staff focused on upgrading facilities, the effort has become more holistic and comprehensive. Data, however, are split across the agency; the ADA officer maintains one database that includes items such as facility width, location and placement of truncated domes. Planning and traffic also maintain databases that presently are not consolidated with the ADA database. Ongoing efforts aim to consolidate all of these sources. The data sources vary and include field review, imagery and permitting, or as-builts. The future desired use noted by the participant was for greater ease of data sharing between departments and teams. Understanding condition of facilities also would be useful to plan- ning efforts. For example, more complete and widely available data may provide insight into safety analyses or evaluation of connectivity and other grant criteria. Facility and Pedestrian Network Definition Data are collected at varying levels of detail. Records are stored for all ADA projects; however, these are not collected comprehensively for all roadways. Sidewalks are reflected in as-built drawings, which are linked to a GIS database. Project extents are shown on a map, but details regarding sidewalks primarily are available via project documents. Trail data are maintained by another agency. Crossings and signals are tracked in a traffic database, as all crossings are permitted. These data typically are accessible only by the traffic department and maintained as an Access database. Data are stored across various departments in a range of formats, with limited sharing among departments. Some data are spatially located and reference other documentation, while others are in nonspatial databases or records. The agency did mention interest in exploring new database infrastructure, in particular related to records presently maintained in Access.

42 Availability and Use of Pedestrian Infrastructure Data to Support Active Transportation Planning Utah The case example interview for the Utah Department of Transportation included three staff members from UDOT: the active transportation planning coordinator, Traffic and Safety Division’s data and asset manager, and a transportation engineer from the Traffic and Safety Division. UDOT is working toward a robust, coordinated system of data collection and analysis that is reflected through the responses shared in this interview. The move toward improved data is based on the idea that the department cannot manage what it does not know it has. Data allow the agency to understand its assets and can lead to more efficient use of constrained funding. In addition to infrastructure data collection efforts, the state uses data from entities such as Strava and bike/scooter share companies. As it looks to expand implementation of automated counters and explores other methods of understanding demand through its partnership with the university, the state is demonstrating the benefit of quality comprehensive data. Assessment of Data Collection and Data Consistency Utah data sources vary by department and use. In most cases, the GIS department sources, but does not maintain, data from various departments. For example, another department sources and maintains trail data, which the DOT only references. Roadway data currently are collected for state roadways as a one-time snapshot of the asset. Following the first collection in 2012, data are gathered every 2 years. Sidewalk data recently were digitized by agency staff in the course of 1 year. Based on aerial imagery, this dataset also provides point-in-time details that allow for visualization of network gaps. For active transportation facilities, the state recently developed a data schema that guides which information is collected for different facilities. Under this data schema, bicycle and pedes- trian master plan data are collected from the local level and rolled up to the state. While this is recognized as a likely challenge for most DOTs, a good relationship with the state’s MPOs facilitates this coordination. There is an effort to develop a data repository for MPOs to provide a central storage location for all active state transportation data. Facility and Pedestrian Network Definition For sidewalks, data are stored on the sidewalk centerline. It represents presence/absence infor- mation and includes an estimate for both width and quality. Roadway characteristics captured through the vendor include shoulders and crosswalks that traverse the state roadway. Crosswalks that run parallel to the state roadway are not captured. Manual collection of curb ramps currently is ongoing, with the goal to move toward a paper- less collection system. The information collected for ramps ideally will include ramp type, whether it meets standards and, if not, why. Attributes collected for ramp data consider the ADA transition plan for the state and are aimed at establishing a method for tracking conditions that better inform project needs. Signal data are maintained by the signal team and available via a web mapping application. The lighting team also maintains existing lighting data. Data Management and Sharing Data storage methods depend on the agency responsible for data maintenance. Some data are stored in GIS format, while project-level work often is completed in CAD and stored as

