Integration of Roadway Safety Data from State and Local Sources (2018)

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.

45 The information in this chapter was derived from detailed interviews with individuals from several organizations in nine states selected to serve as case examples (California, Illinois, Iowa, Kentucky, Louisiana, Michigan, Ohio, Vermont, and Washington). The list of interviewees is provided in Appendix C. This chapter generally addresses the following four topic areas with each agency: 1. Current status of MIRE FDE collection; 2. Tools used for data collection and mechanisms; 3. Maintenance and updating status of MIRE FDEs; and 4. Observed benefits, agency collaboration efforts, and lessons learned with MIRE FDE integration. The original scope of this synthesis was to conduct in-depth interviews with 15 LPAs, but, ultimately, 28 LPAs were interviewed. Several criteria were used to select the agencies to serve as case examples. First, the states that indicated their agency was currently integrating local (non-state-owned) roadway safety MIRE FDE systems into the state DOT MIRE FDE system, whether through the literature review or the survey, were selected. A further criterion applied was whether the state DOT indicated it had a successful program in place to integrate roadway safety MIRE FDE information from both state and local sources. Geographic distribution, differing organizational approaches to delivering local safety projects, and the percentage of local road ownership (more than 60% of the total lane miles in each state) were also considerations. The California and Washington DOTs were also selected because of their planned efforts for collection of MIRE FDEs on all public roads. Table 18 summarizes the survey response from each state and the selection criteria. Information was also gathered through cursory interviews with four additional state DOTs with experience on the topic of MIRE FDEs (Georgia, Kansas, Tennessee, and Wisconsin), FHWA, and TTAP. The information from these interviews is presented at the conclusion of this chapter. Because of the continually evolving nature of the data collection and information technology fields, it should be noted that the information in this chapter requires frequent updating in the future. California The 2015 census revealed a population of 39.14 million in California, where there are 18 MPOs, 10 regional transportation planning agencies (RTPAs), 58 counties and 482 cities. With approxi- mately 180,000 centerline miles, the California public roads are divided into two main systems by ownership: the State Highway System (SHS), which is owned and maintained by Caltrans, and the non–State Highway System (non-SHS) which is owned and maintained primarily by C H A P T E R 4 Case Examples of Data Integration and Maintenance Efforts

46 Integration of Roadway Safety Data from State and Local Sources local agencies along with some tribal, federal, or other state agencies. Caltrans is organized into 12 districts, and some of the MIRE FDEs on local roads are collected by the Research, Innovation and System Information Division, HPMS Branch. As required for all other states, California’s Traffic Records Coordinating Committee was developing a MIRE FDE Project Management Plan, which was to be included in its strategic plan by July 1, 2017. This plan is intended to guide California’s efforts in (1) collecting and maintaining MIRE FDE elements and (2) developing systems to store, manage, and allow for access to MIRE FDEs by state, local, and tribal agencies. Caltrans performed a gap analysis in December 2016 as part of the effort to meet the new MIRE FDE requirement. The three major sources of roadway and traffic data on California public roads were as follows: • Transportation System Network (TSN) database, which is owned and maintained by Caltrans; • HPMS, the federally mandated inventory system; and • Local databases, for which the extent of data coverage varies depending on the size and resources of the local agency. By defining a gap as when an FDE has never been collected and maintained in an existing database (or exists but is outdated), the analysis found that most of the data gaps between the required FDE and the existing inventory databases were associated with the non-SHS. These data gaps were AADT, number of through lanes, ownership, and intersection-related FDEs. Some of the recommended actions for addressing the identified gaps were development of a program for coordination with stakeholders and local agencies, additional funds allocated for AADT collec- tion on the non-SHS, development of a data management and integration plan (e.g., MIRE FDEs with crash data, HPMS data, and other data types), and the creation of a standardized data file format or structure. State Local Road Ownershipa (%) Organizational Approach of Delivering Local Safety Projectsb Is DOT currently integrating local (non- state-owned) roadway MIRE FDEs into the state DOT MIRE FDE system?c Does state DOT have a successful program in place to integrate roadway safety MIRE FDE information from both state and local sources?c California 87 Both Central Office (planning and programming) and District Office staff (implementation) na na Kentucky 64 Both Central Office (planning and programming) and District Office staff (implementation) Yes Yes Illinois 89 Both Central Office (planning and programming) and District Office staff (implementation) Yes Yes Iowa 92 Both Central Office (planning and programming) and District Office staff (implementation) Yes Yes Louisiana 72 Central Office staff, District Office staff, and assistance from LTAP Yes Yes Michigan 91 Central Office staff Yes Yes Ohio 84 Combination of Central Office, District Office, and LTAP staff Yes Yes Vermont 80 Central Office staff Yes Yes Washington 80 Both Central Office (planning and programming) and District Office staff (implementation) na na Note: na = not applicable. a Source: Highway Statistics 2015, Table HM-10 (FHWA 2016b). Percentages represent centerline miles. b Source: NCHRP Synthesis 486: State Practices for Local Road Safety (Park et al. 2016). c Source: Appendix B. Table 18. Summary of selection criteria for nine case example states.

Case Examples of Data Integration and Maintenance Efforts 47 Interviews with representatives from Orange County, the Kern Council of Governments, and the City of Watsonville indicated that the routine HPMS and pavement condition data (submitted in Excel file format to Caltrans) constituted the MIRE FDE-related data shared. An example of the type of HPMS data submitted is included in Appendix D. In the case of the pavement condition data, the local agencies reported that the StreetSaver© software (StreetSaver 2017) was applied to calculate the distress level and that those results were submitted to Caltrans on a biannual basis. Illinois At the Illinois DOT, the Office of Planning and Programming oversees the collection, mainte- nance, and updates of MIRE FDEs as part of the base requirement within the roadway inventory system that also includes the local (non-state-owned) roadway data inventory. The Illinois DOT posts a year-end GIS file to its Internet site for access by local agencies and practitioners. Champaign County Regional Planning Commission The Planning and Community Development Division at the Champaign County RPC (the MPO for the Champaign–Urbana urbanized area) indicated that its role is to serve as a data clearinghouse for local agencies and the Illinois DOT. Until 2011, the MPO conducted traffic counts every 5 years and provided the data to the Illinois DOT. In 2016, the Illinois DOT con- tracted a consultant for this effort, and, at the time of the interview, the MPO collected traffic counts on an as-needed project basis. The statewide traffic count database is maintained by the Illinois DOT (https://www.gettingaroundillinois.com/gai.htm?mt=aadt); however, for data related to specific roadway sections, the MPO has access to the Illinois DOT GIS layer. The MPO is responsible for maintaining and updating the data that reside under the Champaign–Urbana Urbanized Area Transportation Study (http://data.cuuats.org/), an open data portal that provides various transportation data (e.g., crashes, traffic counts, bike facilities, pedestrian facilities), land use, and socioeconomic data for Champaign County or Champaign– Urbana. The data are updated automatically whenever there is any modification, and the Illinois DOT can request data access from the MPO by sharing shapefiles. This database serves as a measure in terms of objectives and goals for the MPO when it establishes the long-range transportation plan. The Champaign County RPC works closely with the Illinois DOT, and the continuous com- munication allows the MPO to inform the Illinois DOT of its projects and plans. The observed benefits of working with the state agency on data integration include resource savings (e.g., cost and time) and a data collection and management process streamlined by having a common data set and maintaining data consistency. Iowa In the state of Iowa, the Iowa DOT Office of Research and Analytics oversees MIRE FDE collection and maintenance. The Iowa DOT reported that with the use of an LRS, integrated MIRE FDE information is stored in the Roadway Asset Management System, which is built in the Esri Roads and Highways ArcGIS extension platform (Esri 2017). A recent report, State of Iowa Traffic Records Assessment (Florey et al. 2015), indicated that the state does not yet have a data governance process, and the Iowa DOT is working on a governance process from a state- wide perspective. Iowa’s Statewide Traffic Records Coordinating Committee is exploring data governance guidelines that will be implemented for improved integration and access. The report

