A Guide for Reducing Collisions Involving Heavy Trucks (2004)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

V-1 SECTION V Descriptions of Strategies Objectives To reduce the number of heavy-truck fatality crashes, the objectives should include • Reducing truck driver fatigue, • Strengthening commercial driver’s license (CDL) requirements and enforcement, • Increasing public knowledge about sharing the road, • Improving maintenance of heavy trucks, • Identifying and correcting unsafe roadway and operational characteristics, • Improving and enhancing truck safety data, and • Promoting industry safety initiatives. Explanation of the Objectives Because truck crashes result from the interaction of so many factors, reducing crash probability requires comprehensive efforts addressing programs affecting truck drivers, other roadway users, vehicle condition, carrier operational practices, highway design and characteristics, and traffic records. Commercial Driver Fatigue In a major national forum on truck safety, the primary safety issue identified was driver fatigue. The reasons for driver fatigue are many, and only some of them may be addressed through state programs. However, states can take steps to increase the efficiency of use of existing parking space for drivers needing rest and/or required to stop driving because of hours-of-service regulations; states can also modify existing space and create new space to provide additional parking facilities. Rumble strips can alert tired drivers that they are leaving the traffic lane. Commercial Driver’s License The commercial driver’s license (CDL) established national standards for acquiring a license to operate heavy trucks. It has been fully implemented since April 1992. Although the CDL has achieved major improvements, e.g., reducing the problem of multiple licensing and consolidating driver history information, problems remain. The administration of the test can be improved, and measures can be taken to reduce fraud and improve the quality of both state and third-party testers. Improved Driver Behavior Most truck fatalities occur in multivehicle crashes, and in 2000, 78 percent of all heavy- truck–related fatalities occurred to occupants of the other vehicles. In crashes involving a heavy truck and a passenger vehicle (80 percent of all fatal truck crashes), it appears that the

SECTION V—DESCRIPTIONS OF STRATEGIES V-2 principal culpability most often lies with the driver of the other vehicle. Consequently, some effort needs to focus on drivers in general to reduce truck fatalities. Drivers need better information on how to share the road with large trucks. Improved Maintenance Heavy trucks generally accumulate high mileage. In 2000, combination trucks averaged almost 65,000 miles, compared with almost 12,000 for passenger vehicles. State vehicle inspection programs (and not all states have them) are designed for passenger cars and usually require inspection only once a year. Large trucks need to be inspected much more frequently. Roadside inspections invariably identify sizeable proportions of trucks that need to be taken out of service immediately because they are considered too hazardous to continue operating. In-depth inspection of trucks in fatal crashes indicates that about one- third would have been removed from service if inspected just prior to the crash. Improved Road Design and Operations Highway configuration can create hazards for some large trucks. Programs to identify and correct highway segments that pose significant hazards to trucks can reduce crashes. While making changes to the highway is costly, providing information to drivers concerning upcoming hazards and providing real-time feedback on excessive speed for safe maneuvering can be implemented at relatively low cost. Enhanced Safety Data Good data are the backbone to any successful highway safety program. Timely and accurate data are required to identify problems (with both vehicles and drivers), establish priorities, design interventions, evaluate countermeasures, and detect emerging problems. Important data on heavy trucks and their operators come from law enforcement, the judicial system, driver records, vehicle registration, and motor carrier records. Rapid entry and compilation of such data can greatly improve the detection of problems and enable immediate intervention. Promotion of Industry Safety Initiatives Unlike the general population of drivers and vehicles, commercial drivers and trucks operate under management supervision. Effective supervision of drivers and the vehicle fleet requires active and systematic management to ensure compliance with all federal and state regulations. Moreover, regulatory compliance is not the only goal. Many safety management activities of the most safety-conscious fleets go well beyond minimum compliance requirements. States and motor carrier industry leaders can work together to provide consultation to carrier safety managers on best practices to ensure both regulatory compliance and safety practices above and beyond compliance. One practice to enhance fleet safety above minimum required levels is the purchase and use of truck safety technologies (such as improved brakes) and advanced technologies (such as collision avoidance warning systems). Exhibit V-1 lists the objectives and several related strategies for reducing heavy-truck crashes. Details of these strategies are covered in the following narrative. It should be noted that this list does not represent all possible strategies to reduce heavy-truck crashes. AASHTO has chosen to concentrate efforts in this guide on lower-cost strategies that can be

SECTION V—DESCRIPTIONS OF STRATEGIES implemented relatively quickly, including strategies that can be applied to “spots” on the roadway (e.g., curvature on off-ramps). With few exceptions, these lower-cost, quickly implementable strategies are covered in the following pages. V-3 EXHIBIT V-1 Emphasis Area Objectives and Strategies Objectives Strategies 12.1 A Reduce fatigue-related crashes 12.1 B Strengthen CDL program 12.1 C Increase knowledge re: sharing the road 12.1 D Improve maintenance of heavy trucks 12.1 E Identify and correct unsafe roadway infrastructure and operational characteristics 12.1 F Improve and enhance truck safety data 12.1 G Promote industry safety initiatives a Explanations of (E), (T), and (P) appear on page V-4. b Mandated technologies are not considered to be experimental, although it would be advisable to carefully monitor and report them, as early as possible, upon an assessment of their introduction. 12.1 A1 Increase efficiency of use of existing parking spaces (E)a 12.1 A2 Create additional parking spaces (T) 12.1 A3 Incorporate rumble strips into new and existing roadways 12.1 B1 Improve test administration for the CDL (T) 12.1 B2 Increase fraud detection of state and third-party testers (T, E) 12.1 C1 Incorporate Share the Road information into driver materials (T) 12.1 C2 Promulgate Share the Road information through print and electronic media (T) 12.1 D1 Increase and strengthen truck maintenance programs and inspection performance 12.1 D2 Conduct postcrash inspections to identify major problems and problem conditions (E) 12.1 E1 Identify and treat truck crash roadway segments—signing (E) 12.1 E2 Install interactive truck rollover signing (P) 12.1 E3 Modify speed limits and increase enforcement to reduce truck and other vehicle speeds (T) 12.1 F1 Increase the timeliness, accuracy, and completeness of truck safety data (NA) 12.1 G1 Perform safety consultations with carrier safety management (P) 12.1 G2 Promote development and deployment of truck safety technologies (E)b Types of Strategies The strategies in this guide were identified from a number of sources, including the literature, contact with state and local agencies throughout the United States, and federal programs. Some of the strategies are widely used, while others are used at a state or even a local level. Some have been subjected to well-designed evaluations to prove their effectiveness. On the other hand, it was found that many strategies, including some that are widely used, have not been adequately evaluated.

The implication of the widely varying experience with these strategies, as well as the range of knowledge about their effectiveness, is that the reader should be prepared to exercise caution in many cases before adopting a particular strategy for implementation. To help the reader, the strategies have been classified into three types, each identified by letter throughout the guide: • Proven (P): Those strategies that have been used in one or more locations, and for which properly designed evaluations have been conducted that show them to be effective. These strategies may be employed with a good degree of confidence, but understanding that any application can lead to results that vary significantly from those found in previous evaluations. The attributes of the strategies that are provided will help the user judge which strategy is the most appropriate for the particular situation(s). • Tried (T): Those strategies that have been implemented in a number of locations, and that may even be accepted as standards or standard approaches, but for which there have not been found valid evaluations. These strategies—while in frequent, or even general, use—should be applied with caution, carefully considering the attributes cited in the guide, and relating them to the specific conditions for which they are being considered. Implementation can proceed with some degree of assurance that there is not likely to be a negative impact on safety and very likely to be a positive one. It is intended that as the experiences of implementation of these strategies continues under the AASHTO Strategic Highway Safety Plan initiative, appropriate evaluations will be conducted, so that effectiveness information can be accumulated to provide better estimating power for the user, and the strategy can be upgraded to a “proven” one. • Experimental (E): Those strategies that are ideas that have been suggested and that at least one agency has considered sufficiently promising to try them on a small scale in at least one location. These strategies should be considered only after the others have proven not to be appropriate or feasible. Even where they are considered, their implementation should initially occur using a very controlled and limited pilot study that includes a properly designed evaluation component. Only after careful testing and evaluations show the strategy to be effective should broader implementation be considered. It is intended that as the experiences of such pilot tests are accumulated from various state and local agencies, the aggregate experience can be used to further detail the attributes of this type of strategy, so that it can be upgraded to a “proven” one. Targeting the Objectives The first objective, reducing truck driver fatigue, is related to growing recognition of its role in both truck and passenger vehicle crashes. Truck driver fatigue is of special concern, because of the long hours of driving demanded by trucking and greater potential hazard posed by the heavy vehicle. Recent investigations (e.g., Fleger et al., 2002; Chen et al., 2002) have documented the shortage of adequate parking spaces for truckers seeking rest along Interstate and other major highways. This shortage makes it difficult to comply with federal hours-of-service restrictions. Although the CDL program has achieved many of its goals, there remain serious problems in the overall licensing program, including truck driver training schools using copies of the knowledge test to “teach the test” to candidate drivers and lax, and even fraudulent, administration of tests by third-party testers and state examiners. Because the general driving public is a major contributor to truck crashes, some effort needs to be targeted at this population SECTION V—DESCRIPTIONS OF STRATEGIES V-4