Case Examples 43 PDFs via ProjectWise. The state is moving toward a paperless system that focuses on model- based design and construction documents. The state does not have any concerns about sharing infrastructure data. Data typically are shared with the public through KMZ files, though not stored in this format for internal use. For collision data, however, the state does have concerns about personal identifying information that may be associated with the data. Data Maintenance, Update Strategy and Program Funding Roadway data collected through the vendor are updated every 2 years and funded with state dollars by the Traffic and Safety, Maintenance and Right of Way divisions. No plan currently is in place for sidewalk data, as they were collected opportunistically based on staff time. There is concern that this effort will not be replicated or updated in the future. Overall, the relationship among departments is crucial for ongoing data maintenance. For data items such as Strava or other third-party data sources, interview participants indi- cated that centralizing purchasing with the agency would help minimize duplicate efforts and increase efficiency in data use. Washington The case example interview with Washington State Department of Transportation (WSDOT) included members from the Active Transportation Division; Transportation Data, GIS & Modeling Office; Asset Management team; and the Information Technology Division. Eight WSDOT staff were present for the call. Participants currently interact with pedestrian data in various ways, including data ownership and storage, network and program planning, and ADA planning. At the time of the call, WSDOT also was working on updates to the state- wide Active Transportation Plan, which has included consideration for current sources of pedestrian data. Through this interview, staff noted that discussion of pedestrian and roadway infrastructure data sources was informative and highlighted the need for increased communication among departments to determine data needs, available data and coordination opportunities. Facility and Pedestrian Network Definition WSDOT staff noted a wide variety of data types and sources related to pedestrian infrastruc- ture data. Responses first centered on counts and use patterns and considered infrastructure elements such as sidewalks, crosswalks and signals second. For participating staff, questions regarding infrastructure and network needs were tightly linked. The connection with counts was reflected in the agency’s focus on understanding where current infrastructure is not meet- ing user needs and how the agency can assess the pedestrian network in terms of comfort and access for all. Primary data collection drivers are safety and a desire to manage assets better. Assessment of Data Collection Current data collection primarily is focused on count and project-level infrastructure data. However, the agency expressed interest in using LiDAR data to collect comprehensive pedes- trian network information. A preliminary pilot was underway at the time of the interview to test the viability for statewide data collection. Interview participants expressed interest in understanding the way vendor data, such as Strava or Waze, can inform understanding of

44 Availability and Use of Pedestrian Infrastructure Data to Support Active Transportation Planning travel patterns using the pedestrian transportation network. The agency also is aware of efforts such as Open Sidewalk Project but expressed concern regarding crowd-sourced data specifi- cally related to accuracy and liability. Currently, data are used primarily for project-level work and ranking grant applications. However, the agency would like to use data to better inform project selection. Considering the agency’s strong focus on a multimodal transportation system, the addition of pedestrian infra- structure data can help with selection of project type, location and timing. With additional data, the agency can better assess issues such as connectivity and how the pedestrian transportation network can help different user segments. WSDOT believes this understanding of pedestrian infrastructure data also would include overlaying information with other data outputs. For instance, how can the pedestrian network be better assessed using equity analysis outputs? Sharing data with the public also is a stated DOT goal. The agency also noted interest in further exploration of cell phone data use; sustainable methods for data collection, maintenance and management; and how to better integrate local and regional efforts across the state. Data Collection Methods and Data Consistency WSDOT data vary across the state and by type. Historically, data have been collected through several methods and for numerous purposes. The sources and attributes were described as: • Pavement Management System – Roadway surface. • Transportation Information Planning and Support System (TRIPS) – Primary roadway database; provides attributes regarding roadway widths and mileposts, among others. • CAD System – CAD-based documents, such as as-builts for individual projects, are stored in a statewide system that can provide project-level details. This is the primary source of new data. Integration among these systems has not occurred on a standard basis, and access to these sources among departments is varied. Consolidation of various datasets – including integra- tion of project-level data from the CAD system – is a current state goal. However, there have been software-related challenges. Software updates over time limit compatibility between data formats, thus making data merging more challenging. Further, compatibility among different software vendors affects data consolidation. With regard to ADA data specifically, data collection has occurred at various times with varying levels of detail. The most recent effort ceased in 2012. The data acquired during this time were considered to be inconsistent and did not undergo a quality control process. Interest in data acquisition was revived in 2016, and current efforts require all WSDOT contracts to specify ADA self-assessment data. However, a recent review found that not all contracts collect these data in full. Multiple sources of data and incomplete collection efforts have affected data consistency. For example, mileposts have changed over time due to the addition of new roadways and realignment of existing roadways. This affects linear referencing systems, and project data integration requires an understanding of the period of both the project and the corresponding milepost system. At this time, there is no process in place for consolidating and unifying data products for use across departments. Additionally, there are no standard definitions for asset types that would facilitate data consolidation. However, there is an ongoing effort to develop these definitions.