48 Integration of Roadway Safety Data from State and Local Sources also noted that the Iowa DOT has achieved a high level of data integration of its state-level roadway data systems and crash database. Through the use of common services and tools, the integrated roadway information is shared with local transportation agencies. A future task identified was the establishment of integration performance measures based on a formal process of metrics, baseline measures, and monitoring of progress over time as compared with expected goals or targets. Interviews with the City of Ankeny, Iowa, and the City of Des Moines, Iowa, emphasized the significant role of Iowa State University’s Institute for Transportation (InTrans) in roadway data collection and analysis. Institute for Transportation at Iowa State University The Center for Transportation Research and Education at InTrans at Iowa State University in Ames has been responsible for populating a statewide intersection database and refining and supplementing a roadway horizontal curvature database, both of which are significantly related to the MIRE FDE components. The ongoing Statewide Intersection Database was established in 2013 and includes approxi- mately 166,315 intersections (561,152 approaches) on all public roads. The following MIRE FDEs are collected through this project: Unique Junction Identifier, Location Identifier for Road 1 and 2 Crossing Points, Intersection/Junction Traffic Control, and Intersection/Junction Geom- etry. The initial comprehensive version has been completed, and the emphasis areas during the creation of the database included stressing the importance of the database and its end results to coders, standard processes, QA/QC, and program monitoring (e.g., peer monitoring, continuous communication, and feedback). The details of the project are presented in Appendix D. Another ongoing project related to MIRE FDE collection is the Curvature Database for all pub- lic roads. The University of Wisconsin Traffic Operations and Safety Laboratory algorithm is used to extract horizontal curve data and characteristics from the Iowa DOT’s LRS data from mostly rural areas. InTrans reported that it has been responsible for reviewing and refining the horizontal curve data as well as integrating additional attributes into the data, such as curve-related signage. Iowa State University’s InTrans assists the Iowa DOT with monitoring the collection of pavement condition data, quality assurance, and data delivery to the locals. The Iowa DOT hires a contractor to collect pavement condition data, which it then analyzes for use in pavement maintenance and rehabilitation decisions. InTrans does comprehensive QA on pavement distress data, summarizes the condition, and then communicates it to local agencies. Pavement data on state National Highway System routes are collected annually, while information on the remaining roadways is collected every 2 years. City of Ankeny The City of Ankeny is located in Polk County in central Iowa and has a population of 45,562 (as of the 2010 census). The City of Ankeny reported that MIRE FDEs exist to an extent in the crash reports and in an in-house GIS street database. The City of Ankeny indicated the significance and benefits of the Iowa DOT’s Crash Mapping Analysis Tool (CMAT), a GIS-based crash analysis tool developed by the Center for Transportation Research and Education at Iowa State University. Crash reports from the police are completed with Traffic and Criminal Software, a data collection, forms management, and reporting software for public safety professionals. The coded crash data are linked to and processed by CMAT, a tool that was reported to be very valuable to both local agencies and to the consultants hired to support local agency efforts. The City of Ankeny has an in-house GIS technician who provides the shapefile for the street database with updated streets to the Iowa DOT. The new street information is typically updated

Case Examples of Data Integration and Maintenance Efforts 49 annually, while the traffic count data are updated every 4 years. The City of Ankeny does not collect pavement data; rather, the Iowa DOT collects state-level pavement data with a special van equipped with a camera. These state-level pavement data could potentially be better utilized by the City of Ankeny in the future if workload allows or staff levels are expanded. Another benefit mentioned was to leverage the resources available through InTrans and collaborate with the Institute. City of Des Moines The City of Des Moines is the capital of Iowa and is also located in Polk County. It has a population of 203,433 (as of the 2010 census) and owns 913 miles of roadways in the Des Moines metropolitan area. The City of Des Moines maintains centerline information for its enterprise applications and 911 dispatching in a geodatabase. Both new and modified roadway centerlines are provided to the Iowa DOT on a biannual basis; AADT is collected by the Iowa DOT every 4 years. Pavement data are collected by the Iowa DOT and then are analyzed by InTrans to obtain a pavement condition index that is provided to the City of Des Moines for use in scheduling roadway maintenance. Kentucky At the Kentucky Transportation Cabinet (KYTC) Department of Highways, the Data Man- agement Branch of the Planning Division oversees the collection and maintenance of MIRE FDEs KYTC contracts with Area Development Districts (ADDs, similar to an RPO) to work with local agencies to collect non-state-road locations, surface types, ownership, street names, and one-/two-way operations. KYTC requires the data collection to fit into the DOT’s system, which uses a well-established data dictionary, Local Road Update Standards (Kentucky Trans- portation Cabinet 2017) (see Appendix E). KYTC mentioned another useful manual, Standards for Road Data Collection & Maintenance Using Global Positioning System Techniques (Kentucky Transportation Cabinet 2004), which is used for effective data collection with GPS (see Appendix E). Currently, KYTC is planning to develop AADT estimates on local roads where no actual values are being collected. Gateway Area Development District The Gateway ADD is located in Morehead, Rowan County, in eastern Kentucky and over- sees Bath, Menifee, Montgomery, Morgan, and Rowan Counties. The Gateway ADD has been involved in collecting MIRE FDEs since 2010. The ADD has acted as the clearinghouse for data being submitted for the statewide centerline project. Tools applied to collect roadway traffic data include the Trimble Geo7x, a GPS tool that has a Zephyr Model 2 antenna and an image system. All collected data are saved in shapefile format and are sent over a SharePoint site in a zipped folder to share with KYTC. The roadway inventory data updates are carried out on an as-needed basis, and a city or county LPA will inform the ADD of any change. The ADD will go out to collect new data and information as a result. Data maintenance is the responsibility of the Gateway ADD, with support from LPAs. The observed benefits noted by the Gateway ADD of collaborating with KYTC on MIRE FDE collection is the trust built between the two agencies. Lake Cumberland Area Development District The Lake Cumberland ADD is located in Russell Springs in south central Kentucky and serves Adair, Casey, Clinton, Cumberland, Green, McCreary, Pulaski, Russell, Taylor, and Wayne