SECTION V—DESCRIPTIONS OF STRATEGIES to communicate ways to share the road safely with trucks. Truck maintenance is a chronic problem, and vehicle defects contribute to serious crashes. Certain stretches of roadway, such as off-ramps or extended downgrades, are particularly hazardous for large trucks. Although making major changes in roadways may be prohibitively expensive, much less costly measures such as signs to alert drivers can reduce the risk. Trucks cross jurisdictional lines much more than other traffic, and enforcement of CDL requirements is much more effective if relevant data are made available quickly, accurately, and completely. Finally, industry—both the motor carrier industry and truck manufacturers—can play an active role in enhancing heavy- truck safety if encouraged and given good information on safety practices and equipment. Related Strategies for Creating a Truly Comprehensive Approach The strategies listed above, and described in detail below, are largely unique to this emphasis area. However, to create a truly comprehensive approach to the highway safety problems associated with this emphasis area, related strategies may be included as candidates in any program planning process. These are of five types: • Public Information and Education (PI&E) Programs: Many highway safety programs can be effectively enhanced with a properly designed PI&E campaign. The primary experience with PI&E campaigns in highway safety is to reach an audience across an entire jurisdiction, or a significant part of it. However, it may be desired to focus a PI&E campaign on a location-specific problem. While this is a relatively untried approach, as compared with areawide campaigns, use of roadside signs and other experimental methods may be tried on a pilot basis. Within this guide, where the application of PI&E campaigns is deemed appropriate, it is usually in support of some other strategy. In such a case, the description for that strategy will suggest this possibility (see the attribute area for each strategy entitled “Associated Needs”). In some cases, when PI&E campaigns are deemed unique for the emphasis area, the strategy is explained in detail. As additional guides are completed for the AASHTO plan, they may address the details regarding PI&E strategy design and implementation. • Enforcement of Traffic Laws: Well-designed and -operated law enforcement programs can have a significant effect on highway safety. It is well established, for instance, that an effective way to reduce crashes and their severity is to have jurisdictionwide programs that enforce an effective law against driving under the influence (DUI), or driving without seatbelts. When that law is vigorously enforced, with well-trained officers, the frequency and severity of highway crashes can be significantly reduced. This should be an important element in any comprehensive highway safety program. Enforcement programs are conducted at specific locations by the nature of how they must be performed. The effect (e.g., lower speeds, greater use of seatbelts, safer vehicle-condition, and reduced impaired driving) may occur at or near the specific location where the enforcement is applied. Coordinating the effort with an appropriate PI&E program can often enhance this effect. However, in many cases (e.g., speeding and seatbelt usage) the impact is areawide or jurisdictionwide. The effect can be either positive (i.e., the desired reductions occur over a greater part of the system) or negative (i.e., the problem moves to another location as road users move to new routes where enforcement is not applied). A pilot program is recommended when it is unclear how the enforcement effort may impact behavior or V-5

where it is desired to try an innovative and untried method. Within this guide, where the application of enforcement programs is deemed appropriate, it is often in support of some other strategy. Many of those strategies may be targeted at either a whole system or a specific location. In such cases, the description for that strategy will suggest this possibility (see the attribute area for each strategy entitled “Associated Needs”). In some cases, where an enforcement program is deemed unique for the emphasis area, the strategy will be explained in detail. As additional guides are completed for the AASHTO plan, they may address the details regarding the design and implementation of enforcement strategies. • Strategies to Improve Emergency Medical and Trauma System Services: Treatment of injured parties at highway crashes can have a significant impact on the level of severity and length of time that an individual spends in treatment. This is especially true when it comes to timely and appropriate treatment of severely injured persons. Thus, a well-based and comprehensive emergency care program is a basic part of a highway safety infrastructure. While the types of strategies that are included here are often thought of as simply support services, they can be critical to the success of a comprehensive highway safety program. Therefore, for this emphasis area, an effort should be made to determine if there are improvements that can be made to this aspect of the system, especially for programs that are focused upon location-specific (e.g. corridors) or area-specific (e.g., rural areas) issues. As additional guides are completed for the AASHTO plan, they may address the details regarding the design and implementation of emergency medical systems strategies. • Strategies Directed at Improving the Safety Management System: The management of the highway safety system is essential to success. There should be in place a sound organizational structure, as well as infrastructure of laws, policies, etc., to monitor, control, direct, and administer a comprehensive approach to highway safety. It is important that a comprehensive program not be limited to one jurisdiction, such as a state DOT. Local agencies are often responsible for the majority of the road system and its related safety problems. They also know what the problems are better than others do. Moreover, commercial vehicle operations are unique in that they occur under the safety management—good or bad—of a transport company. Objective 12.1 F specifically addresses the data aspect of this area. As additional guides are completed for the AASHTO plan, they may address further details regarding the design and implementation of strategies for improving safety management systems. • Strategies That Are Detailed in Other Emphasis Area Guides: Trucks operate around the clock, and the risk of certain types of crashes (e.g., fatigue and other road departure crashes) increases during the early morning hours. Falling asleep at the wheel is a particular problem for commercial drivers. Seatbelt use is a safety concern with passengers in heavy trucks just as it is with drivers of passenger vehicles. There are several strategies that can be used to alleviate the various problems. Some roadway-related strategies are briefly presented herein, but are not discussed in detail. More extensive information on applicable roadway design and operational strategies may be found in the following companion guides: – 5.1—Crashes Involving Alcohol Use – 8.1—Increasing Seatbelt Use – 15.1—Run-Off-Road Crashes – 15.2—Crashes on Highway Curves – 18.1—Head-On Crashes SECTION V—DESCRIPTIONS OF STRATEGIES V-6

SECTION V—DESCRIPTIONS OF STRATEGIES Objective 12.1 A—Reduce Fatigue-Related Crashes A major problem for many truck drivers, particularly over-the-road drivers, is finding a place to stop and rest at night, as well as for short periods during the day. Most states have existing space that could be used for additional truck parking during time periods that the space is not being used for its original intended purposes. Rest area spaces designed for passenger vehicles are usually underutilized during late night hours and could be made available to truckers seeking rest. Likewise, truck weigh stations that are not in use could be made available to truckers. Modifications to make such facilities available to truckers will not provide sufficient parking to meet the full needs of truckers, but they offer relatively inexpensive ways to quickly address the shortage of truck parking spaces. Tennessee conducted an analysis of trucks parked adjacent to the Interstate highway and involved in crashes either while parked or when moving from a parked location along the Interstate, an Interstate ramp, or a rest area ramp. Tennessee found that, although such crashes are relatively rare, they have a much higher likelihood of resulting in fatality (5.3-fold) and a somewhat higher probability of injury (1.27-fold, Wegmann and Chatterjee, 1999). Although truck parking space shortages are evident to any traveler on certain major Interstate highways, the dimensions of the problem were not established until 1996, when two reports were published documenting the severe shortage of parking spaces in both public and private facilities (Trucking Research Institute, 1996a; 1996b). The reports estimated that more than 28,000 additional parking spaces were needed nationwide. Three groups were surveyed—truck drivers at public and private rest areas, motor carrier executives, and truck stop operators. It was found that more than 90 percent of commercial drivers surveyed felt there was a shortage of parking facilities, especially for long-term or overnight parking. Drivers also expressed a preference for private facilities (truck stops) rather than public rest areas, citing both security concerns and the availability of amenities such as food and shower facilities. Motor carrier executives also believe there is a shortage of longer-term parking facilities, reporting that their drivers have to find places to park and sleep at shipper or consignee locations, shopping center parking lots, and exit and entry ramps to Interstates. However, truck stop operators are much less likely to recognize a problem. Nearly 7 in 10 respondents said there is no shortage of public rest area parking for truckers. All respondents reported regional differences, with the Northeast considered to have the greatest shortage. Cost and land availability pose major barriers to truck stop expansion. In June 1999, the FHWA conducted a rest area forum in Atlanta, Georgia. There were more than 70 attendees representing various government, industry, and driver groups. Proceedings of the conference were published in December 1999 (Hamilton, 1999). A variety of issues were addressed, including safety and security, comparisons of commercial truck stops with public rest areas, alternative parking sites, funding issues, time limits, and quality of rest parking services. Participants made a number of specific suggestions for improvement of truck rest parking, some of which are discussed below. More recently, in 2002, FHWA published two reports written in response to congressional direction to assess and analyze commercial truck parking demand and supply. In the first report, Fleger et al. (2002) conducted an inventory of public and commercial truck parking V-7

spaces on the National Highway System, developed a truck parking demand model, surveyed truck drivers on the problem, estimated parking demand, identified major parking deficiencies, and identified improvements recommended by state partnerships to mitigate the problem. In the second report, a commercial driver survey (Chen et al., 2002) revealed widespread driver dissatisfaction with many different rest parking elements. Exhibit V-2 provides a summary of major findings. SECTION V—DESCRIPTIONS OF STRATEGIES V-8 Increasing lengths of trucks plus trailers also compound the problem. Drivers of 75-foot car carriers report that they cannot fit into diagonal parking spaces designed for 45- or 53-foot trucks (Wegmann and Chatterjee, 2002). Even when they find a parking space, problems often develop. In another survey conducted by the OOIDA Foundation (1999), 15 percent of commercial driver respondents reported that they were awakened and told to drive on more than six times in the previous year. In some states, parking is limited to 2 or 3 hours, even when moving on would violate federal hours-of-service regulations. Almost three-fourths of respondents reported that this has happened to them. Furthermore, finding a space in a rest area does not necessarily allow restful sleep. Eighty-eight percent of the drivers expressed concern about robbery, with almost as many concerned about assault or theft. More than 10 percent report having been the victims of robbery in rest areas, and more than half say prostitutes have awakened them. Private truck stops and public rest areas meet different needs, and both are needed. In the FHWA survey (Chen et al., 2002), drivers exhibited a strong preference for commercial truck stops for extended rest, meals, using phones, and performing minor maintenance. Public rest areas were preferred only for quick naps (2 hours or less) and for other quick stops like stops to use vending machines. EXHIBIT V-2 Truck Driver Assessment of Rest Parking Characteristics, in Percent Almost Almost Usability Characteristic Always Frequently Sometimes Rarely Never Find available space, truck 9 25 51 12 4 stop Find available space, rest 2 9 41 34 14 area Parking convenient to 9 30 41 12 7 highway Facility has features needed 15 36 38 7 3 Parking time limits allow 15 22 30 18 15 enough time Enough room to drive in and 8 24 48 15 6 out Truck spaces used only by 9 25 34 20 12 trucks From Chen et al., 2002.

