Implementing Information Findability Improvements in State Transportation Agencies (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.

14 This chapter provides guidance for improving findability from five different perspectives: • Agency champions – knowledge management lead, chief information officer, chief informa- tion architect, chief data officer, communications office director, learning and development director, and strategic or organizational performance lead • Information managers – librarian, webmaster, content management system manager, records manager, GIS manager, and data manager • Information creators – planners, analysts, engineers, administrators, and anyone else who produces information during the course of their job or receives information from external sources (e.g., contractors) • Information seekers – all employees, external partners, and the general public • Information technology staff This guidance supplements the material presented in NCHRP Research Report 846 (1). 4.1 Agency Champions 1. Advocate for resources to support improved information management, including but not limited to shared content repositories; search tools; and staff with expertise in taxonomy development, metadata management, machine learning, and data science. 2. Sponsor an effort to identify priority search/discovery needs – who needs what information and why. Use what is learned through this effort to prioritize and advance findability initia- tives based on business value. 3. Undertake pilot projects to learn about use cases and opportunity to add value. Use the results of these pilots to create a business case for enterprise search systems and supporting meta- data and processes. (See Figure 5 for sample DOT use cases.) 4. Create a list of priority content types that should be searchable across the agency. Examples might include maintenance agreements, project design plans/as-builts, deeds, environmental documents, contracts, and design exceptions. Establish a taxonomy or controlled list of content types and get agreement to use it to tag documents. 5. Establish a team to build and maintain agency vocabulary resources such as glossaries and thesauri. Link this effort to ongoing efforts to catalog data. 6. Sponsor an effort to create an onboarding package for new employees so they understand what information is available, where to store different types of content, how to search, and other best practices for managing their information. 7. Sponsor an effort to develop a metadata architecture including workflows for metadata assignment and integration/operability across systems. 8. Educate staff about available text analytics tools and capabilities for automating document classification, extracting metadata, cleaning up duplicate or near-duplicate files, and con- ducting sentiment analysis of social media related to the agency. C H A P T E R 4 Guidance for Improving Findability

Guidance for Improving Findability 15 9. Identify areas where text analytics applications could add value and undertake pilot efforts. If these prove successful, consider installing a commercial text analytics platform. Identify and deploy search-based applications that add value to the organization—for example, an expertise locator, a project information search engine, tools supporting Freedom of Infor- mation Act (FOIA)/public disclosure-related searches, tools supporting identification of duplicative and outdated content, and tools supporting identification of personally identi- fying information (PII). 4.2 Information Managers 1. Make an effort to ensure that documents are text readable; encourage the use of available optical character recognition (OCR) tools prior to document ingestion, or establish bulk OCR processes. Establish special procedures for documents that are password-protected if there is a need to make these searchable (but not necessarily accessible by those without permissions); these documents may not be processable by all OCR conversion programs. 2. Assign responsibility for search monitoring and tuning. Use results of search logs to under- stand the most common searches and tune the search engine to make these more successful. 3. Scope a project to identify “personas” representing different types of information users and characterize their information needs and information seeking behaviors. Use this informa- tion to improve the design of website navigation and search features. 4. Ask target end users to describe specific situations where they needed information and how they went about finding it. Perform structured testing and refinement of navigation and search tools based on these user stories. 5. Obtain feedback from target end users about available navigation and search tools: what they like and what they wish worked better. 6. Collect stories from users about how they benefit from improved search and navigation capabilities (e.g., time savings and improved effectiveness). Document these stories and use them to advocate for needed resources to maintain and enhance capabilities. 7. Use a standard template for reporting errors to be fixed by information technology staff so that they understand the current and expected behavior of the system. Construction Project Scoping: Find old concept reports/scoping studies, design or as- built design plans, right of way plans, core samples, utility location plans (presence of underground utilities), environmental commitments (e.g., for noise walls or wetlands), traffic analyses, and agreements. Corridor Study Information Gathering: Find any documents related to the corridor (as defined by route, from-to) including datasets, traffic studies, safety studies, environmental studies, agreements, concept reports/scoping studies for projects in the corridor, design exceptions, old plan sets, and prior corridor studies. Responding to Public Records Requests: Locate data and documents in response to a wide variety of requests, including email related to a project or incident, accident reports, as-built plans, environmental reports, traffic control plans, signal timing, and old concept reports for a project. Locating an Old Agreement or Commitment Impacting a Current Project: For example, find an old railroad agreement that restricts the DOT’s ability to remove an old railroad bridge. Environmental Analysis: Locate prior cultural resource or wetlands reports produced within the area of interest; find a GIS wetlands delineation file. Stormwater Analysis: When mapping out the drainage system as part of a stormwater process, identify which components exist by agreement. Find agreements to determine actions needed to meet water quality standards. Figure 5. Sample DOT use cases for search and discovery.

16 Implementing Information Findability Improvements in State Transportation Agencies 4.3 Information Creators 1. Develop an understanding of how your final work products are used by others. 2. Store your final work products in officially designated locations (if specified) or in available locations so they can be discovered and used appropriately should you change jobs. 3. Develop and use standard document templates for deliverables so that information included in a specific content type is consistently presented in the same format. This will make docu- ments easier to use, classify, and find. 4. Review existing agency retention schedules pertaining to your area of business. If you feel that the current retention schedules do not meet business needs, propose revised schedules to the individual in your unit responsible for updating these schedules. 5. Clean up your files on a regular basis (e.g., daily or weekly) to delete duplicate files or older versions that do not need to be retained. 6. At the end of each project or task, designate the final files to be kept. Review and update documentation about these files so that people using them in the future will understand how they were produced and how to use them. 7. Before storing or posting a file for others to access, make sure that it is text readable. Apply OCR processing to make it text readable if it is not. After applying OCR, verify that the drawings and graphics in the document are presented accurately; specifically, make sure that the graphic is not changed by the OCR process. 8. Before storing or posting a file for others to access, make sure it is free of revision markings. 9. Before storing or posting a file for others to access, remove password protections that might interfere with the ability of others to search for the file. 10. Before storing or posting a file for others to access, make sure the document creator is clearly specified and (where applicable) a list of team members is included. 11. Document your files; add and check the accuracy of metadata in the file properties and/or the system where the document lives (date, author, title, status, content type, and others), following applicable agency standards. 4.4 Information Seekers 1. Familiarize yourself with the different information sources in your agency. 2. Review available instructions for constructing efficient search queries that will return the best result set. 3. Improve your search skills. Become familiar with how to formulate an effective search query, taking advantage of available advanced search features, including use of quotes to find an exact match and use of Boolean operators such as AND and OR. 4. Provide feedback to managers of search tools if you are having trouble finding content or if your searches aren’t performing well. Be specific about your search query, the results you expected, and the results you actually obtained. 4.5 Information Technology Staff 1. Develop the capability in the IT function to support the technology stack used to store, classify, and retrieve content. 2. Develop the capability to correct search related problems when users or information managers identify bugs and misclassified content or identify searches that don’t yield correct results. 3. Identify and facilitate implementation of technical solutions to metadata management and integration across systems.