50 Integration of Roadway Safety Data from State and Local Sources Counties. The Lake Cumberland ADD has been involved in collecting MIRE FDEs since 2000, when the state of Kentucky’s 15 ADDs began collecting data for all of the state’s 120 counties. KYTC decided there was much value in having highly accurate data, as hardware and tech- nology became more accurate and affordable. KYTC currently collects all data related to state roads, while the ADDs collect local road information. In previous years, ADDs collected some state centerline and other miscellaneous data. Whenever the ADD is notified of any changes to county and local roads, the ADD staff drive the corresponding roadway segments with GPS tools to collect data. All GPS data are processed with ArcGIS and submitted to KYTC. The ADD acts as a liaison between the 10 counties and the state, collecting data changes from municipalities and counties and providing them to the state. Changes are made in ArcGIS and then submitted to KYTC via a secured FTP site. KYTC has also created an interactive ArcGIS-based website (http://maps.kytc.ky.gov/photolog/?config=LocalRoads) that enables LPAs to view their current road system. Although access is provided to LPAs, actual updates to the map are submitted via a secure FTP site, and access is limited to KYTC and the ADDs. Data are updated whenever changes are made to state and local roads, as KYTC aims to maintain a living road update system. Data maintenance falls under the responsibility of KYTC, but the ADDs’ interaction with county and municipal governments is important as well. The Lake Cumberland ADD noted that collabo- ration with KYTC has built trust among the agencies and kept the LPAs informed about the importance of any project, which is crucial to successful data collection and integration practices. Lincoln Trail Area Development District The Lincoln Trail ADD is located in Elizabethtown in north central Kentucky and serves Breckinridge, Grayson, Hardin, LaRue, Marion, Meade, Nelson, and Washington Counties. The Lincoln Trail ADD has been involved in collecting MIRE FDEs since the start of the statewide centerline project in 2000 and has acted as the clearinghouse for data being submitted for the centerline project. The majority of inputs that the Lincoln Trail ADD provides to the state are roadway surface types, road names, one-/two-way operation, and government ownership. The centerlines and their lengths are mainly collected with a GPS tool and an aerial image view- ing system to better locate the exact site. Data are provided to KYTC as shapefiles via a secured FTP site. At one time, data were updated once or twice annually, but updating is now done more on an as-needed basis. Usually, one of several entities—the 911 dispatch, the Planning and Zoning Office, the County Judge/Executive Office, the County Road Department, or County Engineers—will inform the ADD of any changes that necessitate the collection of new informa- tion. The data collection process is ongoing throughout the year, and the statewide data set is updated within 2 to 5 days after data are sent to KYTC. If there is a rush by the county or city, a change can be submitted immediately, and the state will process the update upon receiving it. Otherwise the Lincoln Trail ADD collects data over time and submits them as the agency feels is needed or quarterly as a minimum requirement. It is the ADD’s responsibility to maintain the data and, thus far, the ADD’s collaboration with KYTC on MIRE FDE collection has built more trust between the agencies. To improve data integration in the future, the Lincoln Trail ADD has recommended developing both a common template and a dictionary to familiar- ize everyone involved with the standard terminology as well as to maintain good relationships between agencies. Louisiana In the state of Louisiana, the collection and maintenance of MIRE FDE information is per- formed by the Louisiana DOTD Office of Planning and Programming as well as by consultants who are contracted by various agencies to collect and deliver the data to the Louisiana DOTD.

Case Examples of Data Integration and Maintenance Efforts 51 Data are updated every 2 years, while updates for local road data depend on each LPA’s own timing. One of the challenges identified by the Louisiana DOTD was having local agencies update their own data in the future. To address this challenge, the Louisiana DOTD plans to meet with LPAs prior to initiating data collection efforts in order to obtain their support, although the level of commitment from the LPAs was reported to be unknown at the time of the interview. Interviews with a few local agencies highlighted the important role of the MPO and the RPC in MIRE FDE collection and updates. The GIS and Mapping group in the Office of Planning and Programming at the Louisiana DOTD indicated that the state is leading the MIRE FDE collec- tion in collaboration with LPAs. The Louisiana DOTD is responsible for data QA/QC on the information exchanged via the FTP site. According to the Louisiana DOTD, of 64 parishes, 24 have delivered their inventory of local road assets to the Louisiana DOTD, 39 have collected their asset data and are awaiting QA of the data, and the remaining parish asset information has not yet been delivered to the Louisiana DOTD for QC. Baton Rouge Metropolitan Planning Organization The Baton Rouge MPO, the Capital Region Planning Commission (CRPC), is located in southern Louisiana. The metropolitan area includes five parishes/counties (2 in entirety and 3 partially) and 18 municipalities and has a population of 714,254 (as of the 2015 census). The Baton Rouge MPO has been involved in collecting MIRE FDEs since 2014 with the use of Fugro Roadware (Fugro 2017) for non-state-owned roadways. As part of its Transportation Asset Management Program, the Baton Rouge MPO collaborated with the Louisiana DOTD to collect location and condition data for pavements and various transportation assets on non-state-owned roads. The Louisiana DOTD collected similar data on state-owned roads. The purpose of this project is to foster systemic safety improvements and also optimal use of limited dollars for effi- cient preventive maintenance. The raw right-of-way and pavement videos were post processed by Fugro Roadware to extract the location and estimate the roadway asset condition informa- tion. The asset and pavement condition data were tied to a linear-referenced map. The roadway condition data along with the raw videos that were collected were visualized for quality control purposes in the web-based tool iVision, a Fugro Roadware product, provided by the consultant. The MPO utilizes the Deighton Total Infrastructure Management System (dTIMS) to analyze data obtained from Fugro Roadware, tests various budget scenarios, and provides the results to the municipalities. The goal is to analyze these data for each parish and municipality for which the MPO has pavement condition data and publish the analysis results on a web-based dashboard where LPAs can visualize the results. Although reporting the analyzed data back to the state is not required, the Baton Rouge MPO shares the results with the Louisiana DOTD to facilitate consistency. The dTIMS is the pavement and asset condition analysis software that assists in optimizing the prioritization of preventive maintenance projects and ultimately saves time and costs for the agencies. An important data source for CRPC’s transportation planning efforts is the Regional Traffic Count Program, which began in 2017. CRPC conducts its traffic volume data collection (traffic counts) program for its designated transportation study area, which includes the Ascension, East Baton Rouge, Iberville, Livingston, and West Baton Rouge Parishes. The CRPC traffic count database system is presented on an interactive map through GIS Online for the Baton Rouge MPO’s transportation study area. The goal is to systematically collect traffic volume data on key links in the Baton Rouge MPO network so that these data can be used for analysis that supports transportation research or land use and development. Figure 21 is a screenshot of the Baton Rouge MPO’s Regional Traffic Count Program. The Louisiana DOTD and Baton Rouge MPO reported they are working well with one another in terms of exchanging data. The benefits listed by the Baton Rouge MPO include improved data

52 Integration of Roadway Safety Data from State and Local Sources content, use of a data collection software, and an analysis program that helps all of the parties involved in maintaining consistency. City of Baton Rouge and Parish of East Baton Rouge The City of Baton Rouge and Parish of East Baton Rouge have been involved in the collection of MIRE FDEs since 2013. The agencies provide a shapefile export of the following data to the Louisiana DOTD: • Street centerline data, including segment ID; • Route number; • Street name; • Route type; • Segment length; • Roadway functional class; • Number of lanes; and • Ownership. The data are collected via field surveys, aerial imagery, and subdivision plat digitization from AutoCAD files. The data are updated whenever streets are created or revoked, which the LPAs reported typically happens once per month. These edits to street layers are the responsibility of Figure 21. Capital Region Planning Commission traffic count program.