SECTION V—DESCRIPTIONS OF STRATEGIES As noted above, drivers generally prefer commercial truck stops to public rest areas. The FHWA study of the adequacy of truck parking (Fleger et al., 2002) found that, according to driver preferences, the total demands for public and commercial rest parking were 23 per- cent and 77 percent, respectively. However, their inventory found that the actual number of spaces was about 10 percent public and 90 percent commercial. Thus, relative to spaces available, demand for truck rest parking at public rest areas is far more intense than that at commercial truck stops. Clearly, the inadequate number and quality of public rest parking for trucks undermines the quality of work life and contributes to fatigue for the nation’s truck drivers. Among the many possible interventions to improve truck rest parking (Hamilton, 1999; Fleger et al., 2002; Chen et al., 2002) are the following: • Expand and improve public rest areas. • Expand and improve commercial truck stops and travel plazas. • Form public-private partnerships to support joint solutions, such as public purchase of lots adjacent to truck stops to increase capacity, or low-interest public loans to truck stops for expansion. • Use alternative parking sites, e.g., weigh stations, government facilities (e.g., fairgrounds, stadiums), and receiving and shipping facilities. • Improve parking layout to make parking easier and safer, e.g., employing more pull- through configurations. • Improve amenities, lighting, and rest area design. • Improve security at rest parking sites. • Provide better information to truck drivers about rest parking availability, including real-time information on available spaces. • Change enforcement practices, e.g., step up enforcement of parking rules to remove vehicles from ramps and shoulders, but extend time limits to allow better sleep. • Conduct additional studies to continually assess demand and operational issues. The remainder of this section describes three possible strategies for overcoming fatigue: one to increase efficiency of use of existing parking spaces, one to create additional parking spaces, and one to incorporate rumble strips into new and existing roadways. Strategy 12.1 A1: Increase Efficiency of Use of Existing Parking Spaces General Description Drivers seeking a place to stop are rarely knowledgeable about space availability until they have actually driven into a rest area. If there are no spaces available, they are likely to “make” a space by parking on the exit ramp or in areas not designated for trucks. Knowledge about space availability prior to reaching the rest area would be helpful to drivers making decisions about whether to pull off the road in a given location. V-9

Trombly (2003) reports a survey of state officials indicating that the provision of improved information on space availability to truckers is regarded as an improvement strategy of high potential effectiveness. The study suggests various ways that drivers can be better informed, including ways that real-time information on space availability can be provided. On I-95, near the Virginia border, North Carolina is using a solar-assisted changeable message sign (CMS) about a quarter mile before the North Carolina Welcome Center. This sign is turned on when the Welcome Center has no more parking spaces available. The message reads “NC Welcome Center, Truck Parking Lot Full, Proceed to Next Exits.” The CMS includes a cell phone that enables remote access. Night custodial staff is trained to observe the truck parking lot. When it begins to fill up, they activate the computer to display the preprogrammed message on the CMS. This usually occurs about 10:30 to 11:00 p.m.; it is usually deactivated around 7:30 to 8:00 a.m. SECTION V—DESCRIPTIONS OF STRATEGIES V-10 EXHIBIT V-3 Strategy Attributes for Making More Efficient Use of Existing Parking Spaces Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Truck drivers seeking a place to stop and rest. In the North Carolina pilot study of providing real-time information on availability of parking spaces, it was shown to reduce overcrowding and parking on ramps. In the 2-week time period prior to sign implementation (in May 2001), at 7:00 a.m. there was an average of 34 trucks parked at and around the Welcome Station, which has only 19 legitimate spaces available. Counts in the 40s and even 50s were not unusual. In the first year following implementation, typical counts were in the low 20s. The number of trucks at the station has risen somewhat since, but is still well below the number before installation of the sign. North Carolina maintains that some crashes have been attributed to truck parking problems around rest areas. Although there are no studies showing a direct relationship between increasing parking facilities for truckers and crash reductions, what is known about fatigue and crashes, as well as truckers’ response to increasing available parking space, indicates that such measures contribute to highway safety. Because parking facilities are often provided at state welcome centers, and because these are placed shortly after entering a state, providing timely information to drivers requires that adjoining states cooperate. This issue is of particular importance for “corridor” states, that is, states that truckers drive through on their way to their ultimate destination. Cooperative programs need to be established to enable presentation of information to drivers in time for them to make decisions about where to stop, and in corridor states, such agreements must be made for trucks entering the state and for those leaving it to enter another state. Although the North Carolina program is monitored by custodial personnel on duty at night, ideally the program would be automated, so that as parking spaces are filled, the sign would automatically be activated. Economical and practical devices to automate this function are not known to be available.

SECTION V—DESCRIPTIONS OF STRATEGIES V-11 EXHIBIT V-3 (Continued) Strategy Attributes for Making More Efficient Use of Existing Parking Spaces Attribute Description Potential Difficulties Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Issues Affecting Implementation Time When power outages occur and the sign is not operating, the pre-implementation parking problems may reappear. Power outages also require that the system be reset. Custodial staff is not generally sufficiently computer-literate to handle this chore. Providing a supervisor with the capability of accessing the CMS from a personal computer should resolve the problem. If direction is provided to private truck stops off the Interstate, it will be strongly advisable to monitor available parking spaces in these private facilities, as well as those available in public rest areas. Such monitoring may be difficult, yet it will be important information for the driver. Unless there is a need for fuel and/or food, drivers are reluctant to spend 15 to 20 minutes seeking a parking space in a private truck stop, since there may not be one when they arrive. If significant overflow truck parking persists in illegal spaces at the rest facility, following implementation, consideration should be given to coordinating with local or state law enforcement officials to ticket illegally parked vehicles. Process measures would include the extent to which the system is operated successfully by the custodial personnel and the extent to which the sign is activated and de-activated at appropriate times. To date, the North Carolina rest area parking staff has been highly reliable in performing these tasks. The number of locations at which a system is implemented is another useful measure. Safety impact measures should center upon crashes involving trucks, including total crashes, night crashes, and those in which drowsy driving was involved. Focus should be upon trucks that are parked or are entering or leaving a parked status. The latter will be difficult to acquire, because crashes involving parked trucks are relatively infrequent. Yet because of their severity, even a small reduction is important. However, to acquire sufficient data to relate improved parking information to such crash reduction requires that illicit parking be widely reduced over an extended period of time. Reduction in parking alongside the highway and on ramps, as well as reduction in illegal parking at rest areas affected by the message sign, can be used as a surrogate safety measure. If information for a welcome rest area is to be provided in time for an informed decision, cooperation with adjoining states is required. An information campaign, targeted at truck drivers, may be helpful to introduce improvements in parking. For all states, but especially for “corridor” states, Interstate agreements will be needed to provide drivers with information in time to make efficient decisions about where to seek a parking space. It would be especially useful if drivers could be given information about availability of parking spaces in private truck stops as well as public rest areas. To acquire this information in real time will require working with the private sector to ensure the timeliness and accuracy of information provided. Programs such as the North Carolina Welcome Center can require implementation periods longer than a year, especially when they are the initial trial. Although the (continued on next page)

Information on Agencies or Organizations Currently Implementing this Strategy The North Carolina Department of Transportation is implementing this strategy on I-95 at the Virginia state line. A point of contact is Ms. Jennifer Pitts, Rest Area Program Coordinator, NCDOT Roadside Environmental Unit, (919) 733-2920, jpitts@dot.state.nc.us. Strategy 12.1 A2: Create Additional Parking Spaces General Description In some places, even more efficient use of existing parking spaces will not solve the problem. There simply may not be enough parking spaces to meet the demand. For example, a study conducted by the University of Tennessee (Chatterjee and Wegmann, 2000) observed 1,224 large trucks parked at night along Tennessee’s Interstates and found that 470 (38 percent) were parked on ramps, shoulders, or other spaces not designated or intended for truck parking. More spaces need to be created. As presented earlier in this chapter, a survey of drivers conducted by Chen et al. (2002) for the FHWA found that 48 percent of respondents felt that parking is rarely or almost never available at public rest areas. Only 9 percent of respondents reported being able to “almost always” or “frequently” find spaces in public rest areas. Trombly (2003) conducted a survey of state highway officials and found that expansion of existing rest areas was rated as one of improvement strategies with the highest potential for effectiveness. Tennessee’s studies indicated a need for nearly 1,500 spaces on Tennessee’s Interstates (Wegmann and Chatterjee, 1999). Kentucky determined that it needed more than 700 more parking spaces for truckers. Some modifications can be made to existing rest SECTION V—DESCRIPTIONS OF STRATEGIES V-12 EXHIBIT V-3 (Continued) Strategy Attributes for Making More Efficient Use of Existing Parking Spaces Attribute Description Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes strategy was identified in North Carolina in early 2000, it was not acted upon until February 2001. Night and early-morning data were compiled to determine the magnitude of the problem. Existing custodial personnel had to be trained to monitor and activate the system. The program was fully implemented May 20, 2001, less than 4 months from the outset of moving on the project. The costs for changeable message systems should be minimal. In North Carolina the cost is about $25,000 to $30,000, but because the DOT already owned this equipment, it was not an extra purchase. A permanent installation would be about the same cost or maybe slightly more. Cell phone usage costs about $20 monthly. The computer cost would be about $1,200, but in this case one was already owned. No additional personnel are needed. The nighttime custodial personnel already covering the Welcome Center had to be trained to monitor the number of trucks and to use the computer to activate the sign when the spaces were filled. There do not appear to be any special legislative needs. None identified.

