CHAPTER 3

Survey of Agencies

A survey was conducted to gather information about the automated processes that IOOs have implemented or are planning to implement. The main purpose of the survey was to identify the applications that agencies have piloted or implemented.

A draft survey was developed and then updated based on comments from the NCHRP Project 14-42 panel. An electronic version of the survey was created and then beta-tested and updated several times to ensure that it was understandable, responses could be coded appropriately, and the survey could be completed in a short amount of time (i.e., 10 to 15 minutes).

The survey was sent to all members of the following committees:

1. AASHTO Committee on Maintenance,
2. AASHTO Committee on Transportation System Operations,
3. AASHTO Committee on Traffic Engineering, and
4. AASHTO Subcommittee on Asset Management.

Additionally, several city associations were identified, particularly those whose members have implemented Smart City applications.

The survey was sent in early March 2023. Reminders were sent 1 week and then 2 weeks after the initial survey was sent. The survey was closed 1 week after the final reminder. A total of 123 individuals started, partially completed, or completed the survey. Responses where a survey was started but no questions were answered were deleted. This resulted in 56 usable survey responses. In some cases, respondents answered one or two questions but not others. These responses were retained, and the metrics for each question were based on the total number of responses obtained for that question. In order to encourage agencies to respond, respondents were not required to provide contact information. Among those respondents who provided contact information, the 29 states shown in Figure 3-1 were represented. As noted in Section 2.3, this group included 29 state DOTs and two cities. As noted, some agencies did not include their affiliation. The total number of responses used was 34. The actual survey is provided in the appendix. Not all agencies answered every question. As a result, the responses were tabulated by the number of agencies that answered for each question.

In several cases, more than one participant within an agency responded. Survey responses were combined for each agency, resulting in one response per agency. As part of this process, conflicting responses were reconciled. For instance, if one participant indicated that UAVs are used for bridge inspection and another participant from that agency indicated that UAVs are not used for bridge inspection, the response was coded as “using UAV for bridge inspection.”

In this case, it was likely that the second respondent was not aware that another group at the agency was using UAVs for bridge inspection. Respondents were not required to provide their contact information. As a result, if a response was not associated with an agency, it was not possible to flag it as one of multiple responses for a given agency. In this case, the responses were treated as if they were from separate agencies.

Respondents were also asked if the team could follow up for more information.

A summary of survey responses is provided in the sections that follow. Unless otherwise noted, a response was received from each of the states noted in Figure 3-1. Otherwise, the sample size is noted in the accompanying section. The questionnaire was structured in such a way as to differentiate between processes that agencies are currently utilizing, processes that have been evaluated but not necessarily implemented, and processes that are planned for implementation in the near future (within the next 3 to 5 years). Questions were placed in the following three categories, each containing principal questions:

• Automated processes used by agencies, excluding UAVs;
• Use of UAVs for various processes; and
• Processes that would benefit the most from automation.

3.1 Automated Processes

This section covered information about the automated processes the agency used. The principal question was as follows:

What automated processes does your agency currently use? Check all that apply.

The following processes were provided as choices:

• PAVEMENT: Pavement data collection.
• PAVEMENT: Pavement repair (i.e., pothole).
• MAINTENANCE: Autonomous/remote mower or vegetation control.
• MAINTENANCE: Snowplow.
• MAINTENANCE: Other maintenance activities.
• TRANSIT: Autonomous shuttles.
• TRANSIT: Autonomous buses.
• CONSTRUCTION & SAFETY: Crash abatement vehicles for work zones.
• CONSTRUCTION & SAFETY: Automated lane closures for work zones.
• CONSTRUCTION & SAFETY: Construction monitoring.
• CONSTRUCTION & SAFETY: Smart arrow boards.
• MUNICIPAL: Street cleaning.
• MUNICIPAL: Garbage collection.
• Other, please describe below.

