In 2015, large trucks (trucks with a gross vehicle weight rating of more than 10,000 pounds) were involved in 414,958 crashes that resulted in 116,000 injuries and 4,067 fatalities (Federal Motor Carrier Safety Administration, 2016). The AAA Foundation for Traffic Safety identified the potential of several large-truck advanced safety technologies as promising countermeasures to reduce these crashes. Advanced safety technologies may use sensors or alerts to warn a driver of a possible collision, actively assume control of a vehicle in situations where a driver does not react to the threat of an imminent crash, or improve driver and fleet management (e.g., monitoring vehicle safety systems and drivers’ hours-of-service status). Although some advanced safety technologies may be effective at preventing crashes, it is also important to know whether they are cost-effective, as this information may assist consumers in purchasing advanced safety technologies and/or government regulators in mandating their use. The objective of this research was to provide scientifically-based estimates of the societal benefits and costs of advanced safety technologies in large trucks (i.e., the impacts an advanced safety technology may have across the entire society if implemented) in order to (1) allow the Department of Transportation to make informed decisions related to potential regulations on advanced safety technologies, and (2) promote the adoption of cost-effective advanced safety technologies to motor carriers. To accomplish this objective, an in-depth literature synthesis of 14 advanced safety technologies was completed, an expert advisory panel informed cost and benefit estimations for all advanced safety technologies (based on the literature review and their experience and knowledge), and benefit-cost analyses were performed on selected advanced safety technologies. The advisory panel recommended the following four technologies for benefit-cost analysis: video-based onboard safety monitoring systems, lane departure warning systems, automatic emergency braking systems, and air disc brakes. This report presents the results related to video-based onboard safety monitoring systems. See other AAA Foundation reports for analyses of automatic emergency braking systems, lane departure warning systems, and air disc brakes. Video-based onboard safety monitoring systems incorporate in-vehicle video technology that records the environment surrounding the vehicle, as well as the driver’s behaviour and performance. These types of systems typically have at least one camera but usually have two. One camera records the forward roadway and shows what the driver can see through the front windshield. The second camera faces and records the driver, showing how they behave behind the wheel and respond to driving situations. Some video-based onboard safety monitoring systems incorporate additional cameras. Additional camera feeds may include a rear camera that shows what is behind the vehicle, and rear-facing left-and right-side cameras that show the lanes adjacent to the vehicle. Video-based onboard safety monitoring systems also capture driver kinematic performance through vehicle telematics. Some of the driving behaviours that these systems can measure and record include speeding, hard braking, rapid acceleration, quick cornering, seat belt use, turn signal use, driver distraction, following distance (if the vehicle is equipped with forward radar), and lane departures (if the vehicle is equipped with a lane departure warning system). This combination of visual and kinematic data provides a wealth of video and vehicle information to pinpoint problems and safe and unsafe behaviours. Most video-based onboard safety monitoring systems continuously monitor the driver whenever the vehicle is on. When a specific threshold is exceeded (e.g., a hard-braking event that exceeds 0.3 g), the system automatically saves a predefined amount of data (e.g., 30 seconds prior to the event and 60 seconds after the event) to the memory card. These safety events in the memory card are used in driver coaching and feedback sessions. The literature review only identified two empirical studies that evaluated crash reductions associated with video-based onboard safety monitoring systems. However, eight case studies were found on technology vendor websites. The empirical studies found video-based onboard safety monitoring systems may have prevented 38.1% to 52.2% of large-truck safety-critical events, 20% of large-truck fatal crashes, and 35.5% of large-truck injury crashes. The case studies found that 44% to 86% of safety-critical events may be prevented, and 61% to 80% of crashes may be prevented with video-based onboard safety monitoring systems. No published documents provided costs associated with video-based onboard safety monitoring systems. However, one technology provider shared detailed cost data. These costs are listed below. • Hardware: $300-$600 (per system) • Installation: $0-$150 (per system) • Monthly service fee: $20-$60 (per system) • Training costs included • Free integration with other Advanced safety technologies • Coaching - Averages 10 minutes per driver - During first two months, approximately 25% of drivers receive coaching - After one to two months, only 1% of drivers require coaching - On average, one manager is responsible for coaching 75 drivers An Expert Advisory Panel convened May 17, 2016, at the AAA Foundation for Traffic Safety headquarters in Washington, D.C. This advisory panel consisted of six individuals representing various aspects of the industry, including representatives from a commercial motor vehicle carrier, a trucking insurance company, the Federal Motor Carrier Association, the National Highway Traffic Safety Administration, and an advanced safety technology vendor. The panel also included an industry safety consultant. The purpose of this meeting was twofold: (1) to assist the research team in selecting technologies that require a benefit-cost analysis, and (2) to identify the appropriate efficacy rates and costs to be used in the benefit-cost analyses. Following this discussion, a