SECTION V—DESCRIPTIONS OF STRATEGIES stations to create more parking places in the space available, as well as allowing nighttime truck parking in spaces used by passenger vehicles during the day. Kentucky also developed the concept of “Rest Havens,” whereby modifications are made to existing truck weigh stations to create additional parking spaces. Other possibilities include adding and/or opening of pullout areas and building entirely new rest areas. Ideally, Rest Havens also include restrooms and vending machines. Clean restrooms help ensure that the overall facility is an attractive, quality location for drivers to rest. The Rest Haven concept is one that can make optimal use of available space, but for this approach to work, truck drivers must be assured that their parking there will not subject them to increased probability of inspection, thus delaying them and increasing their risk of fines and/or being placed out-of-service. V-13 EXHIBIT V-4 Strategy Attributes for Creating Additional Parking Spaces Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Truck drivers seeking a place to stop and rest. Additional parking spaces should allow for more available rest for truck drivers and thus reduce fatigue-related crashes. Although there are no studies showing a direct relationship between increasing parking facilities for truckers and crash reductions, what is known about fatigue and crashes, as well as truckers’ response to increasing available parking space, indicates that such measures contribute to both driver wellness and highway safety. Several factors will increase the efficient use of additional parking spaces for truckers. First, if possible, drivers should be informed of space availability in time for them to make a judgment about whether to stop or continue to drive. Second, drivers need to feel secure in the available parking space. Robbery, assault, and soliciting for drugs and/or prostitution are problems that truckers confront in many rest areas. Lack of security reduces the attractiveness of parking spaces. Truck drivers prefer areas where passenger vehicles are not allowed, because the common belief among truckers is that “cars bring trouble.” Kentucky is installing security cameras at Rest Havens to promote safety. Signs are displayed to indicate spaces that are reserved for trucks. Third, truckers need to feel secure from aggressive enforcement. Some states limit the length of time a trucker may stop and rest, even though requiring a trucker to move on may result in a violation of hours-of-service regulations. If truck weigh stations provide Rest Haven parking, drivers must be assured that using the spaces will not subject them to inspections they would not otherwise undergo. Publicizing Rest Haven availability can be accomplished through the trucking industry, trucking publications, and the media. Although not absolutely essential, amenities such as rest rooms and food services increase the attractiveness of rest areas. Success also depends upon support from the highest level, e.g., head of the department of transportation, for creation of Rest Havens in weigh stations. (continued on next page)

SECTION V—DESCRIPTIONS OF STRATEGIES V-14 EXHIBIT V-4 (Continued) Strategy Attributes for Creating Additional Parking Spaces Attribute Description Potential Difficulties Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Potential difficulties include the obverse of the keys to success. First there is the problem of funding. Kentucky added $2 million to the budget for new weigh stations to cover the cost of additional parking spaces, a building to house rest rooms, vending machines, and a monitor providing weather and road conditions in the area. The goal is to expand five stations in the state and create additional Rest Havens. Second, concerns about aggressive enforcement will reduce the use of available facilities. Motor vehicle enforcement personnel who staff weigh stations are likely to object to exempting truckers from inspection if they use the parking available in the Rest Haven. This opposition must be overcome, if the concept is to succeed. Third, concerns about physical assault and/or unwanted solicitations will discourage use of the Rest Haven. Also, if a driver needs food, or desires to use a rest room, the absence of these amenities may require continuing to another location. Fourth, some opposition may come from private truck stop operators who may consider the state’s providing these facilities to be undue and unnecessary competition with private industry. Process measures include the change in the number of parking spaces, or space- hours, made available to truckers. Safety impact measures should center upon crashes involving trucks, including total crashes, nighttime crashes, and crashes in which drowsy driving was involved. Focus should also be on trucks that are parked or are entering or leaving a parked status. The latter will be difficult to acquire, because crashes involving parked trucks are relatively infrequent. Yet because of their severity, even a small reduction is important. However, to acquire sufficient data to relate improved parking information to such crash reduction requires that illicit parking be widely reduced over an extended period of time. Reductions in parking alongside the highway and on ramps, as well as reductions in illegal parking at rest areas affected by the message sign, can be used as surrogate safety measures. Where Rest Havens have been constructed, feedback provided in suggestion boxes can be used. In addition to providing parking spaces, it would be beneficial to provide information to drivers concerning availability of space. It is also important that drivers know that parking in weigh stations will not subject them to an inspection that would not otherwise occur. An information campaign, targeted at truck drivers and employing trucker-oriented media or publications, may be helpful to introduce improvements in parking. Some arrangement must be made for custodial maintenance of the Rest Haven. There must be backing from the highest level for the Rest Haven concept to succeed. Motor carrier enforcement personnel, as well as the trucking industry, need to be onboard. Commercial truck stop owners and their trade associations (e.g., the National Association of Truck Stop Operators) are likely to point out that many private truck stops have unutilized spaces and that drivers should be responsible for planning ahead to park at legal locations and in accordance with hours-of-service regulations.

SECTION V—DESCRIPTIONS OF STRATEGIES Information on Agencies or Organizations Currently Implementing this Strategy Appendix 1 is a state agency profile providing more information on the Kentucky Rest Haven initiative. NCHRP Synthesis of Highway Practice 317: Dealing with Truck Parking Demands (Trombly, 2003) includes 10 specific strategies for expanding and enhancing public facilities. V-15 EXHIBIT V-4 (Continued) Strategy Attributes for Creating Additional Parking Spaces Attribute Description Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes In Kentucky, it was estimated in 1996 that they had a shortage of 733 commercial parking spaces. Using available parking areas at loadometer stations, they were able to open their first Rest Haven in 1999 and are working to create others. Funding and availability of land are prime factors in the speed with which the Rest Haven concept can be realized. Costs will vary widely, based upon the option used to provide additional space. Renovating existing rest areas or creating Rest Havens in spaces already available in weigh stations will be relatively inexpensive. Providing new rest areas will be costly. Virginia estimates that its cost of creating a new rest area solely for trucks comes to about $51,000 per building. However, Kentucky estimates a cost of $10,000 per parking space in its Rest Havens. Kentucky also provides a 1,000-square-foot rest room/vending building with utilities that cost $150,000 each. Where these buildings are not yet constructed, “porta- potties” are provided. There should be no special training required for this strategy. However, an understanding and appreciation of the role of fatigue in truck crashes may generate more enthusiastic involvement than would otherwise be the case. Implementing this strategy will primarily require the skills and expertise of existing personnel in departments of transportation. However, there will be a need for custodial care of the new facilities. There do not appear to be any special legislative needs, although legislative appropriations may be required for funding. The issue has been raised as to whether states should assume responsibility for providing adequate parking spaces for truckers. It is argued that truckers know where to find a place to park, and if they experienced severe consequences for parking on the roadside, they would stop parking there. It is not the state’s responsibility to make life great for the trucking industry. Bigger companies often make arrangements for places to park their trucks. They have regular routes, know who their customers are, and know how/where their drivers can sleep. The opposing line of reasoning is that while this may be true for large operations, by far the majority of motor carriers have fewer than 10 trucks, with most having fewer than five. Simply ticketing drivers is problematic, in that the shortage of resting space arises from a wide range of sources, including shippers, motor carriers, and the competition for loads. Ticketing drivers and moving them on may create public safety problems.

Strategy 12.1 A3: Incorporate Rumble Strips into New and Existing Roadways General Description Factors influencing driver alertness include time of day (i.e., “circadian troughs” like 3:00 to 5:00 a.m.), amount of sleep, and time on task (hours driving) (Wylie et al., 1996). Many commercial drivers obtain far less sleep than desired for full alertness (Wylie et al., 1996). One of the signature effects of reduced alertness is a deterioration of lane keeping (Wierwille et al., 1994). Thus, single vehicle run-off-the-road crashes are typical of drivers who fall asleep at the wheel. Head-on crashes may also result in two-lane and multilane undivided highways. Rumble strips have been shown to be effective in reducing this risk on Interstate highways and are now being pilot tested on other types of roads. Although most fatigue-related crashes involve drivers of passenger vehicles, truck drivers may be especially benefited because of their long hours of driving and the necessity of frequent nighttime driving. The use of rumble strips is discussed at length in the guides addressing run-off-road crashes (Strategy 15.1) and head-on collisions (Strategy 18.1). Objective 12.1 B—Strengthen CDL Program In 1986, Congress enacted legislation providing for the commercial driver’s license (CDL), that is, a special license for drivers operating large vehicles, transporting more than 15 passengers, or carrying hazardous materials. The legislation established mandatory federal standards for state licensing programs. These standards required that the driver hold only one license and meet certain minimum standards for testing and licensing. This legislation was implemented over a period of years, so that all drivers of the affected vehicles were covered by spring 1992. The Motor Carrier Safety Improvement Act of 1999 strengthened the earlier legislation and established a new federal agency, the Federal Motor Carrier Safety Administration, to administer the program. Even with strong federal legislation mandating CDL requirements, there remain major problems with the program. Not all states comply with all of the provisions of the CDL, especially when it comes to Interstate reporting of infractions (FHWA, 1999a). Another serious concern is the fraudulent issuing of licenses, either by state examiners or by third- party testers. Because a CDL is, in effect, a license to hold a job, both drivers and motor carriers may resort to extreme measures. Surveys conducted by the American Association of Motor Vehicle Administrators (AAMVA) in 2000 also found that some states routinely disregard some CDL program requirements. Of 10 states visited, 5 reported occurrences of not disqualifying commercial drivers due to convictions received through the Commercial Driver License Information System (CDLIS). Seven of the 10 states visited and 15 other states that responded to AAMVA reported that their courts enable drivers to avoid disqualification that technically should be implemented, through special licenses or permits to operate commercial motor vehicles (CMVs). The surveys uncovered other major problems as well. Public health and safety in every state are placed at risk if even a few states fail to meet federal CDL requirements. This is because a CDL from one state allows a driver to operate in any other, and heavy trucks typically operate across state lines. SECTION V—DESCRIPTIONS OF STRATEGIES V-16