Case Examples 45 Data Maintenance and Update Strategy The agency is under an executive order to maintain data quality, for which a governance policy is in place. Previously, a governing body worked to centralize agency data and eliminate data silos. Each time this body convenes, progress is made, but challenges arise when working across departments. Work is underway to reconvene this group, and it is hoped to be in place within the next year. One goal of the group is to develop metadata for agency datasets. Data Sharing A range of data products is available through WSDOT’s open data portal, including usage data. CAD data, however, are available only by request and not widely accessible by the public. The agency’s concerns about data sharing relate primarily to privacy and liability. Privacy concerns are present with items such as crash data. However, following national standards, WSDOT has been successful in maintaining privacy. Liability also is a concern with regard to crash and safety data, and the state soon will apply privacy practices similar to general data protection regulation (GDPR) standards commonly used by the European Union. As WSDOT explores more data options from other sources, including vendors, contrac- tual concerns also were noted. In general, the agency aims to apply correct considerations to the correct data source, given the wide variety of sources available through WSDOT. Assessment of Data Collection Although all five states indicated their agency collected data, the definition of pedestrian infra- structure data varied widely. At least one participant from each state identified facilities such as sidewalks and crosswalks as pedestrian infrastructure. However, when asked about infrastruc- ture data, other participants also identified data related to pedestrian counts and collisions. The discrepancy in pedestrian infrastructure data definitions is due in part to the current use of these data within the state agency and/or what facilitated collection of data originally. Examples include: • Washington DOT (WSDOT) currently has a robust counts program that it is actively seek- ing to expand. State-owned pedestrian infrastructure data (e.g., sidewalks, crosswalks and similar pedestrian infrastructure) are limited, while knowledge of the counts data is more prevalent. Further, programs seeking to expand placement of automatic counters highlight the consideration of counters as infrastructure. • Three states explicitly discussed ADA mandates and transition plans as the catalyst for data collection. While the resulting data elements – including format and associated attributes of the data – varied in response to this mandate, meeting the requirements of ADA was a significant factor in collecting pedestrian infrastructure data. • Other reasons for collecting pedestrian infrastructure data included project-level planning, tracking progress on implementing Complete Streets and improving efficiency and cost of infrastructure maintenance programs. For example, Kentucky responded that demand for quality bicycling and pedestrian facilities, along with frequent overbudget projects (includ- ing ADA upgrades), necessitates better data to inform project selection, project funding and network needs. • Pedestrian infrastructure data may be defined and/or collected differently based on depart- ment, use and collection method. For example, project as-builts might provide the basis for project planning, while safety divisions may store information about roadway and pedestrian infrastructure in a tabular format to evaluate crash patterns.

46 Availability and Use of Pedestrian Infrastructure Data to Support Active Transportation Planning Data collection practices typically were not informed by desired future use of pedestrian infrastructure data. In fact, the states interviewed most often identified current use of data as the desired future use. However, the most frequently noted difference was for pedestrian data to be more comprehensive and easier to navigate to better inform activities such as project- level and ADA planning. Specifically, states desired for pedestrian data to be more consistent with roadway infrastructure data in terms of detail, completeness and maintenance built on the principles of asset management and TPM. Washington hopes to better integrate pedestrian data into project selection techniques that allow the agency to assess a multimodal system, including ways to gauge connectivity within the pedestrian network. This may affect the state’s approach to data collection and maintenance over time. Facility and Pedestrian Network Definition The definition and use of pedestrian infrastructure data are influenced by the method of data collection. Collection refers specifically to how the information for a particular facility type or pedestrian-related dataset is gathered and recorded, whether through aerial imagery, field observation or another strategy. Further, states reported different methods of data collection depending on the type of data and, in part, the intended use. The spatial extent of data collected varied due to factors such as funding, staff time and knowl- edge, and intended data use. As explored in the previous section, data use often is determined in