Case Examples of Data Integration and Maintenance Efforts 53 the GIS coordinator, while the data are maintained by the Department of Information Services in the city and parish. Although the Louisiana DOTD’s collection practices are separate from those of the parishes, the parish offers assistance in providing updated local street information. The City of Baton Rouge and Parish of East Baton Rouge reported they have not yet integrated the safety data with their LRS. However, the Parish of East Baton Rouge acquired crash data to analyze pedestrian and bicycle crashes to determine where the most vehicle crashes occur. As for recent updates to the LRS system, the Parish of East Baton Rouge has added additional attributes to the LRS data set in order to conform to National Emergency Number Association standards for improved emergency response. City of New Orleans The City of New Orleans reported that it is involved in collecting MIRE FDEs through different divisions. Approximately 50% of the street attributes (e.g., surface type, segment length) are collected, but not all of the information is in digital format (e.g., street direction is in paper format). Because of the city’s location near the Mississippi River and other natural barriers, most of the updates to the roadway data stem from renovations and road repairs. The City of New Orleans Public Works and Public Roads offices met with the Louisiana DOTD in 2016 to discuss the details of data collection. Data are collected and stored using Esri’s ArcGIS Local Government Information Model, which is a harmonized information model of GIS data sets designed to support the data management and analysis maps and applications deployed with an ArcGIS geodatabase. The data are then provided to the Louisiana DOTD by using the open data portal for centerline information, which was initiated in 2012 to improve access and allow for automatic updates and modification (https://data.nola.gov/login). Data are updated whenever a change occurs (e.g., a street name is changed to honor an individual in the community), at which point the information is sent to the city and posted to the open data portal. The open data portal was reported to have expedited the process of sharing data by removing the long legal process of cooperative endeavor agreements. The biggest chal- lenges for data collection reported by the City of New Orleans have been limited resources and reduced staff. New Orleans Regional Planning Commission The New Orleans RPC is based in New Orleans and includes the eight parishes of Jefferson, Orleans, Plaquemines, Saint Bernard, Saint Charles, Saint John the Baptist, Saint Tammany, and Tangipahoa. The New Orleans RPC gathers geospatial data from local agencies and then sends those to the Louisiana DOTD. The New Orleans RPC then coordinated with the Louisiana DOTD for the data transfer. In 2012, the Louisiana DOTD returned the updated geospatial data to the New Orleans RPC, which then began mapping its projects and planned projects using the Louisiana DOTD geodatabase. The practice reported at the time of the interview was that the New Orleans RPC gathers geospatial data from the LPAs and sends it to the Louisiana DOTD as part of its general workflow. More specifically, the New Orleans RPC provides the local road layers from parishes and cities that were originally used for determining local road categories. While the New Orleans RPC does not provide the Louisiana DOTD with specific segmentation elements, it does forward information on local road layers and average daily traffic (ADT) collected by the LPAs and by RPC contracts to the Louisiana DOTD. Various software packages ranging from GIS software, to Trimble, to Excel are used as data collection tools. All data are stored, analyzed, and amalgamated at the New Orleans RPC with ArcGIS version 10.3.1 (as of the time of the interview). The data are then sent to the Louisiana DOTD in the form of shapefiles and geodatabases. As of the time of the interview, the New Orleans RPC sends new roadway projects

54 Integration of Roadway Safety Data from State and Local Sources directly to the Louisiana DOTD as they are developed and encourages the LPAs to send all updates to the roadway geodatabase directly to the Louisiana DOTD. Thus far, the New Orleans RPC indicated, the only known direct contact has been the City of New Orleans with the Louisiana DOTD. The New Orleans RPC geospatial staff works with project managers to manage local projects, but its staff does not have the authority to mandate data sharing. Local governments are respon- sible for their local data layers, and the New Orleans RPC is responsible for its regional data layers. The New Orleans RPC updates its regional roadway project files on an as-needed basis and publishes data every December for any obligated projects. The TIP and MTP geodatabases are updated on an as-needed basis. The Louisiana DOTD included the New Orleans RPC as an example of a planning organization that uses the geodatabase. The Louisiana DOTD has made a concerted effort to include the New Orleans RPC as a pilot first user in terms of examining how it uses the geodatabase. The Louisiana DOTD has had its consultant work directly with the New Orleans RPC in order to better understand any necessary revisions, specifically in maintaining unique small-segment identifiers for road segments. The New Orleans RPC maintains a relational Access database by assigning unique ID numbers for tracking all projects and has developed the data to include the history of using these identifiers. This has allowed the New Orleans RPC to track investments along corridors of any kind over time including future planned projects and studies. The benefits reported by the New Orleans RPC include using the geodatabases and providing MIRE FDEs to the Louisiana DOTD. The New Orleans RPC reported that it has experienced a level of communication that has eased the process and helped to remove the confusion of past editing processes. Also, it reported that using the same format and interactive geospatial data helped to coordinate updates, particularly the streamlined and interactive updates associated with the roadway functional classification for Federal-aid projects. The reported lessons learned include the importance of understanding the interest of the locals in terms of attribute features of geospatial data. For example, a 911 dispatcher will be more interested in the address point at a street intersection, whereas a city engineer would be more interested in the ingress–egress point. Also, rather than asking LPAs to follow and use the standard format of GIS data, the New Orleans RPC reported that it was more effective to approach LPAs by providing data that would help them solve some of their own issues and ask for the updates and maintenance. A simple lookup table by unique ID should be the next step: the unique ID from the local data- base matching as close as possible to a unique ID in the state database, a process that requires significant resources (e.g., time, staff, funds) as it is not a one-to-one match. The unique ID on a small linear segment that would match among different databases would encourage LPAs to incorporate the Louisiana DOTD geodatabase into their daily workflow. Michigan At the Michigan DOT, the Traffic and Safety Division and Data Integration and Inventory Division are collecting and maintaining MIRE FDEs. Currently, the Michigan DOT is develop- ing a plan for specific quantifiable and measurable anticipated improvements for the collection of MIRE FDEs and inputting them into the Michigan DOT’s Traffic Records Strategic Plan. The plan presents a high-level description of how the Michigan DOT will accomplish the collection of all MIRE FDEs in the state. More specifically, the Michigan DOT will collaborate statewide with its partners, including MPOs, RPCs, LPAs, and other state departments. The Michigan DOT will lead the discussion on determining necessary performance metrics, how to collect data, and how to achieve target values related to data errors, data sharing, timeliness, accuracy, completeness, uniformity, integration, and accessibility of available roadway information.

