CHAPTER 3

Application of Emerging Technologies in SWZ Use Cases

This guide is focused on emerging SWZ technologies that had a maturity rating of 1 or less (≤1) and SWZ technologies suggested by stakeholders who participated in workshops held as part of the development of the guide. Two technologies were chosen on the basis of each use case. These technologies are discussed in this chapter.

Traveler Information Systems

Connected and Automated Vehicles

Connected vehicle technologies have been used to disseminate information about work zones to approaching drivers. In Wyoming, a connected vehicle pilot program used traveler information messages and showed that they reduced deceleration rates and lowered crash risk (Yang et al., 2020). Further, Misra et al. (2018) found that vehicle-to-infrastructure (V2I) and infrastructure-to-vehicle (I2V) communication could be more effective than vehicle-to-vehicle (V2V) communication because it allowed state DOTs to send customized in-vehicle messages to drivers. Finally, a driver smart assistance system alerted drivers to an approaching work zone by using radio frequency identification to send alerts in auditory as well as visual modalities, which resulted in reduced speeds and helped by prompting vehicles to decelerate sooner (Qiao et al. 2014). Most of the CAV deployments for traveler information systems are in experimental or pilot testing stages.

Crowdsourced Data/Smartphone Applications

Some of these technologies—email alerts, text messages, social media, applications (e.g., Waze), websites, and media alerts—can also be used to both gather and provide information to travelers about work zones. In general, these technologies make it easy to disseminate information and are low-cost solutions. These systems are commercially available and well understood by state DOTs. However, they are also limited by access and availability of infrastructure to send the data. For example, a person may need to be subscribed to an email alert or text message option to receive the information. In the case of a website, a person needs to know about the website, have the app installed on his or her cellphone, and have Internet access to be able to access traveler information.

Queue Warning Systems

Connected and Automated Vehicles

CAV-based queue warning systems have been proposed in the literature, but few field evaluations exist. In CAV-based queue warning systems, the vehicles themselves act as probes on the traffic stream. The speeds from equipped vehicles are then used to detect the existence of a

traffic queue or significant speed differences. A CAV-based queue warning can be executed either by V2V communications between vehicles in the vicinity or by reporting speeds to an IOO, which then creates and disseminates warnings via V2I communications. To date, V2I systems that have relied on CAVs to generate the data for the queue warning system have only been examined via simulation, as the market penetration rate for CAVs has been insufficient for field testing in which the vehicles serve as the data collection tool. Although most V2V systems have been evaluated with simulation analysis, a few field evaluations of V2V-based systems, in the form of rear-end collision warning systems, have been identified (Khazraeian et al. 2017; Stephens et al. 2015; Zhao et al. 2019).

CAV-based systems are still in an experimental stage. The CAV market penetration rate remains low, and the required penetration to support this application remains unclear. As a result, the ability to directly use CAVs to generate data for queue warning is still limited. While using CAVs as a data source for queue warning is currently experimental, it has a great deal of promise as market penetration increases, because data generation would not be tied to specific physical locations where sensors are deployed. Khazraeian et al. (2017) found that a 3% to 6% market penetration rate would yield accurate queue warnings and that the CAV-based estimation of the back of the queue on a congested freeway can provide more accurate location than that based on detectors. The Texas A&M Transportation Institute conducted a demonstration with 21 vehicles equipped with connected vehicle systems that showed that queue warning using CAV technology can locate the end of the queue more accurately than loop detectors with a 1-mile spacing (Stephens et al. 2015). CAV technology could provide a long-term robust way for vehicles approaching a work zone to identify traffic queues.

Crowdsourced Data/Smartphone Applications

Queue warning functions in smartphone applications are rarely stand-alone applications. Instead, they are usually part of navigation software that provides broader traveler information and routing functions. Navigation applications on smartphones are a well-developed, commercially available technology with multiple vendors. The smartphone user can install the navigation software directly from the app marketplace, usually at no cost, but not every user may have the software installed or be actively using it when approaching a work zone. The queue warning function of navigation software is available not only at the pre-trip-planning stage, but also throughout the trip if an Internet connection is available. Being a data element within navigation software, the queue warning function integrated with the navigation software can easily disseminate comprehensive information from local work zone conditions, alternative routing, and suggested departure times. However, no explicit studies of the effectiveness of these systems on preventing end-of-queue crashes could be located. Because this dissemination method requires no deployment of physical devices on the road, the queue warning on smartphone applications works under all traffic conditions at any sites, given that an Internet connection is available. It has been observed that drivers trust navigation software on their mobile devices, and as drivers often follow the detour suggested by a navigation app, significant shifts in traffic patterns may be observed.