benefit-cost analysis was recommended for video-based onboard safety monitoring systems, and upper- and lower-bound efficacy rates and costs were selected for video-based onboard safety monitoring systems. For video-based onboard safety monitoring systems, the advisory panel recommended efficacy rates of 20% and 52.2% to reflect current performance capabilities of video-based onboard safety monitoring systems (instead of systems that were under development). This recommendation was based on current carrier conservative efficacy estimates, Hickman and Hanowski (2012), and Soccolich and Hickman (2014). Additionally, the panel recommended the cost described above for use in the benefit-cost analysis. The benefit-cost analysis followed conventional methods used in similar studies (e.g., Hickman et al., 2013) to estimate the societal benefits and costs of implementing video-based onboard safety monitoring systems in the trucking industry. Societal benefits of video-based onboard safety monitoring systems associated with a reduction in crashes were compared to the costs of deploying video-based onboard safety monitoring systems across the entire U.S. fleet of large trucks. The benefit and cost factors considered in this study are discussed below. Benefit Factors: • Medical-related costs • Emergency response service costs • Property damage • Lost productivity • Monetized value of pain, and the suffering and quality-of-life decrements experienced by families in a death or injury Cost Factors: • Video-based onboard safety monitoring systems’ hardware purchase, installation, and financing costs • Video-based onboard safety monitoring systems’ maintenance costs • Video-based onboard safety monitoring systems’ replacement costs • Costs associated with video-based onboard safety monitoring systems training for drivers and managers • Costs associated with coaching drivers To assess the impact video-based onboard safety monitoring systems could have on reducing crash rates (and the costs associated with the systems), national crash databases were used to identify video-based onboard safety monitoring systems’ target population. These crash databases included the Fatality Analysis Reporting System (FARS) and the General Estimates System (GES). The FARS database was used to determine the number of fatal crashes and their associated fatalities and injuries, and the GES database was used as an estimation for injury and property damage only crashes. The GES database also was used to estimate the number of injuries as a result of injury crashes. Queries were developed for video-based onboard safety monitoring systems and information was extracted for different vehicle types for a period of six years (2010 to 2015). When filtering the GES and FARS crashes, the research team carefully considered the scenarios where video-based onboard safety monitoring systems may have prevented the crash. Whereas other advanced safety technologies prevent specific crash types, video-based onboard safety monitoring systems are applicable to many different crashes, as long as the large-truck driver could have done something to prevent or mitigate the crash. Thus, all large-truck crashes that resulted from an error by the driver of the large truck were included. The research team used the same GES/FARS filtering criteria found in Soccolich and Hickman (2014). The complete list of GES/FARS variables is located in Appendix B. Two sets of benefit-cost analyses were performed for video-based onboard safety monitoring systems. The first set of analyses included retrofitting the entire U.S. fleet of large trucks. This approach assumed all new vehicles added to the fleet would be equipped with video-based onboard safety monitoring systems and old vehicles would be retrofitted. This analysis approach represented the scenario with the most benefits but also the highest costs. The second set of analyses used an annual incremental costs analysis approach. This approach assumed all new vehicles would be equipped with video-based onboard safety monitoring systems (starting in 2018) and did not include retrofitting old vehicles. Societal benefits were assessed over the life of the vehicle. Additionally, for each analysis approach, an analysis was performed on different types of large trucks. The first analysis included all large trucks (gross vehicle weight rating greater than 10,000 pounds). The second analysis was performed only using class 7 and 8 combination unit trucks (CUTs). The third analysis was performed only using single unit trucks (SUTs) with gross vehicle weight rating greater than 10,000 pounds. Finally, separate analyses were performed to account for the rate of monetary discount, in the present value, of the cost and benefits in any future year. Following guidance from the Office of Management and Budget (OMB, 2003) analyses were performed using a 0%, 3%, and 7% discount rate. Video-based onboard safety monitoring systems were evaluated using a low and high efficacy rate (20% and 52.2%, respectively) and a low, average, and high hardware cost ($350, $525, and $750, respectively). Additionally, monthly service fees and driver coaching costs were incorporated into each analysis. Table 1 shows the benefit-cost ratios for video-based onboard safety monitoring systems when equipping all trucks (new and old). The analyses with a benefit-cost ratio greater than 1.00, which indicate that the benefits outweigh the costs, are highlighted. For example, the first row of results in Table 1 shows the results for all large trucks using a high efficacy rate for video-based onboard safety monitoring systems. When the costs of video-based onboard safety monitoring systems are equal to the average cost used in the analysis and the discount rate is 0%, the estimated benefits of video-based onboard safety monitoring systems are 4.51 times the estimated costs. However, when the costs of video-based onboard safety monitoring systems are high and the discount rate is 7%, the estimated benefits of video-based onboard safety monitoring systems are 3.13 times the estimated costs. Sensitivity analyses were performed with a