Case Examples of Data Integration and Maintenance Efforts 55 The integration, accessibility, and usability of roadway safety MIRE FDEs to facilitate mean- ingful crash analysis by all road agencies are the main goals of the implementation of geographic information technologies (GITs). The Michigan DOT will lead the selection and implementation of GITs, which will allow for future roadway data storage as well as the exchange and utilization of MIRE FDEs collected by the Michigan DOT and its collaborating partners. Early input from a select group of MPO and RPC partners emphasizes the need for the plan to identify the time and costs associated with collecting MIRE FDEs. Though the participation of MPOs, RPCs, and local agencies is not a guarantee, a sufficient level of MPO and RPC involvement is dependent upon the Michigan DOT including partners’ input and obtaining enough resources to accomplish the task. Table 19 shows the overall schedule of the proposed plan for collecting MIRE FDEs. The plan will be submitted in the SHSP by the Traffic Records Coordinating Committee. The plan shows that the Michigan DOT will lead the efforts and MTU will perform support services through its activities with the Roadsoft program. The Michigan DOT and MTU Center for Technology and Training indicated the importance of the Roadsoft software, in that while the majority of MIRE FDEs exist in Roadsoft (especially those for urban areas), several MIRE FDEs were found to be missing, namely, route number, direction of inventory, access control, median type, and one-/two-way operations. The research team at the MTU Center for Technology and Training reported that it was working on adding the missing elements and that the fields that already exist in Roadsoft would be upgraded so as to be populated automatically (Table 20). Fiscal Year Plan Description 2017 The Michigan DOT will collaborate with the Michigan Technological University (MTU) Center for Technology and Training to make Roadsoft a MIRE FDE–compliant collection tool. 2017–2018 The Michigan DOT and MTU will conduct a survey focusing on identification of (1) collected MIRE FDEs and (2) the data format. 2018 The Michigan DOT will build a state-level MIRE FDE repository using GIT that includes server configuration, a MIRE FDE model, a web-mapping application, two-way web services, and a reporting tool. The Unified Work Program (UWP) asks for MPOs and RPCs to participate in the state’s outreach efforts to refine the MIRE FDE collection plan. A main goal of the outreach is to determine the time and cost required. 2019 With a few select partners, the Michigan DOT will establish a pilot that uses the new tools and the data repository to collect and exchange MIRE FDEs. The goal is for the Michigan DOT to enhance the MIRE FDE model, web-mapping app, two-way web services, and reporting tools as determined appropriate on the basis of the pilot. The Michigan DOT will work with MTU to develop training and outreach materials. Continued partner participation is the goal in the FY 2019 UWP (details to be determined). 2020–2026 The first year of MIRE FDE collection will be 2020. The Michigan DOT will contract with MTU to create Roadsoft MIRE FDE training materials and webinars that will be used by the staff of partners and Michigan DOT staff. The Michigan DOT will engage partners to determine the best methods for collecting MIRE FDEs and the optimal frequency of collection to ensure that the state is compliant and can report all the required data by September 30, 2026. Continued partner participation is the goal in the FY 2020–2026 UWP (details to be determined). Table 19. Schedule for Michigan DOT MIRE FDE collection plan. Surface Type Automated value fill based on current value and options to be added Type of Government Ownership Automated value fill based on the legislated roadway classification per Michigan Act 51 Unique Approach Identifier Automated value fill based on leg direction Intersection/Junction Traffic Control Legacy field to be retained and options added Intersection/Junction Geometry Automated value fill based on geospatial line work MIRE FDE Automated Fill Scheme (https://www.michigan.gov/documents/act51simple_28749_7.pdf) Table 20. Michigan DOT MIRE FDE automated fill scheme for Roadsoft software.

56 Integration of Roadway Safety Data from State and Local Sources A few of the lessons learned mentioned by the Michigan DOT regarding data integration efforts included contextualizing the data element and collection effort for LPAs so that they can better understand the importance and necessity of each data element. This approach will also address the issues associated with LPA relationships with the Michigan DOT in highlighting that the relationship is not just for reporting. In local data collection efforts, the Michigan DOT reported it aims to make data collection on the local level a worthwhile endeavor for the LPAs by providing tools (e.g., Roadsoft software) to collect and use the data for each LPA’s own purposes. The Michigan DOT reported that it emphasizes to LPAs that the ability to have data from other agencies within the same land boundaries or adjacent to the LPA’s boundaries, along with the tools for crash analysis, are benefits that the process led by the Michigan DOT can provide. The Michigan DOT reported that capturing the extent of data collection accurately is an important practice because in many cases, the cost has often been underestimated, albeit being a benchmark for any effort. Ohio Over the past 10 years, the Ohio DOT has been working with the Ohio Department of Public Safety (ODPS) and county engineers through agreements to collect and maintain a common file of roadway inventory data that includes the MIRE FDEs needed for analyzing safety. This project, the Location-Based Referencing System (LBRS), is being funded through a variety of sources, including ODPS Section 408 grant funding under SAFETEA-LU, Ohio DOT HSIP funds, and county contributions. Before this project, Ohio DOT reported that it could only locate about 30% of all crashes within a given county because of inaccurate or incomplete local (non-state-owned) roadway databases. With the roadway inventory file now completed for 80 out of 88 counties in Ohio, approximately 90% of all crashes are locatable, regardless of jurisdiction. The Ohio DOT is currently looking at providing more HSIP funds to fund the remaining eight counties. Figure 22 presents the status of the LBRS in Ohio [Ohio Geographically Referenced Information Program (OGRIP)]. The interview with the Ohio DOT’s Office of Technical Services indicated the need to use a proactive approach, which helps to highlight the merit of accurate data, rather than a reactive approach with LPAs. Through this approach, the Ohio DOT initiated a process with the LPAs that benefits them because it should facilitate data use for many parties (e.g., 911 dispatchers, tax accounting office, school districts for school bus routing). The Ohio DOT mentioned that it hosted a meeting for county commissioners to explain the importance of an accurate and consistent database. Mid-Ohio Regional Planning Commission The Mid-Ohio Regional Planning Commission (MORPC) encompasses Delaware, Franklin, and Fairfield Counties, and it covers a combined population of 850,106 (as of the 2015 census). Although not directly involved in collecting MIRE FDEs, MORPC focuses its efforts on facilitating the maintenance of county-collected data. MORPC is not a data authority, but it works to manage regional data sets. The only exception is with Franklin County, in which the MORPC works with the county’s GIS users to edit the LBRS and to gather address point data. The MORPC makes the data available via the Esri Open Data site. The LBRS project is county based, so each county is responsible for getting its LBRS data to the Ohio DOT and OGRIP. MORPC relies on the LPAs to edit the files that are sent to the state. Esri ArcMap software is used to edit and maintain the data. The local GIS users replicate the data in their network, edit the data, and then synchro- nize the data back to the main files stored at the MORPC. For internal purposes, the MORPC periodically merges the data of 15 counties together. To provide the data to the state, MORPC

Case Examples of Data Integration and Maintenance Efforts 57 uses the OGRIP system that collects the county data in various ways. The biggest benefits that MORPC reported from overseeing the data-managing process have been enhanced systemic safety analysis and enhanced hot spot location analysis. MORPC reported it aims to balance the states’ and locals’ role in the future. At the time of the interview, LPAs had access to the database and updated their own data fields, which could cause confusion later (e.g., change of field attributes, formats that do not match). MORPC would like to help the LPAs to understand the importance of other MIRE FDE elements (e.g., median type) that are important to the Ohio DOT for planning and design of roadways. MORPC also reported that it plans to develop official county and city boundary files. Erie County Erie County is located in northern Ohio and has a population of 75,828 (as of the 2015 census). There are 401 lane miles of state highway in Erie County, and the county has been collecting some of the MIRE FDEs since 2001, with updates provided in 2003; the majority of these data come from the street centerline project and the address system for 911 dispatchers. Erie County has also applied to the Local Government Safety Capital Grant Program for funding for GIS database updates. The data were originally collected by Digital Data Technologies Inc. Figure 22. Status of LBRS in Ohio (OGRIP, http://ogrip.oit.ohio.gov/ProjectsInitiatives/LBRS/ LBRSStatus.aspx).