part by state or federal guidance, requirements or mandates. These uses may expand as FHWA policy encourages more comprehensive TPM. Project As-Builts Several states indicated that data are collected based on individual roadway projects, in most cases recorded primarily in the project’s as-built documents. These documents most frequently are CAD or PDF files, and the data typically are not georeferenced. These files are referenced as needed, often to inform project-level planning efforts. Since this method of collection is project- based, states identified that this information was collected for some or all projects. Aerial Imagery In some cases, data are collected opportunistically based on identified data needs, available staff time or other resources. For example, the Utah Department of Transportation recently used aerial imagery to capture the locations, relative quality and estimated widths of sidewalks along state roadways. The use of aerial imagery and an established GIS database has enabled the collection of sidewalk data as time and skillsets allow. Third-Party Vendors Third-party vendors were an identified source of current or potential future data collection by several states. For third-party vendor data, the relationship among attributes collected, extent of data and use of data clearly are demonstrated. Examples include: • Louisiana – A third-party vendor, FURGO, was hired to complete a comprehensive data col- lection on all state routes and local roads to meet federal guidelines for data collection. While the intent of this data collection initially related to ADA Transition Plan efforts, limitations

Case Examples 47 of the data collection technique did not allow for the required level of detail. Because of this constraint, more detailed ADA data were collected by the Office of Compliance, which differs from the Highway Safety Office, where the vendor-collected data are housed. • Utah – Roadway data, including roadway shoulders and crossings of the state roadway, are collected via LiDAR data by a company called Mandli. However, items such as crossings adjacent to the state roadway, curb ramp locations and whether a curb ramp meets standards are not viable data outputs from the LiDAR system. For this reason, the state has explored options such as manual data collection of curb ramps to inform ADA planning and use of aerial imagery for sidewalks and adjacent crossings. Other data sources noted by interview participants included Strava, StreetLight, Sugar Access, bike and scooter share companies, and OpenSidewalks. In general, there is interest from the interviewed states to explore the opportunities, benefits and challenges that datasets from these sources might provide. Utah, for example, uses Strava data to help understand current use pat- terns. While the state recognizes the limitations of this dataset, it has found the information it provides to be beneficial as one piece of the discussion. The primary questions asked by partici- pants regarding these data sources included: • What are the limitations or biases of the data? • How accurate is information related to infrastructure? • Specifically, for questions related to ADA compliance, are there concerns regarding the authority of the data source? District, MPO and Local Data Consolidation Highway district staff, MPOs and jurisdictions often are the most familiar with the assets in their area. For this reason, these staff and agencies may be able to provide detailed information for many locations across a state. Examples provided for using this data source include gathering datasets from local jurisdictions and MPOs and providing quality checks and updates on data- sets acquired through other methods. These examples are consistent with current recommended data management practices. Concerns noted about this approach include consistency in datasets – related to the types of attributes collected, possible field values and geospatial qualities of the data, such as centerline location or line segmentation – as well as varying levels of staff knowledge and experience related to GIS or database management. Kentucky is working to establish a comprehensive bicycle and pedestrian database for all state roadways and roadways in incorporated areas. While data were initially collected via van-mounted cameras, staff in the ADD offices provided reviews to check the quality of the sidewalk data extracted from this imagery. As these staff work directly with counties and other jurisdictions, they have been able to assist with identification of these facilities. Current efforts coordinating with ADD office staff are focused on trails and shared-use-pathway data collection, specifically where they do not follow the existing right of way. For several states, collaboration with MPOs, jurisdictions, other state agencies and/or advo- cacy groups provided the primary source of available trails data. However, due to trails often not following existing rights of way, interview participants gave less emphasis to trails data collection processes. Trails were noted as a vital aspect of transportation networks and tourism efforts, although the departments participating in the interviews typically did not collect, maintain or use existing datasets.