SECTION V—DESCRIPTIONS OF STRATEGIES Some states conduct careful monitoring of CDL testing, whether by third parties or by their own examiners. Such monitoring can reduce the incidence of licensing unqualified drivers. Failure of states to fully implement all elements of the CDL act could explain the negative results of a California study by Hagge and Romanowicz (1996). Using intervention time- series analyses, these authors found no evidence that California’s implementation of the major provisions of the act in 1989 had any effect on subsequent heavy-vehicle crash rates. However, not all of the provisions were implemented, many drivers received waivers, and the commercial driver road test used by California was found to be unreliable in a parallel study by Clark (1995). The need to strengthen the CDL program has been duly recognized at the state and federal levels and in 2002 resulted in a joint effort by AAMVA and FMCSA to reduce licensing and testing fraud by implementing a comprehensive set of 14 initiatives. The strategies embedded in these initiatives embrace and go well beyond the scope of the strategies presented in this guide. As such, an administrative mechanism has already been established to develop and implement the CDL enhancements proposed in this guide. The 14 initiatives are summarized below, along with their respective AAMVA staff contacts. 1. Develop and implement a comprehensive fraudulent document recognition training program for driver licensing and law enforcement personnel. Contact is Brett Robinson, 703-908-2808, brobinson@aamva.org. 2. Complete and distribute a model program of uniform identification procedures for issuing driver licensing and identification documents. Contact is Harold Kocken, 703- 908-5774, hkocken@aamva.org. 3. Convene an international symposium on driver licensing fraud and document identification (completed). 4. Develop and maintain a manual setting forth “best practices” policies and guidelines for improving CDL competency assessment, driver control, and fraud detection. Contact is Kevin Lewis, 703-908-2823, klewis@aamva.org. 5. Develop and maintain an accessible online database and library for law enforcement and issuing agencies containing samples and exemplars of valid and counterfeit identification documents. Contact is Harold Kocken, 703-908-5774, hkocken@aamva.org. 6. Establish an electronic warning system for reporting incidents of licensing and identification fraud within jurisdictions. Contact is Harold Kocken, 703-908-5774, hkocken@aamva.org. 7. Develop an online DMV verification system to enable private-sector and governmental entities to verify the validity of state-issued driver’s licenses and identification cards. Contact is Randy Holleger, 703-908-2844, rholleger@aamva.org. 8. Perform a pilot evaluation of a digitized driver licensing/identification image exchange program. Contact is Randy Holleger, 703-908-2844, rholleger@aamva.org. 9. Expand the computerized Commercial Driver License Information System (CDLIS) to provide Interstate license status and driving record history message exchanges among all V-17

SECTION V—DESCRIPTIONS OF STRATEGIES V-18 states. This capability currently exists for all CDL holders, but not for non-CDL truck drivers in some states. Contact is Randy Holleger, 703-908-2844, rholleger@aamva.org. 10. Develop an online verification system that interfaces “vital event” records (Immigration and Naturalization Services [INS], birth certificates, etc.). Contact is Randy Holleger, 703- 908-2844, rholleger@aamva.org. 11. Identify a “best method” for implementing a uniform state digital certification procedure for use in certifying and validating documents and authenticating the issuing agency. Contact is Rich Carter, 703-908-8296, rcarter@aamva.org. 12. Establish a document security laboratory staffed by persons with expertise in licensing procedures relating to document security, maintenance, and machine-readable verification technologies. Contact is Rich Carter, 703-908-8296, rcarter@aamva.org. 13. Determine the minimum biometric identifiers for licensing and identification and define the operational requirements of individual matching. Contact is Rich Carter, 703-908- 8296, rcarter@aamva.org. 14. Create and implement a new and enhanced driver license agreement (DLA) setting forth specific state procedures and policies relating to licensing, record maintenance, and monitoring of all drivers. Contact is Brett Robinson, 703-908-2808, brobinson@aamva.org. The above initiatives are in various stages of progress, and some are not due for completion until 2005. Initiative 4 has the most relevance to the present implementation guide because it directly relates to driver competency assessment and the objective of reducing fraud among DMV and third-party testers. In fact, the development of a “model program” or “best practices” manual addresses the same problem and strategy that are outlined in the following two strategies (12.1 B1 and 12.1 B2). Strategy 12.1 B1: Improve Test Administration for the CDL General Description The knowledge test for CDL is lengthy and must cover many different areas or domains. Knowledge test construction is extremely difficult, so that usually only a few versions of a test in hard copy exist. However, printed versions of the test can fall into the hands of potential applicants. There is no way to totally prevent this from happening, but some states are moving toward computerized administration of knowledge tests, with randomized selection of questions, as well as randomized answer choices, so that no two applicants are likely to receive the same test. Even when not all licensing offices are equipped with computer test administration capability so that some tests must be offered in paper and pencil format, the computerized capabilities may allow the examiner to print out hard copies, each of which is a unique combination of test items and answer choices. The AAMVA has developed computerized testing software and a pool of over 600 test items for state use in testing CDL drivers. This system, which is currently in use or under consideration in many states, can develop unique tests of randomly selected items for each applicant. States that do not have automated testing capability can use the software to generate paper tests that are changed every several weeks.

SECTION V—DESCRIPTIONS OF STRATEGIES V-19 (continued on next page) EXHIBIT V-5 Strategy Attributes for Improving Test Administration for the CDL Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Issues Affecting Implementation Time Applicants for CDL. The administration of the knowledge test, using randomized items with randomized answer choices, should reduce advance knowledge of specific test items. Since both items and answer choices will differ from one test to another and over time, memorizing items for later recall would be ineffective. Computerized creation of tests and administration of tests precludes copies of the test from being removed from the licensing office. In the case of stations that are not equipped to provide computerized testing, randomized copies of the test can be printed ahead of time to be used in those stations. Every test will be different, making it difficult for truck driver training schools to “teach the test.” The computerized test is scored automatically and provides immediate feedback on incorrect answers, thus making the testing a potentially more effective learning situation. When an applicant reaches the point that he or she has failed or has answered enough questions correctly, the test is terminated. Because hard copies of the test are generated for manual administration in stations without computerized capability, there is the possibility of questions becoming publicly available. It is important that test security be maintained. Start-up costs could be a problem for some states. Process measures would include the number, rate, and percentage of applicants processed using the improved techniques of administration. Reductions in examiner time involved in testing can also be measured. It will not be feasible to measure this strategy’s effect on crash experience, as there are many other factors more directly involved in a crash. It is possible that one could measure the recidivism rate (both citations and crashes) of drivers taking the computerized exam versus those taking the paper-and-pencil version, but it is likely that these applicants differ in other ways as well. Surrogate impact measures include length of time taken to complete the computer- administered test compared with previous paper-and-pencil administration; the ease of test administration, that is, how well the programmed materials work; and the initial failure rate immediately following implementation of the computerized test administration. If copies of the test were previously available, it may be anticipated that failure rates would increase when the test administered is not the same as the previous versions. Good computer hardware and software are necessary, as well as computer support services. None identified. The availability and costs of the necessary hardware and software, as well as the computer support required to implement and maintain the system, will affect the time required to implement.