58 Integration of Roadway Safety Data from State and Local Sources (DDTI, https://www.ddti.net/), and at the time of the interview, data editing was done with the LBRS application from DDTI. The county reported that the SQL-based address and road centerline maintenance application was a helpful tool. The application contains a built-in QC function that allows consistent input of data as well as a data modification tracking function. Erie County provides the data to the state by using the Ohio DOT’s DDTI web update server. Data are updated whenever there are changes to addresses or roadways, and Erie County has 17 addressing authorities (villages, cities, and townships) that use the LBRS application to edit or add new addresses for its respective jurisdictions. The road editing is done by DDTI and the mapping staff of the Erie County Auditor’s Office, and the LRS system provides the report to the county engineer. The Ohio DOT LRS system provides the county engineer with a road mileage report that is used to reimburse local jurisdictions for road mileage. The creation of the GIS Advisory Board was reported as a successful practice, along with the LBRS application and with allowing the Ohio DOT web access to the data. The Erie GIS Advisory Board is unique and was initially created to include political subdivisions in Erie County (cities, townships, villages, and county organizations). The purpose of the Erie GIS Advisory Board was to share the cost of data, software, and professional services to facilitate delivery of data and solutions. As an example, a sample GIS Advisory Board agenda, meeting minutes, and memorandum of understanding are included in Appendix D. The state of Ohio was reported to benefit from the Erie County GIS Advisory Board as a result of the data sharing and relationships and participation in state programs such as the LBRS and Ohio Statewide Imagery Program. Erie County explained that the benefits of data collection include more exact 911 dispatch locations and the creation of secondary data sets to use for comparing with others. The key lessons learned reported by Erie County in terms of work- ing with the Ohio DOT are maintaining a good relationship between state and local agencies, employing quality control and standards, and communicating with data maintenance and end users regularly. Fairfield County Fairfield County is located in south central Ohio and has a population of 150,381 (as of the 2014 census). It has been involved in collecting MIRE FDEs through the centerline project, which includes the street centerline and the address system for 911 dispatchers. The project was developed to meet the needs of various LPAs; it started with all public roads designated as “rideable” and then expanded to unpaved and private roads. Fairfield County reported that roadway data are obtained by a consultant who manually collects information and has also used advanced survey equipment. Any new centerlines are copied in from recorded platforms by using coordinated geometry. FDEs are then provided to the Ohio DOT via the OGRIP system (http://ogrip.oit.ohio.gov/) through the FTP site. The data are updated through the LRS system, and afterwards a report is generated and reviewed by the County Engineer. The roadway center- line database is shared weekly with the LPAs and also through the OGRIP quarterly report and the Ohio DOT annual report. Fairfield County reported that it coordinates closely with the Ohio DOT and maintains an effective working relationship with the state. One benefit reported was the provision of more exact 911 dispatcher locations. The lessons learned reported for safety data integration included the importance of understanding each data field and its various attributes and of communicating effectively with the Ohio DOT to best understand each agency’s needs. Fairfield County also mentioned the need for consistent terminology and consolidation of attribute terminology to more successfully implement the collection of MIRE FDEs.

Case Examples of Data Integration and Maintenance Efforts 59 Vermont In the state of Vermont, there are 10 RPCs and one MPO, the Chittenden County Regional Planning Commission (CCRPC). The Vermont Agency of Transportation (VTrans) collects and maintains the master road centerline data that are used for transportation mapping efforts. This data set represents all public highways, including Interstates, U.S. and state highways, and town highways. With these data and the LRS created for local roads, VTrans has nearly all of the roadway segment MIRE FDEs, with intersection data as yet being only partially populated. More specifically, the VTrans Asset Management and Performance Bureau’s Data Section over- sees the majority of MIRE FDEs, while the Traffic Research Unit covers AADT and traffic-related items. VTrans collects road centerline data, with an analog process for LPAs and the GIS road centerline built and maintained by VTrans. Local municipalities, RPCs, and the MPO do not have direct access to update or modify road centerline data but do have the ability to provide feedback on any modifications that may be necessary, and municipalities annually file mileage certificates that document any changes that have occurred during the year. VTrans receives updates on town highways from RPC staff, who send an updated shapefile of a town with notations of the changes that RPC made in the data. VTrans is then able to easily transfer these changes into its master centerline data layer, thereby leveraging good local knowledge and research on the highways. This process was reported to follow a protocol that was established but did not necessarily follow the database schema that was once used (see http://vcgi.vermont.gov/sites/ vcgi/files/standards/partii_section_g.pdf, page 26). VTrans receives local AADT counts from the RPCs, and these counts are then integrated into the VTrans traffic monitoring system. The RPCs send VTrans their automatic traffic recorder counts every year as directed in their annual work plan. The CCRPC shares the VTrans database that enables its own traffic counts. VTrans reviews the counts and loads them into the database, a cloud-based system hosted by Midwestern Software Solutions (MS2). VTrans reported that the RPCs are timely in submitting their counts and that the quality of the counts is generally good. In terms of QA, VTrans uses standards as specified by the Procedure for Reviewing Counts and Assigning Station IDs (see Appendix D). Central Vermont Regional Planning Commission The Central Vermont RPC is located in Montpelier and encompasses Washington County’s 23 municipalities in central Vermont and has a population of 58,612 (as of the 2015 census). The Central Vermont RPC has been involved in collecting MIRE FDEs since 2011 with GIS and Esri software that constitute the primary means for sharing shapefiles. The Geospace database system was reported to allow easy input in the field, to help in standardizing the process, and to ease integration and migration of the data files. The RPC collects data for local municipalities on a request basis rather than on a scheduled basis. The RPC manages, updates, and maintains the data annually, although this is not a mandatory task. The raw basic files are then sent directly to VTrans by e-mail. Overall, the RPC has worked well with the state in terms of data collection, and, in return, the state was able to properly map the data. Since it began working with VTrans to provide MIRE FDE updates, the Central Vermont RPC has become well-connected with the state. The collaboration was reported to have introduced a peer review type of informational exchange at monthly meetings of the State Transportation Planning Initiative. Chittenden County Regional Planning Commission The Chittenden County RPC is located in Winooski, Vermont, and encompasses the 25 munic- ipalities of Chittenden County in western Vermont. The county has a population of 156,545 (as of the 2015 census), and the Chittenden County RPC collects AADT and turning movement