Responses related to pavement applications are shown in Figure 3-2. As the figure shows, 68% (n = 21) of responding agencies are currently using some type of automated processes for pavement data collection, with 13% (n = 4) planning to use or evaluate automated processes. Around 19% (n = 6) are not using or planning to use automated processes. Figure 3-2 also shows agencies’ use of automated processes for pavement repair. As the figure shows, 9% (n = 3) are using some automated processes for pavement repairs, such as fixing potholes, with 25% (n = 8) indicating that they have some plans for its use. The majority (66%) (n = 21) were not using or had no plans to use automation for pavement repair.

Figure 3-3 shows responses for automated maintenance activities. As the figure shows, 18% (n = 6) of agencies that responded use some type of automation for mowing and vegetation control, with 35% (n = 12) indicating that they are planning to use automation in the next 3 to 5 years. Almost half have no plans to use automation.

Figure 3-3 also shows agencies’ use of automation for snowplow operations, with 18% (n = 6) responding that they use some type of automation and 15% (n = 5) planning to use or evaluate automation. The remainder (68%, n = 23) have no plans to utilize automation. Agencies were also asked about other maintenance, with 15% (n = 5) indicating that they had used some type of automation and 26% (n = 9) indicating that they are planning to use automation. Finally, 59% (n = 20) have no plans to use automation. Respondents were not given an option to indicate what those activities entailed.

Agency responses for the use of autonomous transit are shown in Figure 3-4. As the figure shows, 16% (n = 5) of agencies that responded were using some type of autonomous shuttle. None were using autonomous buses. Another 75% (n = 24) indicated that they were planning

to use or evaluate the use of autonomous shuttles in the next 3 to 5 years, and 84% (n = 27) had plans to use or evaluate the use of some type of autonomous transit bus. Nine percent of agencies (n = 3) indicated that they were not using or planning to use autonomous shuttles, and 16% (n = 5) indicated that they were not using or planning to use autonomous buses.

Agency responses for the use of automation in work zone or safety applications are provided in Figure 3-5. The most common automated construction process was the use of smart arrow boards, with 45% (n = 14) of respondents noting that they use this technology and 23% (n = 7) indicating that they plan to use or evaluate the use of this technology in the next 3 to 5 years. The use of automation for construction monitoring was the next most commonly used application (23%, n = 7) with 30% (n = 9) of respondents indicating that they plan to use this technology. Around 13% (n = 4) of respondents noted that they were using crash abatement vehicles in work zones, and 43% (n = 13) were planning to use this technology. Around 13% (n = 4) of respondents used automated work zone lane closures, with 16% (n = 5) planning to use or evaluate the use of this technology in the near future.

Agency responses for automated street cleaning and garbage collection are provided in Figure 3-6. As the figure shows, most agencies are not using these technologies, (93%, n = 26 for street cleaning and 96%, n = 27 for garbage collection), with a small number saying that they may use or evaluate these technologies in the next 3 to 5 years (7%, n = 2 for street cleaning and 4%, n = 1) for garbage collection. It should be noted that these two activities are not primarily the duties of a DOT. They were included in the survey with the expectation that more responses would be obtained from cities.

3.2 Use of UAVs for Automated Processes

The next question was about the agency’s use of UAVs:

How is your agency using unmanned aerial vehicles (UAVs)? They are also called unmanned aircraft, drones, etc. Check all that apply.

The use of UAVs for traffic operations is shown in Figure 3-7. Around 9% (n = 3) of respondents are using UAVs for traffic monitoring, and 27% (n = 9) have piloted the technology while 45% (n = 15) are not using the technology. Another 18% (n = 6) plan to use or pilot the technology in the next 3 to 5 years. Twenty-eight percent (n = 9) of agencies indicated that they have used UAVs for traffic incident detection or emergency response, and 13% (n = 4) have piloted the technology for this application. An additional 13% (n = 4) plan to use or pilot UAVs for traffic incident detection/emergency response and 47% indicated they are not using the technology (n = 15). Around 7% (n = 2) of respondents have used UAVs for traffic enforcement, and 3% (n = 1) have piloted the technology. Ten percent (n = 3) are considering using the technology for traffic enforcement in the near future and 80% are not using the technology (n = 24).