Information on Agencies or Organizations Currently Implementing this Strategy Georgia is in the process of implementing this strategy, and Pennsylvania, Delaware, and New York are using it statewide. A point of contact in Georgia is Mickey Rawls (678-413- 8495, mrawls@dmvs.ga.gov). The Pennsylvania point of contact is Pam Gabriel (717-705- 2418, pgabriel@state.pa.us). A point of contact for AAMVA is Kevin Lewis (703-908-2823, klewis@aamva.org). California has been considering this strategy but has not yet adopted it due to operational and budgetary constraints. Instead, it uses three equivalent test forms developed from a large item pool and randomizes the sequence of items on each printed form quarterly. It is believed that this reduces, but does not eliminate, the problem of applicants memorizing specific test items. Points of contact for California are Martha Boudreau (916-657-8267, mboudreau@dmv.ca.gov) and Robert Hagge (916-657-7030, rhagge@dmv.ca.gov). Strategy 12.1 B2: Increase Fraud Detection of State and Third-Party Testers General Description Possession of a CDL is essential to hold certain kinds of employment. As such, it is highly prized, and some drivers, as well as some motor carriers, are willing to pay premium prices to see that CDLs are issued. According to AAMVA (1999a), most states have provisions for third-party testers. Many states do not cap the fee that may be charged, and average fees may run as high as $150. Given the importance of the CDL and the amount of money involved, there is room for fraudulent issuance of licenses in the absence of quality control (FMCSA, 2000b). Federal regulations require that third-party testers be audited at least once a year, but fraudulent issuance of licenses has been detected in several states. Not all states are adhering to federal requirements. To ensure the integrity of the licensing process, states need to have in place procedures that closely monitor both state examiners and third-party testers. Although there are no state or national statistics quantifying the magnitude of the problem, the documented cases and anecdotal evidence suggest that fraud involving third-party testers and/or DMV personnel is cause for concern. In Pennsylvania, for example, a large number of third-party test providers were detected accepting bribes in connection with the training, testing, and licensing of truck drivers. The problem required the state to retest over 1,000 CDL drivers in order to certify their competency and withdraw licensure from those SECTION V—DESCRIPTIONS OF STRATEGIES V-20 EXHIBIT V-5 (Continued) Strategy Attributes for Improving Test Administration for the CDL Attribute Description Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes Costs include the hardware and software required, as well as the cost of computer support personnel and training. Some training of driver license personnel will be required. There do not appear to be any special legislative needs. None identified.

SECTION V—DESCRIPTIONS OF STRATEGIES who could not legitimately pass the skill and road test. In response, Pennsylvania instituted a number of truck safety initiatives to reduce third-party testing fraud. The program consists of the following strategies: • Improved on-site auditing procedures, • Computer analysis of transactions to identify suspicious or statistically deviant third- party providers and testers (i.e., analysis of test scores, failure rates, number of tests per day, etc.), • Covert surveillance, and • The requirement that third-party testers post a performance bond to cover the cost of any retesting. The covert surveillance is performed by a private firm contracted by the state. An innovative feature of the covert surveillance is the use of undercover operatives posing as applicants and testing observers in follow-up vehicles. The authority to use covert observation is explicitly stated in the contractual certification of each third-party tester. As noted above, AAMVA and FMCSA have been working to develop a “best practices manual” to assist the states in detecting and controlling fraud among both third-party and DMV test administrators. This manual is not yet published or otherwise available, but it is possible to outline here some of the concepts and issues that ultimately need to be addressed in developing a comprehensive program of fraud reduction and test quality control. Test Auditing Standards. The use of covert observation as in the Pennsylvania program is a highly innovative approach to fraud detection that is promising. However, its expense and operational complexity might limit its use in some states. It therefore should be considered as one element of a comprehensive auditing model, as it is in Pennsylvania. The current federal audit regulations require that each certified third-party tester be audited once a year. The regulations do not specify the technical requirements of the audit with respect to sample size (number of tests reviewed) or the method of detecting statistical and psychometric aberrations in the audited tests or in other indices that might be indicative of fraudulent or sloppy test procedures. California operates an employer testing program (ETP), since only third-party testers affiliated with a commercial driver employer are certified to give the test to their applicants. Approximately 1,000 ETP providers operate in California and roughly 7 percent of all commercial drivers are tested by ETP providers. These ETP providers frequently offer training, which means that the trainer and test examiner are not always independent and can be the same person. Pennsylvania licenses 60 percent of its truck drivers through third- party testers, but the third-party testers do not have to be employers; most of Pennsylvania’s third-party testers also provide the training. In most disciplines, it is normally not desirable for training and competency-certification to be performed by the same individual because the trainer, in a sense, ends up evaluating his or her own training and may also have an interest in pleasing the client. However, this tendency might be lessened when the training and testing are done by an employer because the employer has a strong vested and corporate interest in not employing incompetent drivers. V-21

California audits each ETP provider once a year, and roughly 10 percent of the test score sheets are randomly selected for inspection in each audit. Over the past 3 years, approximately 32 sanctions per year have been taken against ETP providers, but there have been no instances of legally verifiable fraud. (However, there have been fraud cases involving DMV personnel.) Like most states, California has no quantitative management information system or statistical quality control database for quickly detecting scoring variations and patterns among either DMV examiners or ETP providers/examiners. California, however, has performed periodic psychometric reliability and validity evaluations of its passenger vehicle road test. The California commercial driver skill and road tests are patterned after an objective route sampling and scoring procedure used for the passenger skill and road test and on the so-called Essex test described by Mackie et al. (1989). Since this approach has been found to be valid and reliable, the DMV believes that the current commercial road test is psychometrically sound. However, generalizing this deduction to tests administered by non-DMV staff involves a number of inferences and assumptions. California has performed statistical analyses of the comparative crash rates of ETP-licensed and DMV-licensed commercial drivers (Chapman, 2003). It was found that ETP-tested drivers had 22 percent more fatal/injury crashes during the 2-year period following testing—a difference that was highly significant statistically (p < 0.001). However, Chapman properly cautions that the difference could be attributable to a number of confounding variables that were not available for analysis, most notably driving mileage and type of exposure. The above analysis was made possible by the availability of the necessary data elements on the California driver record file, and the approach could be extended to examine test scoring indices and patterns among DMV and third-party testers if all test results were computerized and made linkable to individual driver records. In addition to identifying providers and testers with suspect score patterns, the existence of such a system could provide some deterrence against shoddy and fraudulent test practices. The major requirements for such a system would be converting test results to electronic format and establishing a common identification linkage across record systems. It is possible to tabulate test volume and fail rate counts without linking data elements to individual driver records. However, most states do not currently maintain test score information that can be easily tabulated by individual state examiners or third-party providers. The 1997 survey by AAMVA (1999a) reported that the majority of states were unable to provide counts of either the volume or failure rate of CDL road tests administered by either state or third-party providers. Only 15 states could supply these statistics for CDL road tests administered by the state, and only 9 reported these statistics for third-party tests. The absence of such information limits the ability to (1) identify statistically aberrant state and third-party testers; (2) establish statistical baselines for use in assessing the effects of improved auditing procedures; and (3) implement statistical quality control systems for monitoring CDL road testing on an ongoing basis. One of the strategies in the Pennsylvania fraud-detection program is to inspect data on the test scores, failure rates, and daily number of tested commercial drivers. This information is currently accessible electronically, and the potential exists for developing a statistical SECTION V—DESCRIPTIONS OF STRATEGIES V-22

SECTION V—DESCRIPTIONS OF STRATEGIES management information system that would tabulate these statistics by area, month of testing, individual examiner, third provider, and organization (state versus private). A final type of auditing that warrants serious consideration is for states to retest random or representative samples of third-party-certified drivers. In fact, such authorization is authorized in title 49 part 383.75 of the federal regulations and in California is explicitly authorized in the state’s written agreement with each third-party employer. The AAMVA survey of state practices reported that 18 states randomly retest third-party-certified commercial drivers and six states randomly retest samples of state-tested commercial drivers (AAMVA, 1999a). The six states that randomly retest both state- and third-party- tested drivers are Colorado; Washington, D.C.; Kansas; Kentucky; Mississippi; and Montana. If one can assume that the retest is reliable and valid, the retest policy provides a powerful mechanism for estimating the proportion of third-party and state-tested licensees who should have been failed. Although this approach would not be able to identify many individual incidents of fraud, it does address the more general question of the adequacy of a state’s commercial driver licensing program. Moreover, in states where retesting is used for both state and third-party licensees, a comparison of retest failure rates provides information on whether the two test systems are approximately equivalent in terms of competency assessment. A substantial or increasing incidence of fraud would be among the factors causing a higher incidence of retest failures. Third-party providers with suspect retest failure rates and individual retest failure cases would then be subject to additional review for evidence of fraud. An analysis of retest failure rates over time also provides criteria for evaluating the effects of fraud-prevention strategies. Two practical disadvantages of the above approach must be acknowledged. The first is cost. Commercial driver road tests are expensive. However, samples as small as 200 would provide relatively stable estimates and would not be cost-prohibitive for many states. A second problem is negative reactions of drivers to the inconvenience of being tested a second time. Driver Competency Measurement. The detection and elimination of fraud in licensing commercial drivers is a valid goal in and of itself, apart from any nexus with public safety. As a safety initiative, however, it is anchored in the implicit premise that the skill and road test are reliable and valid indicators of driver competency. Guidelines for constructing reliable and valid tests are described by McKnight (1999), and applications to a state DMV program are described by Clark (1995) and Peck (1996). A model commercial driver testing program based on these principles is described in a study commissioned by AAMVA (Mackie et al, 1989). Whether competency assessment is done by state examiners or by third parties, the test should be based on sound psychometric principles and there should be periodic checks of two types of test reliability (internal consistency or “sampling” reliability and inter-rater reliability). Most states periodically review state and third-party exam score sheets but do not include a formal analysis of the reliability indices described by McKnight (1999). As noted above, a state’s assessment of the quality of a third-party tester assumes that the test standards used by the state are reliable and valid. The commercial skill and road test used by many states is based on the Essex model or subsequent variations of that test. To the extent that a state uses this validated model, it is permissible to conclude that the test has adequate sampling reliability based on a simple V-23