60 Integration of Roadway Safety Data from State and Local Sources data throughout the region. Most traffic counts are for specific projects, and the majority of data collection takes place between May and November. A large number of MIRE FDEs comes from the road centerline data managed by VTrans. CCRPC manually uploads individual data sets to the VTrans online database. This takes place annually, usually from January to March. Once the data are uploaded to the online system, VTrans is responsible for maintaining the data. Both agencies reported that since they began providing MIRE FDEs to the state, the result has been a more complete and integrated system that has greatly benefitted planning efforts. Windham Regional Commission The Windham Regional Commission is an RPC located in Brattleboro, Vermont. It encom- passes 27 towns in Windham County and portions of Bennington and Windsor Counties in southeastern Vermont and has an estimated population of 45,564 (as of the 2015 census). The Windham Regional Commission gathers some MIRE FDEs while performing updates of road names, surface types, and traffic data for the 27 towns it encompasses. It serves as a resource for the state in gathering local data, especially for towns that do not have the ability or resources to collect and review the data. The majority of the Windham Regional Commission’s data are collected via GIS and then provided to the state via ArcGIS files. The commission aids munici- palities in data collection, specifically of data about road surface types. Towns often do not report roadway modifications or updates to VTrans, and the Windham Regional Commission reported its attempt to mend the observed disconnect by collecting or updating any missed data. One tactic reported by Windham Regional Commission is to use a town road foreman, who can verify the changes the Windham Regional Commission passes on to VTrans. VTrans is responsible for maintaining the data. Washington State The state of Washington is divided into 39 counties and 281 municipalities. It has a popula- tion of 7.3 million (as of the 2015 census), with approximately 167,132 miles of roadway. In the state of Washington, the Transportation Data, GIS, and Modeling Office within the Multimodal Planning Division collects and maintain MIRE FDEs. The Washington State DOT collects and stewards individual MIRE FDE elements in various systems as part of its standard procedures. For instance, the Washington State DOT’s LRS and HPMS business functions include many MIRE data elements, and the Washington State DOT is working on integrating these data resources into a GIS-based LRS. Furthermore, the Washington State DOT is also exploring the potential for implementing an asset management system. To meet HSIP U.S. 23 CFR 924, which notes that a state DOT should have access to a complete collection of roadway and traffic data for all public roads by September 2026, the Washington State DOT initiated a survey (included in Appendix D, along with the survey results) that was sent to all LPAs in the state to identify which of the MIRE FDEs were being collected by individual agencies. In Washington State, the county data are mostly available through the County Road Administration Board (CRAB) database, but at the time of the interview, there was no equivalent for cities. Another main objective of the survey was to identify the correct contact person at each local agency for future data collaboration associated with MIRE FDEs. County Road Administration Board The Washington State CRAB provides accountability through “standards of good practice,” fair administration of funding programs, and technical and professional assistance to the state’s 39 county road departments in accordance with Revised Code of Washington 36.78.070. The 39 counties own, operate, and maintain the transportation system, which includes approximately 40,000 miles of roads, 3,300 bridges, and 4 ferry systems. CRAB has been involved in collecting

Case Examples of Data Integration and Maintenance Efforts 61 MIRE FDEs, which are stored primarily in CRAB’s Mobility Dashboard (http://www.crab.wa.gov/ Technology/Mobility/Help40/Overview_1.htm), a comprehensive road inventory and manage- ment system software program. The Mobility Dashboard contains four management systems: 1. Infrastructure Asset Management, 2. Pavement Management, 3. Maintenance Management, and 4. Systemic Safety Project Selection Tool. The Mobility program was reported to enhance a county’s ability to make quality decisions through consistent, equitable, and defensible management plans and operations. Each of the 39 counties has a road log manager to update the Mobility database. The counties collect their own data and are also responsible for maintaining and updating the data set. Each county must submit to CRAB an updated road log for its complete road system with all the data elements by the annual deadline of May 1. After this process, CRAB performs QA and certifies the roadway data sets that were provided and receives approval by CRAB. On the basis of these data, CRAB issues a Certificate of Good Practices to the State Treasurer, which directs disbursement of the Motor Vehicle Fuel Tax. CRAB is responsible for distributing the counties’ portion of the Motor Vehicle Fuel Tax the following year. In August, CRAB meets with the Washington State DOT and provides the road log data in the form of CDs and hard copy. CRAB also offers training in Mobility, VisRate (a program for collecting visual pavement distress data), and Design Systems to all county road departments and provides technical support for each county. Overall, CRAB acts as a clearinghouse by ensuring compliance, quality work, and training and by maintaining data for each of the counties. CRAB works with the Washington State DOT to help identify gaps and overlaps in the road inventory system, to increase the success of identifying jurisdictional boundaries, to explore multimodal opportunities, and to foster sharing of data management. CRAB reported that working with the Washington State DOT has also emphasized the importance of data consistency, recognition of new technology, the need for more staffing and funding, and the use of a systematic safety project selection tool—a comprehensive method for safety planning and implementation that supplements traditional site analysis. Figures 23, 24, and 25 present screenshots of the Mobility software that contains a roadway inventory of vertical curves, horizontal curves, and clear zones. A report provided during the interview with CRAB presents a good summary of solutions and the feasibility of addressing issues of roadway data integration (Transcend Spatial Solutions 2015). In particular, the report recommends upgrading the Washington State DOT’s current road inventory systems into more state-of-the-art technologies that would leverage and enable the integration of the GIS, LRS, mobile data collection, and road inventory capabilities into a single system. One suggested solution is the Esri Roads and Highways system (http://www. esri.com/software/arcgis/extensions/roads-and-highways) that would allow maintenance of the GIS, related LRS, and road inventory data in a technical setting and synchronized manner. This effort was reported to eventually minimize the level of maintenance effort and any potential errors. Other suggested solutions that centered on the enhancement of the Mobility system at CRAB included additional road inventory fields based on HPMS data gaps and development of functionality to recognize the extent of HPMS sample segments. The report also acknowledges that funding levels and staff limitations are key constraints. Federal Highway Administration Interviews with the FHWA Office of Safety and Office of Safety Research and Development indicated a need for consistency in the definition of roadway elements, particularly between the HPMS and the MIRE FDEs. One example given was the element of pavement type. At the time of the interviews, revisions to the HPMS were expected within a few months, and consideration of

62 Integration of Roadway Safety Data from State and Local Sources Figure 23. Roadway data inventory of horizontal curves (CRAB Mobility software). Figure 24. Roadway data inventory of vertical curves (CRAB Mobility software).

Case Examples of Data Integration and Maintenance Efforts 63 common data elements within the MIRE FDEs and the HPMS was identified as an item of signifi- cance for review. In addition, the FHWA Safety Data Business Planning program was mentioned as a potential tool for facilitating the collection and integration of MIRE FDE data. Defining the challenges of collecting MIRE FDEs was reported to be difficult, as each state is different in terms of size and various regulations at the state level. In addition, an interview with the LPA Program in the Office of Program Administration indicated a need for more coordination with the Office of Safety with respect to MIRE FDE requirements. Two of the practices identified by FHWA to help address the gaps between the LPA program office and MIRE program office were (1) lever- aging the LPA program in each state to facilitate data submission by LPAs when they apply for federal funding (e.g., HSIP, Safe Routes to School) and (2) adding video clips featuring safety data and MIRE FDEs to the existing Federal Aid Essentials website (FHWA 2017a). Georgia The Georgia DOT reported that it was contracting with the state’s regional planning partner organizations (regional commissions), which already have working relationships with all 690 local governments, under one umbrella called the Georgia Association of Regional Commis- sions. The association first validates some of the minimal data elements that the Georgia DOT already has in its statewide LRS and GIS data systems, which means that the local agencies only need to correct and update data where needed. This first program iteration is designed not to change the actual GIS or positional geometry of any of the local road representations in the state- maintained LRS system, but rather to update the information and associated attributes about Figure 25. Roadway data inventory of clear zones (CRAB Mobility software).