higher value of a statistical life ($13,260,000) and with a lower value ($5,304,000). As video-based onboard safety monitoring systems were cost-effective in all of the analyses above when installing the systems in the entire U.S. fleet for all trucks and for both single-unit trucks and combination-unit trucks individually, using a higher value of a statistical life in the calculations would only make these systems more cost-effective. Thus, only the results with the lower value of a statistical life are shown below (Table 2). The results with the higher value are shown in Appendix C. Using the $5,304,000 value of a statistical life, video-based onboard safety monitoring systems were still cost-effective, regardless of vehicle classification or system cost, when the high efficacy rate was used in analyses. However, only the low- and average-cost systems were cost-effective using the lower efficacy rate and lower value of a statistical life (the average-cost system was no longer cost-effective for single unit trucks). Table 3 shows the benefit-cost ratios for video-based onboard safety monitoring systems when only equipping new trucks. As shown in Table 3, low-, average-, and high-cost video-based onboard safety monitoring systems were cost-effective for both the lower and upper efficacy rate with all truck types. Table 4 shows the sensitivity analyses for only equipping new trucks with video-based onboard safety monitoring systems using the lower value of a statistical life. The results with the higher value are shown in Appendix C. As shown in Table 4, the high efficacy rate still resulted in cost-effective solutions for all cost estimates, regardless of vehicle classification, when using the lower value of a statistical life. The lower efficacy rate also resulted in cost-effective solutions for the low- and average-cost video-based onboard safety monitoring systems with the lower value of a statistical life. However, the high-cost video-based onboard safety monitoring system was only cost-effective at a 7% discount rate when the analysis used both the lower efficacy rate and the lower value of a statistical life. This report presents the scientifically-based estimates of the societal benefits and costs of video-based onboard safety monitoring systems installed on large trucks. The current study used efficacy rates from previously published research and identified crashes that may have been prevented through the deployment of a video-based onboard safety monitoring system. Crashes were identified using 2010 to 2015 GES and FARS data sets. Benefit-cost analyses were performed using varying efficacy rates, vehicle types, system costs, and discount rates (0%, 3%, and 7%). This project was the first to perform empirical-based benefit-cost analyses on video-based onboard safety monitoring systems. Video-based onboard safety monitoring systems were shown to be cost-effective regardless of their cost and efficacy (within the ranges of cost and efficacy considered) when analyses were performed using a $9.4 million value of a statistical life. Video-based onboard safety monitoring systems were found to have a high benefit-cost ratio of 10.17 when new large trucks were equipped with the technology (with a 3% discount rate). The benefit-cost ratio was even higher when only new combination-unit trucks were equipped with the systems. Despite the significant costs associated with driver coaching, the results of this study suggest that the societal benefit of equipping large trucks with video-based onboard safety monitoring systems, expressed in economic terms, substantially outweigh the associated costs. Limitations: Although the analyses used to assess the benefits and costs associated with video-based onboard safety monitoring systems were comprehensive, there were several limitations, including the following: • It is possible the efficacy rates used in this study may not represent the current functionality/effectiveness of the current generation of video-based onboard safety monitoring systems. However, the advisory panel consisted of experts with knowledge of current technology research, and as such, the efficacy rates recommended by the panel should be consistent with the current generation of systems’ efficacy rates. • The technology costs used in this study may differ from current costs (costs typically decrease over time). • This study used estimated crash, technology, and labour costs. It is possible that actual costs may differ and thus impact the cost-effectiveness of video-based onboard safety monitoring systems. • The GES database only included crashes that required a police accident report. However, video-based onboard safety monitoring systems may also prevent less severe crashes. Thus, these additional benefits are not accounted for in the benefit-cost analyses. • The real-world effectiveness of the systems against different severity crashes may differ significantly. However, data limitations precluded the use of separate efficacy estimates for this study. • These analyses did not account for reduced litigation costs associated with reduced crashes or driver exonerations. These may be significant cost savings that were not integrated into the analyses. • The efficacy of video-based onboard safety monitoring relies on carrier management following driver coaching protocols. These protocols include using the data generated by the video-based onboard safety monitoring system for driver coaching. System efficacy and reductions in crashes outlined in this project may not be achieved if coaching protocols are not adhered to. • The efficacy of video-based onboard safety monitoring systems is dependent upon effective introduction, then initial and subsequent ongoing driver and management training. • This study assumed all vehicle systems were functioning as intended. However, this is unlikely to be seen in the real world. For example, anti-lock brakes and foundation brakes have a direct impact on a vehicle’s ability to avoid a crash. If they are poorly maintained, the actual efficacy rates achieved may be lower than those used in this study. (Author/publisher)
Abstract