content analysis and comparison with the model. However, this approach is much more tenuous in assessing the inter-rater reliability of the state test. This latter reliability is heavily dependent on the adequacy of examiner training and periodic examiner monitoring. These exigencies are likely to become more critical in generalizing reliability estimates to third- party testers and tests given by trainers. State Implementation. The intent here is to present some of the elements that should be considered in developing a “best practices” manual for state use in strengthening the CDL through improved testing and fraud detection. Cost and operational feasibility are among the factors to consider in implementing these strategies. SECTION V—DESCRIPTIONS OF STRATEGIES V-24 EXHIBIT V-6 Strategy Attributes for Increasing Fraud Detection of State and Third-Party Testers Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties State examiners and third-party testers using careless or fraudulent methods that assist drivers in meeting CDL requirements. Effective monitoring of third-party testers should reduce CDL issuance to applicants who do not meet CDL requirements. A strong program of auditing also reduces the need for re-testing of applicants who were not adequately tested originally. However, no formal evaluations are available. There must be a system for regular review and auditing of both third-party testers and state examiners. Undercover testing may be conducted in which an applicant deliberately performs poorly, e.g., crossing the center line several times. The performance is videotaped by another undercover person following in another vehicle, so that if the candidate is passed, there is evidence that he should have been failed. A statistical reporting system should be established to provide test scores or failure rates by individual examiners and third-party organizations. The reporting system should have the capacity to tabulate scores by month and provide periodic management information reports. Candidate examiners must be thoroughly evaluated, including a criminal history check and a driver history check. In the case of rejection, a candidate examiner has a right to appeal (e.g., if a criminal offense occurred years ago in one’s youth and there is evidence that the person has led an exemplary life since then). Examiners should be recertified annually. To ensure that there will be third-party testers available in all parts of the state, new testers are solicited through a statewide bulletin akin to the Federal Register. The areas where testers are needed are identified, and testers are added only in those areas. Prior to program implementation, even though no new legislation may be required, the state legislative leadership should be briefed to ensure their understanding of the new program and why it was being implemented. When improper or fraudulent testing is detected, it is essential that all applicants who have been issued a license by the testers involved be recalled for re-testing. Because of the expense involved, third-party testers are required to post bond.

SECTION V—DESCRIPTIONS OF STRATEGIES V-25 EXHIBIT V-6 (Continued) Strategy Attributes for Increasing Fraud Detection of Third-Party Testers Attribute Description Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Issues Affecting Implementation Time Where a performance bond is required, bond companies may have no experience with such a program. A bond requirement may also cause some potential third-party testers to avoid, or withdraw from, the program. Unlike state examiners, third-party testers charge fees that may be as much as $150 or more. Because the candidate is paying so much, the tester may feel that a license should be issued even if the performance is below standard. This tendency to want to pass a candidate, because of the high cost, must be resisted by the third-party tester. Although there is no charge for the exam if a state examiner conducts it, candidates may prefer a third-party tester for two reasons. First, there is no delay in scheduling a test. Second, unlike the state, the third-party tester can usually rent a vehicle to the applicant. Process measures would include the number and type of changes made in the program, as well as the number and percent of applicants processed using the improved techniques of administration. Safety performance measures: It will not be feasible to measure the impact of this strategy on crash experience, as there are so many other factors more directly involved in a crash. Surrogate impact measures can include the rate of detection of fraudulent license issuance and the changes in this rate as tighter monitoring is implemented. Frequency of two types of problems should be monitored, namely, sloppy testing and fraudulent testing. Sloppy testing is when a candidate performs poorly but is still issued a license. Fraudulent testing is when portions of the testing are omitted entirely. In the case of sloppy testing, the testing privilege may be suspended until the third-party tester can provide an acceptable management plan for quality control. Fraudulent testing should lead to dismissal from the program, as well as payment for re-testing of all candidates who have gone through the program. Measuring the number of re-tests required is also of interest. Tracking pass/fail rates can also identify examiners who may be passing or failing too many applicants. Also, a high number of tests administered in a day can identify fraudulent testing, since a valid test takes significant time. None identified. The program would usually be administered through a DMV Bureau of Driver Licensing and requires no participation from other agencies. However, it may be appropriate to inform legislative leadership about what is being done and why. In Pennsylvania, it took about a year and a half to develop new procedures and put them in place. However, there had been a prior system in place that divided the state into three sectors, each handled by a project manager. The project manager trained examiners, managed the program, and conducted audits. Under the new system, there is one contract manager who conducts the training. If the covert testing is to be conducted by a private organization, it will probably be necessary to go through the state bidding process, necessitating delay in implementation. (continued on next page)

Information on Agencies or Organizations Currently Implementing this Strategy The information provided is based particularly on Pennsylvania’s program for ensuring quality control for CDL testing. A contact in Pennsylvania is Joy Gross (717-787-9930, joygross@state.pa.us). The information has also been influenced by joint in-progress activities of AAMVA and FMCSA as described above. A contact for AAMVA is Kevin Lewis (703-908-2823, klewis@aamva.org), and a contact for FMCSA is Bob Redmond (202-366-5014, Robert.Redmond@fmcsa.dot.gov). A contact for the California research and development program is Robert Hagge (916-657-7030, rhagge@dmv.ca.gov). A contact for California’s employer test program (ETP) is Linda Stanley (916-657-8667, lstanley@dmv.ca.gov). Objective 12.1 C—Increase Knowledge on Sharing the Road About 85 percent of vehicle occupant fatalities resulting from large-truck crashes occur not in the truck but rather in other vehicles involved in the crash (FMCSA, 2003a). Analysis of driver-related factors in crashes between large trucks and passenger vehicles indicates that SECTION V—DESCRIPTIONS OF STRATEGIES V-26 EXHIBIT V-6 (Continued) Strategy Attributes for Increasing Fraud Detection of Third-Party Testers Attribute Description Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes With the implementation of the new program in Pennsylvania, state costs were actually reduced, in that three project managers were replaced with one. Auditing of third-party testers is now conducted by about 150 state examiners, but no one examiner audits the same site for more than 2 years to prevent close relationships from developing. With this new system, the rate of tester auditing increased from once a year to twice a year. Private-sector, third-party testers must provide a performance bond. If it is determined that persons they tested must be re-tested, the cost of re-testing must be covered. The amount of the performance bond is based upon the number of tests given. In Pennsylvania, it ranges from $100,000 to $250,000. There are normally no state costs for the third-party testing. With the implementation of the new system, all examiners, both state and those with third-party testers, should be re-certified. Examiners should be given in-service training and re-certified on a yearly basis to ensure quality control. If, through an audit, an examiner is identified as having a problem, that examiner must go through retraining before conducting any further exams. All examiners and third- party testers should be assembled annually to discuss any questions or issues of concern and to receive additional training. Legislation may be needed to adopt the federal CDL requirements as a whole if they are not already adopted. It is advisable that key legislators be briefed about the program, including what is to be done, how, and why. None identified.

SECTION V—DESCRIPTIONS OF STRATEGIES passenger vehicle driver errors or other driver factors are cited in more than two-thirds of these crashes, whereas truck driver errors are cited in less than one-third (Craft and Blower, 2003b; FHWA, 1999c; Blower, 1999). Passenger vehicle driver behaviors such as improper lane changes have been found to be among the highest-risk behaviors around trucks (Kostyniuk et al., 2002). Using a fleet of instrumented trucks, Hanowski et al. (2001) observed and analyzed 210 critical incidents (driver errors resulting in potentially unsafe conditions) involving the interaction of a large truck and light vehicle. The most common errors were lane changes without sufficient gaps, entrance onto the roadway without adequate clearance to the trailing truck, left turns without adequate clearance to the trailing truck, and late braking for stopped or stopping traffic. More than three-quarters of such incidents were attributed to drivers of light vehicles in the vicinity of trucks, rather than to truck drivers. Clearly, the driving public needs to improve its driving practices in the vicinity of large trucks. Drivers of passenger vehicles are making most of the errors leading to crashes and fatal crashes, and they are suffering most of the injury consequences. The above findings indicate a need for a broad-based public understanding of the hazards associated with driving too close to heavy trucks. Public awareness and understanding of this issue is critical to heavy-truck safety, although it may take years to see significant improvements. In many ways, the task is analogous to the challenges associated with achieving public understanding and acceptance of safety belt effectiveness and the hazards associated with drinking and driving. It was not until there was sufficient public understanding of these issues that meaningful measures could be implemented. The effectiveness of these measures accounts for the major gains that have been made in highway safety over the past quarter century. In the same way, public understanding of the need to drive carefully in the vicinity of large trucks must be achieved. Although it will take time, the sooner it is initiated, the sooner the heavy-truck crash toll may be reduced. Until recently, FMCSA outreach and other public information programs on sharing the road with heavy trucks have emphasized the “No Zone” concept. Like passenger vehicle drivers, truck drivers have blind spots in the rear and sides of their vehicle, but for trucks, these blind spots are much larger. It is especially hazardous for other vehicles to position their vehicles in the truck driver’s blind spots, as well as in the area immediately in front of the truck. Truck maneuverability and stopping capabilities are of course much less than those of smaller vehicles. When a collision does occur, the massive size of large trucks compared with light vehicles (up to 25 times more mass) creates a huge injury potential for drivers and occupants of the smaller vehicles. Understanding the “No-Zone” is a fundamental step for safer driving by motorists operating in the vicinity of large trucks. However, in recent years, FMCSA and its partners have shifted the focus of their program to the broader concept of “Share the Road.” The intent is to target all road users and to increase public awareness of a broader range of potentially hazardous interactions among large and small vehicles. These include tailgating trucks and cutting in front of trucks when passing. FMCSA plans to implement a Share the Road pilot in one or more states, combining targeted, multimedia outreach and intensive enforcement of traffic laws prohibiting unsafe light-vehicle maneuvers around trucks such as cutting in front. This program is expected to become a centerpiece of the FMCSA Share the Road program. V-27

