CHAPTER 2

Literature Review

This chapter documents the results of the literature review. The literature included agency reports, peer-reviewed journal articles, web articles, agency websites, and international examples. Literature related to rail crossings’ operation, safety, and security was included in the review. Literature pertaining to descriptions of the technologies involved and available and their implementation at rail transit and commuter rail crossings was of special interest. In addition, it involved documenting emerging trends and future potentials for electronic surveillance of rail crossings.

Rail Crossing Operation, Safety, and Security

There are several issues and needs related to rail crossing operation, safety, and security. These are related to

• Traffic blockage and related impacts on delays to drivers and emergency operations,
• Control and warning devices,
• Physical security,
• Fatalities and injuries related to pedestrians and vehicles,
• Device malfunction and human error,
• Distracted drivers and pedestrians, and
• Trespassers and suicides.

After 9/11, the Transportation Security Administration (TSA) and the Department of Homeland Security (DHS) were established to strengthen interagency collaboration and the government architecture against terrorism. The TSA was created by Congress through the enactment of the Aviation and Transportation Security Act. Although originally within the U.S. Department of Transportation, the TSA and other agencies were transferred to DHS after the passage of the Homeland Security Act of 2002 (DHS 2002). From a security standpoint, transit targets such as transit vehicles, transit and related infrastructure, communications systems, and transit personnel need to be secure. Rail crossings were not the focus but were observed to identify threats to critical rail infrastructure. Thus, earlier monitoring efforts related to rail crossings were from a security standpoint.

The safety of control devices as well as the potential collision of vehicles and pedestrians have always been the focus at rail crossings. Now, additional concerns have emerged. The rail transit and commuter rail agencies face a continuing problem with trespassing incidents that occur on rail and transit systems throughout the United States. This issue was thoroughly investigated in recently concluded research (Warner et al. 2022a; Warner et al. 2022b). There is also more emphasis on near misses and learning from that data to design more effective control devices or enforcement programs. All this has created an urgent and expanded need for safety data with greater granularity.

Grade crossings are a primary source of risk to the safe operation of passenger and freight services globally. This risk is also recognized at a national level by the FRA and manifested in several programs and industry-wide initiatives designed to tackle risk on the railroad network. It is estimated that 94% of fatalities on the U.S. railroad networks relate to road-rail crossings and trespassing. Figures 1 to 4 provide insights into fatality and injury rates per 100 million (100M) vehicle revenue miles (VRM) traveled by different rail transit modes and suggest more attention is needed on rail transit crossing safety. This is meant to provide a general overview of the numbers, not a comprehensive one, since there are two different systems [FRA (heavy rail) and FTA (light rail)].

Systems and Technologies

Previous TCRP reports, like TCRP Synthesis 38: Electronic Surveillance Technology on Transit Vehicles (Maier and Malone 2001) and TCRP Synthesis 80: Transit Security Update (Nakanishi 2009), have studied overall security and the use of electronic video surveillance technology in the transit environment. TCRP Synthesis 21: Improving Transit Security (Needle and Cobb 1997) considered how transit agencies were using video surveillance as part of their overall security strategies, primarily in conjunction with uniformed patrol by police or security officers. Most of the examples and case examples in earlier reports combined discussions of the use of electronic video surveillance cameras in bus and rail systems and a few considered non-security uses of such technology. TCRP Synthesis 90: Video Surveillance Uses by Rail Transit Agencies (Schulz and Gilbert 2011) differed from the earlier ones as it documented the use of electronic video surveillance technology solely by passenger rail agencies and considered the totality of its use, including onboard railcars and along the right-of-way. The synthesis also described

• Current administrative policies on monitoring video images either in real time or for post-event analysis;
• Policies on archiving and storing images and access to them by employees, other public agencies (primarily police), and the general public; and
• Funding sources for installing new or upgrading existing video surveillance systems.

The monitoring of railroad crossings has evolved in both technologies and applications. Photo enforcement was the first monitoring attempted and implemented at rail crossings for security as well as reducing vehicle violations (Carroll and Warren 2002; ICC 2002). A railroad grade crossing monitoring project, the Sugar Land Rail Monitoring System, was developed by the Houston District of the Texas Department of Transportation and Texas A&M Transportation Institute as part of the ITS Priority Corridor Program that examined how real-time detection, communication, and information systems can be integrated to monitor the movements of trains in a corridor to reduce conflicts and delays created by railroad grade crossings for the primary benefit of fire and police personnel while on emergency runs (Goolsby et al. 2003). The system monitored the following parameters: the presence of train and direction, the speed of the train, the length of the train, and crossing gate closures via monitoring of traffic signal preemption relays. Sheikh et al. (2004) presented a real-time computer vision system for the monitoring of the movement of pedestrians, bikers, animals, and vehicles at railroad intersections, which processed the video images for the detection of uncharacteristic events, triggering an immediate warning system. This system performs robust object detection and tracking and uses a classification algorithm to determine whether the detected object is a pedestrian, biker, group, or vehicle, allowing inferences on whether the behavior of the object is characteristic or not. Since then, several other

