By Sultan Ali, Contributing Author
Several highway construction projects involve the application of smart work zone (SWZ) systems to manage traffic. The potential benefits of using the SWZ systems include mobility and safety improvements.
Therefore, it is critical for agencies to estimate the benefits of SWZ systems before deployment and during operations.
Estimating the mobility benefits of the SWZ is relatively straightforward due to the availability of real-time traffic data, including speed, travel time, volume and occupancy. In contrast, estimating the safety benefits has several challenges.
The major challenge being the use of crashes, the widely accepted measures of safety, as the performance measure.
It is challenging to use crashes in an evaluation because the SWZ are active during specific time-of-the-day when there is work going on. Also, the work zones have a specific duration from the beginning of the project to the end, typically a few years.
The small-time window for a work zone and its activities could limit collecting enough crash data to evaluate the SWZ systems. Moreover, using crashes as a performance measure does not allow real-time monitoring of the SWZ systems to help make the required adjustments.
The presence of voluminous and variety of real-time data allows the use of surrogate safety measures (SSMs) to assess the safety effects of SWZ systems.
SSMs offer a viable alternative for safety analysis in work zones, leveraging traffic conflict/near misses analysis. Therefore, comprehending work zone safety indicators through surrogate measures is essential for agencies to effectively monitor real-time safety within these zones.
Traffic data collected within the work zone of the Interstate 395/State Route 836/Interstate 95 design-build Miami Signature Bridge project in Florida served as the basis for demonstrating the application of SSMs in assessing safety implications for road users and workers.
Utilizing unmanned aerial vehicles, commonly known as drones, video traffic data were gathered. Subsequently, the captured video footage underwent processing using video analytics enhanced by Artificial Intelligence (AI) technology, yielding heat maps and SSM performance indices values.
Several performance indicators were considered in the analysis, including Time-to-Collision (TTC), Post Encroachment Time (PET) and Deceleration Rate.
TTC represents the temporal distance from the collision point assuming constant velocity, while PET signifies the time difference between a vehicle leaving the area of encroachment and a conflicting vehicle entering the same area. Deceleration Rate measures the rate at which a following vehicle must decelerate to avoid a collision with the leading vehicle.
These insights, combined with engineering judgment, guided the implementation of traffic safety countermeasures as necessary.
Following the deployment of traffic safety countermeasures, the video footage was once again analyzed to evaluate any improvements in heat maps and performance indices of the SSMs.
This proactive approach aims to preemptively prevent traffic crashes, rather than relying solely on post-incident analysis to apply countermeasure strategies.
Utilizing video analytics powered by AI technology, the collected data underwent thorough processing and analysis.
The methodology employed in this study offers a means for agencies to actively monitor the safety of work zones in real-time, facilitating the deployment or adjustment of various countermeasures as needed.
This methodology can serve as valuable support for agencies when making decisions regarding the implementation of SWZ systems in their projects.
Furthermore, the methodology holds promise for estimating the benefits of SWZ systems in other projects, thereby contributing to the pursuit of Federal Highway Administration (FHWA) Vision Zero objective.
The application of drones and AI represents a proactive approach to enhancing traffic operations and safety. By preemptively addressing potential safety concerns, rather than waiting for incidents to occur, agencies can effectively mitigate risks and prevent crashes.
Drones and AI have the capacity to significantly augment work zone traffic management, offering a promising avenue for improving overall road safety.
Sultan Ali is a Transportation Systems Management and Operations engineer at CHA Consulting, Inc.