Work Zone Intrusion Alert Systems

Smart Vest

Smart Vest systems can identify intrusions and track vehicles, equipment, and personnel as well as consider potential collisions on the basis of trajectory (Roofigari-Esfahan et al. 2021). The Smart Vest is a very new piece of technology, and its evaluation is still ongoing (maturity rating = 1).

The vest employs an algorithm to communicate potential collisions to workers, passing drivers, and CAVs. Human–machine interface (HMI) modes utilize flashing lights for visual signaling, vibrations for haptic signaling, and speakers for auditory signaling. A receiver provides real-time kinematic (RTK) GPS data, and a communication module transmits between the vest and the gateway that houses the algorithm. This module sends algorithm-based RTK GPS data and HMI requests to activate the sensors on the vest. The gateway also interprets work zone boundaries through a polygon plot. The vest features a low-level warning when it is within 1 meter of a work zone boundary and a high-level warning when the boundary has been reached or crossed. Field testing was conducted in Elliston, VA, at an active construction site, where the communication range was verified to be operational in a 500-meter-long zone (Roofigari-Esfahan et al. 2021).

Traffic Guard Worker Alert System

The Traffic Guard Worker Alert System (TGWAS or WAS) uses a pneumatic detection system and provides auditory, visual, and haptic alarms. It is a commercially available system, but so far has been evaluated only in limited pilot deployments. Astro Optics advertises the device as having a 1,000-foot range but specifies that the device is a “Warehouse Audible” system, which suggests that it is marketed for an indoor environment as opposed to an outdoor highway work zone environment (TAPCO n.d.). If a vehicle drives over a pneumatic tube positioned at the perimeter of a work zone, the alarms are triggered, including an audio alarm that achieves 80 decibels at a span of 50 feet (Boodlal et al. 2020). Eseonu et al. (2018) found that workers operating noisy equipment are unable to hear the alarm.

Work Zone Presence/Layout Systems

Connected and Automated Vehicles

Typically, warning drivers of an approaching work zone or end of queue ahead can be done with an I2V system. Parikh et al. (2019) described this technology as able to warn approaching CAVs about upcoming work zones by providing advance warning to approaching drivers about obstructions, lane shifts, lane closures, speed reductions, or maintenance vehicles entering or exiting a work zone area. Most of the work in this domain is still ongoing at an experimental capacity.

Crowdsourced Data/Smartphone App

Crowdsourcing technologies (e.g., Waze, TomTom Go, Nexar) allow information to be collected either manually or automatically and shared with a large number of people who also enlist the services of the technology. These technologies have been deployed successfully for more than 10 years, and the near ubiquitous use of smartphones has made it easy to collect data with minimal effort. Some crowdsourcing applications allow users to opt in regarding the information they share, while other applications capture information without the user’s direct knowledge. In general, these applications allow information to be regularly updated and curated by the public. Some services analyze the environmental aspects of the data, including road inventory and work zones, for organizations to use in their asset management. For work zones, the application detects the presence of cones and aggregates the information where it can be accessed and monitored via an application programming interface. Studies in non-work-zone locations have shown that these crowdsourcing technologies have resulted in traffic incidents being identified 10 minutes sooner, a 30% decrease in vehicle crashes, and a 24% to 27% decrease in traffic delays (Amin-Naseri et al. 2018; Waze 2020). These crowdsourcing technologies can be used by traffic management centers or third-party providers to update road users about lane closures or other

changes to work zones along their route. Further, crowdsourcing features include the ability to upload images and select traffic conditions in active work zones, such as crashes and traffic queues, as well as information on lane activity conditions, including whether a lane is open, restricted, or closed (Adu-Gyamfi et al. 2019).