efforts followed (Courage and Kirkpatrick 2003; Courage and Ko 2003; Ko et al. 2003; Blacketer et al. 2005; Simmons 2007; Medina et al. 2011; Medina and Benekohal 2015, Medina and Benekohal 2016; Thomson et al. 2016; Chen and Rilet 2017; Zhang et al. 2017; Okorodudu et al. 2018; Zaman et al. 2018, Tackett 2018; Zaman et al. 2019; Fayyaz et al. 2021; Nunes et al. 2021; Pamuła and Pamuła 2021; Qian et al. 2022; Herr 2023; Sick 2023; Tang et al. 2023) and even led to several patents [Stevenson and Calixto 2006; Alexander and Haley 2013; Steffen and Isaacson 2013 (Patent US 20130270395 A1); Rempel et al. 2019 (Patent US 20190145791 A1)]. Rail crossing monitoring has also been relevant in understanding driver and pedestrian distractions and behaviors and related safety concerns (Khattak 2007, Khattak 2009; Khattak and Luo 2011; Tung and Khattak 2015; Lee and Khattak 2016).

Custom-made, commercial surveillance or monitoring systems have been developed by ONYX-Rail Safety Solutions (ONYX 2023), Island Radar (Island Radar 2023), Nokia (Nokia 2023), BriefCam (BriefCam 2023), TRAINFO (TRAINFO 2023), Infinova (Infinova 2023), RXM Rail Crossing Monitoring (Voestalpine 2023), Verra Mobility (Verra Mobility 2023), Milestone Systems (Security Informed 2022), LBFoster (LBFoster 2023), and others. Network Rail (Network Rail 2023) has a proprietary system, which has both closed-circuit television (CCTV)-based and LIDAR-/RADAR-based systems, and they use a risk tool to identify and prioritize where they would use one or other or neither. A Rutgers University team received two grants, both from FTA, to develop an artificial intelligence (AI)-based video analytics system to monitor and develop actionable intelligence about movements, potential for collision or near misses, and better safety metrics, which were not possible from paper-recorded incident reports (Zaman et al. 2023). Advancements in acoustic, thermal, and other sensors are expanding capabilities to develop usable information from raw images under good and adverse weather and day as well as night conditions. Video surveillance has been used to help with security and operations compliance. Advancements in fiber optics have improved connectivity to install surveillance systems in more widespread locations. Storage devices’ capabilities have increased significantly with cloud services. The video analytic systems for CCTV traffic videos can help alleviate the workload for traffic operators in traffic operations and management. Cloud computing technologies use an on-demand architecture to dynamically assign computer resources such as storage and computing powers without the need to actively manage a local computing center. Cloud computing has had significant success in many areas such as video conferencing, video streaming, computer gaming, and so forth.

Applications and Implementation of Technologies

Electronic surveillance of rail crossings has been implemented at many crossings, many through pilot programs supported through FTA and FRA grants. Primarily, several crossings were photo-enforced starting in 2002, particularly in California and Illinois. The Utah Transit Authority (UTA) and the Tri-County Metropolitan Transportation District of Oregon (TriMet) have used it extensively in the United States. Network Rail, in the United Kingdom, examines its 6,000 rail crossings each year based on risk assessment tools to decide if a crossing needs CCTV or RADAR system or if no implementation of monitoring is needed. These implementations have been motivated by several needs related to a better understanding of safety and operation at and in the vicinity of rail crossings. Applications have been used to gather data on safety collisions, fatalities, injuries, violations, alerts, the health of devices, trespassing, dwelling, suicides, street blockage, traffic management, traffic advisories, interfacing with positive train control (PTC), and near misses.

There are critical issues related to the implementation of technologies for electronic surveillance of rail crossings. There is anecdotal evidence that there has not been consistent application, so there are no standard operating procedures, nor appropriate plans for maintaining and updating

a system if one is installed, and funding is uncertain or not sustaining. Cost and funding are the biggest barriers to pursuing electronic surveillance. A lack of trained staff and technical resources further exacerbates the situation, not just in setting up the system but also in sustaining it.

Emerging Opportunities and Technologies

Some emerging technologies include the following:

• Technology for cameras has matured.
• Fiber optics has improved connectivity and accessibility.
• Cloud storage has expanded abilities to store terabytes of data.
• Artificial intelligence, machine learning, and deep learning have allowed better data mining and interpretation.
• Video analytics has unraveled new information or improved our understanding.

Some emerging opportunities include

• Safety and security,
• Institutional support and streamlining of processes,
• Improved decision-making through better safety data,
• Going for diversified applications (including both safety and operational data and information), and
• Partnerships with railroad, city, and private entities.

Summary of Literature Review

The literature review informed the type and nature of questions to have in the questionnaire. Literature on the cost of implementations, successes, and failures of electronic surveillance of rail transit and commuter rail crossings is lacking. This necessitated the need for a survey of agencies, individuals, and vendors. The intent of the survey is to obtain insights on

• Motivation,
• Decision criteria,
• Measures of effectiveness,
• Funding,
• Capital costs,
• Operation and recurring costs,
• Reasons for successes and failures,
• Implementation challenges,
• System and technologies, and
• Other related aspects of rail transit and commuter rail crossings.