The use of UAVs for infrastructure inspection activities is shown in Figure 3-8. The most common application used by agencies is bridge inspection, with 58% (n = 18) of respondents having used and 27% (n = 9) having piloted the technology, and with another 3% (n = 1) likely to evaluate the use of the technology in the next 3 to 5 years with 12% indicating they did not use the technology (n = 4). Around one-third of respondents (n = 10) have utilized UAVs for construction inspection, with 23% (n = 7) having piloted the technology. Around 13% (n = 4) of respondents noted that they plan to evaluate the use of UAVs for construction inspection in the next 3 to 5 years and 30% are not using (n = 9). A smaller number of agencies (14%, n = 4) have used UAVs for geotechnical asset inspection, 24% (n = 7) have piloted, 7% plan to use (n = 2), and 55% (n = 16) are not using UAVs for that application.

Figure 3-9 illustrates agency responses about the use of UAVs to monitor other situations. As the figure shows, 42% (n = 14) of respondents are using UAVs to monitor emergency conditions,

and 15% (n = 5) have piloted the use of UAVs for this application with 18% (n = 6) planning to use and 24% (n = 8) not using. Almost 40% (n = 12) of respondents have used UAVs to monitor assets, 6% (n = 2) have piloted the use of UAVs for this application with 23% planning to use (n = 7) and 32% not using (n = 10). For this response option, the question used “assets” as a general category to include assets not already listed under other response options for this question, such as bridge inspection, pavement condition evaluation, and so on. Finally, 7% (n = 2) of respondents are using UAVs to monitor pavement condition, 7% (n = 2) have piloted the use of UAVs for this application while 13% (n = 4) plan to use and 73% are not using (n = 22).

Overall, the most common use of UAVs by agencies that responded was for bridge inspection (58%, n = 19), followed by emergency condition assessment (42%, n = 14). The most common UAV applications that have been piloted are traffic monitoring (27%, n = 9) and bridge inspection (27%, n = 9), followed by monitoring geotechnical assets (24%, n = 7). Finally, the most common UAV application that agencies plan to implement, or pilot is monitoring assets in general (23%, n = 7) followed by traffic monitoring (18%, n = 6) and emergency condition monitoring (18%, n = 8). Agencies were least likely to have used UAVs for traffic enforcement (80%, n = 24) or pavement condition assessment (73%, n = 22).

3.3 Processes that Would Benefit the Most from Automation

The last question asked agencies about the processes that they thought would benefit the most from automation:

What processes do you think would benefit the most from automation? Check all that apply.

The response options were “Yes” or “No,” and respondents were able to select a separate response for each application. As a result, agencies could indicate that they believed that multiple different applications could benefit from automation. The results for applications that agencies indicated that they believed would benefit the most from automation are shown in Figure 3-10. Agencies may have considered the use of UAVs in their responses, but the question was not specifically about the use of UAVs for any particular application. Additionally, each question asked whether the respondent thought the application would benefit (Yes or No). A respondent could have provided a response for one application but not another.

As shown in Figure 3-10, 100% of respondents (n = 30) indicated that traffic incident/emergency response activities would benefit from automation. This was followed by emergency condition assessment (93%, n = 28) and monitoring assets (93%, n = 27). As noted above, monitoring assets was a broad response option, and agencies may have considered assets such as bridges in their responses even though other categories listed specific assets. Bridge inspection and pavement data collection were the next most common processes that agencies thought would benefit from automation (90%, n = 28 and 83%, n = 24, respectively). Agencies were least likely to indicate pavement repair as likely to benefit from automation (46%, n = 13), followed by snowplow operations (52%, n = 15).