48 Availability and Use of Pedestrian Infrastructure Data to Support Active Transportation Planning Data Consistency Establishing consistency among datasets, both in terms of spatial extent and attributes, is a vital component to establishing comprehensive, reliable datasets that serve to inform the data’s intended use. This is particularly concerning when consolidating datasets from MPOs and juris- dictions and using data maintained by different state departments. Although not all states indicated a specific method for ensuring consistency among datasets, several practices were notable toward this end. These include: • Kentucky, where the state transportation cabinet has funded a comprehensive quality review of bicycle and pedestrian infrastructure datasets. A consultant will perform a quality review of the data gathered by the state and ADD offices in recent years to establish a comprehensive, consistent dataset that informs project selection. This is a direct result of local governments requesting a larger number of bicycle and pedestrian improvements in coordination with state funding mechanisms. In fact, it was noted that an estimated 90% of projects submitted to the state for review are bicycle- or pedestrian-related. • Louisiana and Utah, which have worked internally to establish data format requirements for bicycle and pedestrian datasets, cited FHWA guidance related to MIRE and as defined in the Coding Nonmotorized Station Location Information in the 2017 Traffic Monitoring Guide Format publication as the foundation for this information. In addition to consistency of data collected, consistency was referenced in terms of the con- tinuity of data collection programs. States including Washington indicated that data collection occurred previously but ceased as a comprehensive program for various reasons. Other states also noted previous data collection efforts – or, in the case of states such as New Hampshire, multiple departments are undertaking similar efforts to gather data on various infrastructure elements. Data Maintenance, Update Strategy and Program Funding Maintenance and Funding Four of the five states interviewed indicated that they have a formal data maintenance plan in place to keep data up to date over time. This plan most often addresses roadway assets and may cover items such as roadway shoulders, but also include active transportation facilities. Some data plans address who is responsible. Examples include: • Kentucky – Specifically for active transportation facilities, the process is ongoing and evolving toward the end of developing a comprehensive dataset. Actions toward maintaining these data include asking for community feedback during meetings with jurisdictions to verify facilities, comparing existing datasets to publicly available heatmaps to identify areas where a facility may be present, and providing official policies for ADD offices to review data at a minimum of every 3 years. • Louisiana – A third-party vendor updates data inventory every 3 years, and the department updates volume and roadway class with the same frequency. As projects are completed, data also should be updated. Sidewalks are included in the maintenance plan, but the frequency of updates is unknown. • Utah – LiDAR pavement collection is completed through a third-party vendor annually, and additional assets are collected every 2 years. Sidewalk data currently are not part of the main- tenance plan.

Case Examples 49 • Washington – A data governance policy states that data owners are responsible for mainte- nance of their data. Previously, this was advised by a governance body that was established per executive order. The group is to reconvene in the near future. Similar to data collection efforts, maintenance activities rely on funding availability. The states interviewed indicated that project-specific funding was an important source for mainte- nance, with either project files or GIS databases being updated at project completion. Federal funding, including HPMS and HSIP, also frequently was noted. In some instances, program budgets and larger data maintenance budgets also are used. Data Sharing Public availability of data varied among interviewed states. Two states indicated that data are available through an online portal or webmap, two indicated data are available through request, and one indicated that no formal mechanism exists for sharing data. However, it is important to note that format affected the way data may be shared. For example, GIS data were more likely to be available through a web-based format, while CAD or PDF files primarily were available by request only. States that primarily store pedestrian infrastructure information via CAD or PDF file formats also more commonly indicated concern with data sharing due to liability. Contrac- tual and privacy concerns were noted for states using data sources such as Strava or bike/scooter share data. Some states, including Washington and Pennsylvania, have begun the process of applying stricter GDPR standards to protect an individual’s privacy better. Additional Findings Outside of the specific survey focus areas, the case studies revealed several other trends or reiterated findings from the survey results. Among them: • Each state expressed interest in having a comprehensive, complete pedestrian infrastructure dataset. However, understanding where to begin, how to define the pedestrian network and how to fund the efforts were discussed as barriers. Several state agencies expressed being understaffed or having multiple roles, in addition to serving as the bicycle and pedestrian coordinator. • A better understanding of the approach to the pedestrian network as it relates to federal guidelines was noted by several respondents. Requirements related to HPMS or HSIP report- ing were noted, even when primarily applicable to roadway infrastructure, along with ADA requirements. Particularly for states that have multiple departments involved in pedestrian infrastructure data, consistency among understanding of guidelines and agency approach was limited. • Both departmental structure and staff capacity drive the format and comprehensiveness of pedestrian infrastructure data. For states with multiple departments handling data, there was not always clear understanding about data availability and quality among departments. • Data collection practices are built on historical processes, such as relying on as-builts or out- of-date database systems, or data collection practices are not carried through staffing changes, departmental reorganizations or similar occurrences. • Several states identified a certain level of distrust in existing data that they did not specifically have a hand in collecting. For instance, existing out-of-date datasets were referenced because there is little understanding about how the data came to be or how accurate they are. Similarly, as-builts were identified by several participants as questionable due to experience.