Speed Harmonization

Connected and Automated Vehicles

CAV-based speed harmonization systems have been proposed in the literature, but few field deployments have been conducted. In these systems, the vehicles themselves act as probes on the traffic stream. The speeds from equipped vehicles are then used to generate a recommended travel speed to optimize flow and safety. CAV-based speed harmonization typically involves reporting speeds to an IOO, which then generates a recommended speed that is transmitted via V2I communications. To date, systems that have relied on CAVs to generate the data for the speed harmonization system have only been examined via simulation, as the market penetration rate for CAVs has been insufficient for field testing in which the vehicles serve as the data collection tool (Ghiasi et al. 2019; Hale et al. 2016; Ma et al. 2016; Ramezani and Benekohal 2015). Those studies found widely varying results for the market penetration of CAVs required to generate system benefits, ranging from a low of about 10% (Hale et al. 2016) to a high of 80% (Ramezani and Benekohal 2015). The most widely reported CAV-based speed harmonization field test used three CAVs as data collectors and provided automated speed control, but the actual data collected to set the recommended speeds were from roadside radar trailers (Hale et al. 2016; Ramezani and Benekohal 2015).

CAV-based systems are still in an experimental stage. The market penetration rate of CAVs remains low, and the penetration required to support this application remains unclear. As a result, the ability to directly use CAVs to generate data for speed harmonization is still extremely limited. While using CAVs as a data source for speed harmonization is currently experimental, speed harmonization has a great deal of promise as market penetration increases, since data generation would not be tied to specific physical locations where sensors are deployed. This could provide a long-term robust way to characterize flow approaching the work zone taper and throughout the work zone.

Smartphone Applications

Mobile applications and Internet sites have also been used to share information from VSL/speed harmonization systems with road users, but they are not as prevalent as the use of VSL signs. These methods typically show a display that duplicates the speeds that are displayed on the field VSL signs, so they are not the sole means by which information on VSLs is distributed. Websites showing work zone conditions are provided by some commercial vendors. Likewise, some DOTs will archive and share information about real-time messages through online data portals for ingestion by state traveler information websites or other third-party sites. Mobile applications have been developed for VSLs deployed for the purpose of active traffic management, but no example of work-zone-specific applications was found.

Website-based platforms for information sharing are commercially available and well understood by agencies. No studies that determined the effects of disseminating data using these platforms were identified, but they are expected to create relatively small behavioral changes, since they are mirroring the information provided by the VSL signs. The information provided is accessible to everyone and can be accessed en route or before a trip. The information used to populate these sites is derived almost exclusively from SWZ systems where sensors and signs have been placed, so they are limited to locations where there is supporting infrastructure.

Speed Compliance

Connected and Automated Vehicles

CAVs can be used to reduce the speed of drivers approaching work zones by using roadside units to broadcast alerts (Parikh et al. 2019). A simulator study conducted by Whitmire et al. (2011) showed the in-vehicle warnings in connected vehicles can increase driver speed compliance, and these warnings are more effective when used in audio and visual modalities. Use of CAVs for speed compliance in work zones is still ongoing at an experimental capacity.

Speed Safety Camera Systems

While speed safety camera systems have been available for some time, they may be considered an emerging technology because legal and institutional barriers to deployment in many jurisdictions have prevented their widespread adoption. There has been some evidence to suggest that these systems are effective, but there have been limitations as to where they may be deployed. Therefore, speed safety camera systems cannot be fully assessed and considered as mature technologies.

Speed safety camera systems use sensors such as radar or cameras to identify and capture images of vehicle license plates or drivers, or both, of vehicles traveling over the speed limit. Speeding citations are then mailed to the vehicle’s registered owner. These systems have been used successfully by several state DOTs to reduce traveling speeds in work zones. For example, the use of speed safety camera systems in Illinois has reduced the average travel speed by 4 to 8 miles per hour (Benekohal et al. 2008). Furthermore, in Maryland, the percentage of vehicles exceeding the speed limit dropped from 7% to 1% after these systems were implemented (Franz and Chang 2011). More recently, in 2021, the Pennsylvania DOT reduced the total percentage of speeding vehicles in work zones where automated work zone speed enforcement was in place from an average of 30% to 35% at the start of the program in 2020 to an average of 18% to 20% (Pennsylvania DOT 2022). These systems can work in any deployed stationary work zone, although installing cameras and radar sensors at mobile work zones might be problematic. The data generated from these systems are only available where they are deployed and, typically, to the agency that deploys them. Speed safety camera systems can be deployed with the resources available to state DOTs, but implementing these systems requires the collaboration and involvement of multiple government stakeholders, such as law enforcement, the judiciary, and others (Douma et al. 2014).