64 Integration of Roadway Safety Data from State and Local Sources those roads where needed. Depending on the technology resources and GIS capabilities available at the various LPAs, this process was reported to allow the opportunity for GIS-knowledgeable or larger local governments to perform their own analysis, conflation, and updates as they are able. However, the Georgia DOT reported that for LPAs that are limited in terms of technology capabilities or for smaller local governments in rural areas or small cities, the option exists to use the assistance of the GIS staff in the regional commissions in the data collection and integration effort. In many cases, for the smaller or GIS-limited local governments, a series of questions (when presented with the data) may suffice. This process was reported not to require the LPAs to change any of the data structures they were already maintaining to be in exact alignment with Georgia DOT data or vice versa, but the process helps to ensure that the information the Georgia DOT has for them is correct. In addition, the process reported by the Georgia DOT allows for future communication of any changes to the existing system as the program grows and matures in subsequent years. The Georgia DOT reported that by working collaboratively with the regional commissions in this manner, it has been able to determine with empirical data the number of LPAs that are either competent or deficient with GIS tools and can ascertain whether a statewide GIS local update system could be successful. Kansas The Kansas DOT collects the majority of the MIRE FDEs (except intersection information) such as • Dedicated turn-lanes, • Turn-lane restrictions, and, • Intersection control details. The Kansas DOT reported that it would begin to manage MIRE FDEs within the LRS and initiate distribution to different agencies. In addition to the ongoing K-Hub project previously described in Chapter 3, the Kansas DOT is working on the Next Generation 911 (NG911) project to create a GIS-based call-locating system to be used by emergency services (see Appendix E). The Kansas DOT has incorporated the GIS data model and the road centerline geometry from the NG911 project to use as the basis for the LRS routes it will develop. The state’s GIS clearing- house, the Data Access and Support Center, maintains a portal website where data are submitted electronically. The Kansas DOT has access to the portal, as well as to a database connection to current road centerline data for NG911. The Kansas DOT receives quarterly letters from local agencies on any listed changes, which are then provided to the emergency services system. The Kansas DOT stated that the lessons learned in safety data integration and in working with the local agencies included the following: • It is a best practice to ask LPAs for information rather than to send someone from the state out to collect the data themselves. • The NG911 data model and data standards have been well-received by LPAs, and the standards have been effective in providing guidance and expectations. The standards were designed to accomplish the purpose of GIS-based call routing, but incorporation of additional standards such as road centerlines meeting the ARNOLD standard could be viewed as a barrier or an unfunded mandate. Tennessee In the state of Tennessee, the Road Inventory Office and Roadway Data Office are responsible for collection of MIRE FDE roadway data. The data are stored in the Enhanced Tennessee Roadway Information Management System (E-TRIMS), which is a map-centric, web-based

Case Examples of Data Integration and Maintenance Efforts 65 reporting tool used for extracting roadway inventory and crash and project data according to specific criteria. E-TRIMS is updated daily and is used by consultants, contractors, MPOs, RPOs, and LPAs. It is also used internally by the Tennessee DOT, primarily by the planning, engineer- ing, structures, and maintenance divisions. Access to E-TRIMS is granted by the Tennessee DOT after an entity completes an application form and the Acceptable Use Policy Network Access Rights and Obligations User Agreement Acknowledgement form (included in Appendix D). E-TRIMS can export map data to Excel, ArcMap, or Google Earth. Currently the Tennessee DOT receives roadway attributes from LPAs when there is a roadway name change. Tribal Technical Assistance Program An interview with the former Director of the Tribal Technical Assistance Program revealed that efforts related to MIRE FDEs have been under way in the state of North Dakota, where the Three Affiliated Tribes are currently in the initial phase of using the same crash and citation reporting system (Traffic and Criminal Software) as the North Dakota DOT, the North Dakota Highway Patrol, and several counties. The data are being captured in the North Dakota DOT server. Wisconsin The Wisconsin DOT uses the state trunk network and WISLR databases to store and manage its MIRE FDEs. Most of the MIRE FDEs are contained in both databases, with the exception of Interchange Type. The Wisconsin DOT is currently conducting a feasibility analysis and building the criteria that will guide the analysis. The agency reported that one primary factor will be analyzing the compatibility of the different data models. ARNOLD Phase II, which is expected to be completed by mid-July 2018, will produce further information with respect to WISLR and the state trunk network as well as possibly integrate these two systems. Summary The interviewees from the nine focus states consistently made the following observations: • A close relationship between the state DOT and LPA staff is a key factor in facilitating a successful and coordinated data collection and maintenance effort. • A concentrated and consistent approach on the part of the state to demonstrate the mutual benefits of data integration to the LPAs facilitates the collection and provision of MIRE FDEs. • The advancements in technology, especially with respect to GIS, have enabled both easier data collection and better-coordinated QA efforts. • Consistency in naming common features and attributes (perhaps through the use of a data dictionary) is necessary to avoid different interpretations by different stakeholders. • There is a benefit in engaging academic partners, such as state University Transportation Centers, in coordinating and collaborating with state DOTs and local governments to facili- tate and manage data collection (e.g., the examples from Iowa and Michigan). Table 21 describes the current status of MIRE FDE collection on the basis of responses from 12 of the LPAs that provided information in interviews. In the Roadway Segment for Non-Local Paved Roads, the route number, street name, surface type, and type of governmental ownership are all data elements collected by nearly all of the agencies interviewed as case examples for this synthesis. Elements that require more resources, such as AADT, were reported to be collected only by approximately half of the agencies interviewed.

66 Integration of Roadway Safety Data from State and Local Sources Agencies Collecting MIRE FDEs MIRE FDE Rank Total Number Interviewed Percent Roadway Segment for Non-Local Paved Roads Route/Street Name 1 10 83.33 Surface Type 2 9 75.00 Type of Governmental Ownership 2 9 75.00 Route Number 4 8 66.67 Segment Identifier 5 7 58.33 Segment Length 5 7 58.33 Intersection for Non-Local Paved Roads Location Identifier for Road 1 Crossing Point 1 4 33.33 Location Identifier for Road 2 Crossing Point 1 4 33.33 Intersection/Junction Geometry 1 4 33.33 Interchange/Ramps for Non-Local Paved Roads Ramp Length 1 5 41.67 Unique Interchange Identifier 2 3 25.00 Location Identifier for Roadway at Beginning Ramp Terminal 2 3 25.00 Location Identifier for Roadway at Ending Ramp Terminal 2 3 25.00 Type of Governmental Ownership 2 3 25.00 Local Paved Road Type of Governmental Ownership 1 9 75.00 Surface Type 2 9 75.00 Segment Identifier 2 7 58.33 Unpaved Road Type of Governmental Ownership 1 8 66.67 Segment Identifier 2 4 33.33 Table 21. Status of MIRE FDE collection summarized from 12 LPA interviews.