2. Chapter 2 – Research Approach

The research approach consisted of two phases. Phase I consisted of the data-gathering tasks to understand the current state of practice, identify gaps, and document best practices. Phase II consisted of methodology and guide development, as illustrated in the research workflow diagram in Figure 1.

2.1. Phase I – Data Gathering and State of Practice Analysis

Phase I consisted of Tasks 1 through 5. To accomplish the Phase I objectives, the research team conducted a targeted desktop scan to understand the current practice and identify notable agencies to interview. The data-gathering effort was to understand the current practices, organizational requirements and constraints, opportunities, and knowledge gaps related to business-driven data representations and the development and use of data ontologies for data-driven decision-making in State DOTs. Interview requests in the form of questionnaires were sent to 18 State DOTs for their written responses. Nine State DOTs responded to the questionnaire. Follow-up phone interviews were conducted with the responding agencies to clarify some information.

2.1.1. Task 1. Kickoff Teleconference

To ensure the research team and the panel were aligned, we held a meeting to kick off the research. During this meeting, we discussed the comments received by the panel during the selection process. The research presented the Amplified Work Plan and expected deliverables.

2.1.2. Task 2. Literature Review

The literature review aimed to identify strategies for DOTs to implement data ontologies that support agile and efficient data-driven decision-making. It was

conducted via a desktop scan, targeting both published and unpublished materials, including journal articles, research reports, and academic papers relevant to the subject. Keywords and phrases, such as data ontologies, legacy systems, data integration, State DOTs, data management, and data representation, were used to identify relevant material.

2.1.3. Task 3. Targeted Surveys and In-Depth Interviews

Following the literature review and additional input from the NCHRP Panel, the research team conducted targeted surveys and in-depth interviews. These activities aimed to examine how DOTs implement data ontologies to maximize the use of legacy data. The secondary objective of this task was to identify notable practices for case studies and use them as best practices in the guide. A targeted survey was chosen because data ontology is still relatively new among State DOTs, and surveying all 50 states would not be an efficient use of resources.

The approach to accomplishing the stated objectives included:

Step 1: Identify Survey Participants

Through the literature review, we identified specific DOTs to be surveyed. The panel members also recommended to the research team some viable agencies with notable practices to contact and survey. In all, the team identified 18 agencies to survey.

Step 2: Develop a Survey and Interview Guide

The research team created the survey questionnaire and a cover letter for NCHRP’s review. The questionnaire was structured to align with the essential elements of ontology development, including processes and procedures, tools, and people. The interaction among these elements was examined holistically to inform the development of the guide and case study in the next step.

The cover letter outlined the purpose of the survey, key deadlines, and standard definitions to ensure that survey participants could respond with a common understanding. After the NCHRP panel approved the survey questions and cover letter, the research team distributed the survey via email to the 18 state DOTs.

Step 3: Gather and Analyze Survey Responses

The survey was conducted over two months. We received responses from nine state DOTs (Arkansas, Florida, Indiana, Iowa, Minnesota, Texas, Utah, Vermont, and Washington). The respondents included data owners, IT staff, and asset managers. The research team gathered and analyzed responses. The respondents’ experience with data ontologies varied significantly, encompassing everyone from beginners to recognized members of the transportation research board ontology community.

Step 4: Conduct In-depth Interviews

Driven by the survey responses, follow-up interviews were conducted between the research team and agency representatives via Microsoft Teams. The follow-up interviews were used to clarify responses from the survey and to request additional information where necessary, with the goal of enhancing the quality and completeness of the data collected.

Step 5: Identify a List of Agencies for Case Studies and Use Cases

The research team developed a list of criteria for selecting agencies for the case study development. The following criteria were developed in selecting the case studies:

• The agency responded to the survey questionnaire.
• The agency provided at least one example of legacy data system migration.
• The agency expressed a desire to advance ontology development.
• The agency demonstrated the use of one or more data ontology tools.
• Agencies represent a reasonable range in terms of geographic location, size, and data maturity.

2.1.4. Task 4. Synthesis of Task 2 and Task 3 Results

This task involved synthesizing the results of the literature review, targeted surveys, and interviews. The synthesized report was submitted to the NCHRP panel for review before the virtual panel meeting.

2.1.5. Task 5. Virtual Panel Meeting

Following a one-month panel review of the synthesis report, the research team met virtually with the NCHRP panel to discuss the panel’s comments. During the panel meeting, we sought the panel’s authorization to commence Phase II of the research. The research team prepared the meeting materials, including an agenda, presentation slides, and meeting notes. At this meeting, the panel voted to authorize the execution of Phase II of the research.

2.2. Phase II

Phase II consisted of Tasks 6 through 11. The researchers used the information gathered from Phase I to develop case studies, a conceptual framework comprising four strategies, and supporting actions. These serve as roadmap and guide for developing and using data ontologies.

2.2.1. Task 6. Develop Case Studies and Use Cases

This task involved developing case studies and use cases based on the targeted surveys, in-depth interviews with selected DOTs, and a comprehensive literature review. The purpose of the case studies was to showcase industry practices, techniques, and methodologies that facilitate the development of data ontologies and provide examples of successfully implemented strategies.

2.2.2. Task 7. Methodology for Implementing Data Ontologies

The objective of this task was to use the information gathered from the prior tasks to develop a conceptual framework with actionable steps, serving as a roadmap for developing and using data ontologies to support data-driven decision-making. The associated methodology included competency questions to facilitate a straightforward, user-centered approach to ontology development, thereby aligning the methodology with user needs. The methodology also uses practical examples, where applicable, to illustrate implementation tips. The draft methodology was issued to the NCHRP panel for review before the second virtual panel meeting in Task 9.

2.2.3. Task 8. Guide Outline

This task involved developing the outline of the guide. The conceptual framework and methodology developed in task 7 drove the guide outline. The draft guide outline was issued to the NCHRP panel for review prior to the second virtual panel meeting in Task 9.

2.2.4. Task 9. Virtual Panel Meeting

Following a one-month review of the draft methodology and outline, we organized a virtual panel meeting to discuss the panel’s initial comments, solicit additional feedback, and seek the panel’s authorization to develop the research products. The research team prepared the meeting materials, including an agenda, presentation slides, and meeting notes.

During this meeting, the research team facilitated interactive activities to verify the draft methodology’s readability and usability. We also discussed the annotated guide outline to ensure it has the essential elements to address users’ needs. At this meeting, the panel voted to authorize the research team to proceed with the research as planned.

2.2.5. Task 10. Develop Draft and Final Guide

The objective of this task was to develop the draft and final guide, incorporating the findings and feedback from the NCHRP panel members. The draft guide was issued to the panel for review and feedback, and the research team incorporated the comments to develop the final guide.

2.2.6. Task 11. Develop Final Deliverables

The objective of this task was to develop materials that would help market the research and facilitate the transition from research to implementation of the final products. The final deliverables included this Research Report and the following:

• Standalone, customizable communication material—The communication material, PowerPoint Slides, contains the research outcome. Users can customize this to communicate at all levels of management and build awareness of the research products. It can be tailored for management buy-in and engagement, as well as for training technical staff on ontology concepts or how to build data ontology.
• A standalone technical memorandum on implementation—This technical memo defines the research products’ target users, identifies avenues and activities for disseminating them, and suggests additional activities for implementing the research.