Dynamic Merge Systems

Connected and Automated Vehicles

The CAV approach to disseminating guidance for dynamic lane merging involves both notification and vehicle control. In this approach, an IOO provides information on work zone events, potentially via a unified data exchange format (e.g., WZDx). This information would be transmitted to vehicles approaching a work zone via I2V communications. In dynamic lane merging assistance, data dissemination via the CAV approach can be broadly grouped into two types: (1) in-vehicle notification that relies on human intervention to conduct the merging process and (2) cooperative lane changing/merging control that uses an advanced driver assistance system (ADAS) to execute the merging process automatically. In the first type, CAVs in the closed lane can receive notifications about the presence and location of the work zone, while drivers in the open lanes are asked through in-vehicle notifications to cooperate and smooth the merge. A primary benefit of this approach is that it can accommodate mixed traffic conditions that consist of both autonomous and human-operated vehicles. In the second group, assuming 100% market

penetration, the work zone merging traffic enjoys a full suite of CAV benefits, including reduced gaps, increased throughput, and a system-optimal merging plan, that are made possible by full cooperative merging and are not reliant on voluntary compliance. However, because of the low market penetration of CAVs, most systems have only been tested via simulation.

Several studies have proposed and evaluated the effectiveness of various strategies for CAV-based work zone lane merging. Many simulation/modeling studies found that the cooperative control strategy can improve both work zone capacity and safety at 100% of cooperating vehicles. For example, a simulation study by Cao et al. (2021) found that, assuming a 0.5-second headway, a three-to-one lane closure work zone can experience a reduction in mean travel time of more than 50% through the use of CAV technology. Some other studies explored how the market penetration rate affects traffic safety and mobility (Algomaiah and Li 2021, 2022; Liu et al. 2017). Liu et al. (2017) concluded that the number of merging conflicts decreased with increases in the ADAS penetration rate. The simulation study found a 7.2% decrease in conflicts as the penetration rate increased from 20% to 50%. Conflicts decreased by an additional 4.3% as the penetration rate grew from 55% to 90%. Varying results in mobility performance are widely reported in the literature, ranging from a decrease of 1.5% in throughput with 90% ADAS market penetration (Liu et al. 2017) to a 45% increase in capacity with an optimized cooperative merging scenario that used central trajectory planning (Algomaiah and Li 2022).

Using CAVs to provide dynamic lane merging assistance is still in the experimental stage. Most data available on this approach come from simulation studies, and impacts varied significantly between studies. The deployment and field evaluation of these systems are contingent upon having enough market penetration to support the application. A CAV-based lane merging assistance system has obvious potential advantages because it does not require physical roadside signs to share information with drivers. This enables continuous merging recommendations over the course of a work zone and potentially improves flexibility, especially in cases in which a work zone might be in place only for a short duration or experience phasing changes. Furthermore, cooperative lane merging integrated with ADAS will likely create greater safety and mobility effects, as the system will no longer rely on driver compliance with recommendations.

Smartphone Applications

Lane merging assistance functions on smartphone applications are rarely stand-alone applications. Instead, they are usually part of navigation or traffic and traveler information applications. These navigation or traveler information apps on smartphones are a well-developed, commercially available technology with multiple vendors. The navigation software can be installed by the smartphone user directly from the app marketplace or accessed directly via a software website, usually free of charge. However, not every user may have the software installed or be actively using it when approaching a work zone. One challenge with disseminating information to assist dynamic lane merging is that the merging strategy changes with real-time traffic conditions, and pre-trip information may be obsolete when the motorist arrives at the work zone. Therefore, dynamic lane merge systems would require continuous and frequent updates on work zone status, presumably via the WZDx. Most navigation systems include only static work zone data elements (e.g., work zone locations), but not dynamic lane merging information. Because this technology requires no physical devices to be deployed onto the road, the static lane merging assistance function on smartphone applications, if available, works under all traffic conditions at any sites given an available Internet connection. No studies were identified that determined the effects of disseminating dynamic lane merging information using smartphone app platforms.