Strategy 12.1 C1: Incorporate Share the Road Information into Driver Materials General Description The Share the Road Coalition (www.sharetheroadsafely.org) has been established as a public-private partnership of organizations embracing the goal of reducing crashes and fatal crashes involving large trucks and other vehicles. The coalition consists of FMCSA, NHTSA, AAMVA, the American Driver and Traffic Safety Education Association (ADTSEA), state and local governments, law enforcement, motor carriers, industry trade associations, insurance companies, and highway safety organizations. FMCSA public information and education programs have spearheaded the initiative. An important first step is to incorporate information into state driver handbooks and knowledge tests for both passenger vehicle operators and CDL licensure. Most states have already done this to some degree, but many have not. State driver handbooks are read most often by young beginning drivers and, less frequently, by applicants who already hold a license from another state. Renewal applicants are not usually required to take a knowledge test, and even if they are, most do not read the handbook. Nevertheless, the handbook is an important place to start in disseminating information on sharing the road with trucks and avoiding the dangerous areas in the blind spots and directly in front of the truck. ADTSEA is the professional association that represents traffic safety educators throughout the United States and abroad. The ADTSEA Web site (http://adtsea.iup.edu) contains a model curriculum for novice drivers that includes extensive lesson plans and supporting slides. These are available free in Adobe (pdf) format, and more usable formats (such as PowerPoint slides) may be purchased. Unit 8 of the model ADTSEA curriculum includes limited information on driving safety around trucks. Topics mentioned include the possibility of truck driver fatigue, truck wide right turns, side blind spots and safe passing, other “No-Zone” areas around trucks, and the importance of being able to see the truck driver in the truck’s mirrors (so that he or she can also see you). More elaborate instruction on sharing the road with trucks is under development by AAMVA and will be incorporated into the ADTSEA curriculum as it is completed. Because most renewal applicants will not read the driver handbook or take a renewal exam, they must be reached through other channels. One approach is through the renewal notice sent by the licensing authority. This mailing presents an opportunity to alert drivers on how to maneuver safely around large trucks. Brochures or other relevant material may be included in the renewal notice package to reach this audience. Another approach is to include the information in driver safety schools operated in conjunction with courts, where offenders are required to attend courses on driver improvement. An approach under consideration by FMCSA for a major pilot program is to disseminate educational brochures through rental car agencies. SECTION V—DESCRIPTIONS OF STRATEGIES V-28

SECTION V—DESCRIPTIONS OF STRATEGIES V-29 EXHIBIT V-7 Strategy Attributes for Increasing Share the Road Knowledge Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes This strategy targets the general public, including all road users, but especially drivers in the presence of truck traffic. Effectiveness will not be immediate, but over time, as the public becomes aware of how to drive around heavy trucks, there should be a reduction in truck–passenger vehicle crashes caused by passenger vehicles engaging in hazardous driving behaviors in the vicinity of trucks. Further formal evaluations of this strategy are needed. To provide good information to the public requires that there be a clear understanding of the behaviors that create hazardous car-truck interactions. A thorough analysis, preferably including state-based analysis, is an essential component of this strategy. However, as noted above, model instructional material is being developed by AAMVA and will soon be available from ADTSEA’s Web site. Driver handbooks, driver knowledge tests, and courses are revised only periodically, and such information can be incorporated only when new versions are scheduled. Process measures would include whether Share the Road information is included in the driver handbook and mailings, as well as covered on the driver knowledge tests, or in driver improvement courses. Measures may also be made of the estimated number of drivers who have been exposed to the material in some form. Surveys could be conducted to determine the extent to which the public is knowledgeable about Share the Road information. Safety impact measures would be more difficult to obtain, but impact could be measured by surveys of the public to obtain self-reports of changes in behavior based on Share the Road information. There are no known associated needs. It would be useful to work with both the state trucking association and organizations representing the public, e.g., AAA, Parents Against Tired Truckers (PATT). These groups can all review proposed changes to the driver handbook and provide useful feedback. They can also be of assistance in getting the information out to commercial drivers and other road users. Implementation time will depend upon when new versions of the driver handbook and the driver knowledge tests are scheduled to occur. Once new versions are scheduled, implementation time should be minimal. This strategy should involve relatively minimal cost. Conferring with stakeholders and making changes to the driver handbook and driver knowledge tests and courses should not require major additional funding. There will be a need for developing the handbook material and the items for the knowledge testing and courses. Use should be made of materials currently available in the public domain. None. None identified.

Information on Agencies or Organizations Currently Implementing this Strategy Information and materials are available from several sources, including the FMCSA Outreach Division (202-493-0499). Information is also available at the Share the Road Coalition Web site, www.sharetheroadsafely.org. Finally, as noted above, AAMVA is developing Share the Road driver training materials that will be available in Adobe (pdf) format on the ADTSEA Web site (http://adtsea.iup.edu). Virginia established a Large-Truck Safety Task Force in 2002 to develop recommended strategies to reduce large-truck crashes. Among the strategies recommended were to develop training resources and to seek means to educate both large-truck drivers and the general driving public about safe driving practices around trucks. A project survey indicated that 60 percent of large-truck drivers support more education for the public. The task force recommended that the Virginia Department of Education adopt No-Zone and Share the Road programs as part of the high school driver education programs. Information on the Virginia program is available from Robert L. Irving, Motor Carrier Service Operations, Virginia Department of Motor Vehicles (804-367-2865, dmvrli@dmv.state.va.us). Strategy 12.1 C2: Promulgate Share the Road Information Through Print and Electronic Media General Description Because it will take years to reach all drivers through driver-related materials, the Share the Road information should also be promulgated in multiple ways. Newspapers often have a column on safe driving or tips for the road, and this information would be relevant to its readers. Television can also include Share the Road messages in public service announcements. The message needs to come from many sources over an extended period of time, so that drivers in general will be familiar with the material. The experience with safety belt use and drinking and driving is critical to understanding the value of this strategy. A quarter of a century ago, it was generally accepted that we would never get safety belt use laws in the United States, and driving after drinking was widespread and widely considered harmless. The media played a major role in bringing about major changes in how safety belt use and drinking and driving are viewed. General public awareness of the Share the Road issue could be accomplished through similar means. If reports on truck–passenger vehicle crashes were to include this information, the public would eventually gain a deeper appreciation of this safety issue, as has been done with drinking and driving and with safety belt use. There are important differences between the Share the Road concept and safety belt use and drinking and driving, but the similarities are sufficiently strong to consider the latter as models for bringing about public behavior change. The national Share the Road initiative provides a ready source for general public information on this safety topic. At the state level, Colorado developed a customized public information campaign focusing on one rapidly growing, high-truck-crash county (Weld County in Northeastern Colorado, which includes the city of Greeley), with plans to expand the program to the entire state. The Colorado Truck Safety 2000 initiative (Olsgard et al., 2002) included a detailed crash analysis that served as the basis for the development of a multifaceted truck safety initiative, including targeted enforcement, engineering strategies, SECTION V—DESCRIPTIONS OF STRATEGIES V-30

SECTION V—DESCRIPTIONS OF STRATEGIES and public education. This program is described in more detail in the state profile accompanying this chapter. The public information and education initiative was named “Size Matters for Safe Driving in Weld County.” It included the development of printed material featuring a project logo and special graphics. These materials were printed in both English and Spanish and included brochures, information sheets, wallet-sized plastic cards, and posters. Press and media were contacted and enlisted in the campaign to spread the safety message to citizens. The Weld County program will be a template for a larger statewide initiative. V-31 EXHIBIT V-8 Strategy Attributes for Promulgating Share the Road Information Through Print and Electronic Media Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties Appropriate Measures and Data Targets for this strategy are initially print and electronic media that cover highway safety issues and ultimately the motoring public. With appropriate publicizing of Share the Road concepts and associated risks of heavy- truck crashes, the public should, over time, become aware of the problem and know more about how to share the road safely with heavy trucks. However, as in the case of safety belt use and drinking and driving, the impact of such reporting will occur over time. In truck/light-vehicle crashes, it is important that reporting does not imply culpability. Simply passing a truck, or being passed by a truck, means that the light vehicle will be in the dangerous zone at some point, and there is nothing illegal about that. However, it is important that drivers know it may be dangerous to remain for any length of time in such a relationship to a truck. Of course, it is extremely hazardous to swerve into the path of a truck traveling at high speed and assume that the truck will be able to stop as quickly as a passenger car. When such crashes result, it would be appropriate for the media to stress the dangers associated with these maneuvers. Therefore, finding means to involve and interest the media on a regular basis is another key to success. To achieve support for such reporting, it would be useful to have one or more champions in key leadership positions who would communicate the value of educating the public on how to drive safely around large trucks. It may be difficult to get the media interested in the topic. The media may also be especially reluctant to report hazardous driving on the part of the passenger vehicle, since truck–passenger vehicle crashes are more likely to produce serious or fatal injuries for the occupant(s) of the passenger vehicle. The same problem existed in regard to safety belt use. Yet, over time, the media became accustomed to reporting the information. The media must be persuaded that reporting this information should lead to a greater awareness on the part of the public and, ideally, a reduction in risky driving and resulting damage and injuries. Process measures would include standard measures of the conduct of media campaigns, such as column inches or number of articles, minutes of television time, etc. Safety impact measures would include reductions in car-truck crashes, but it should not be expected that this relationship could be definitively established. Surrogate safety measures might be generated through a survey of the public’s knowledge of both the concept of Share the Road and the possible contributing factors for specific relevant crashes in the geographic area being surveyed. (continued on next page)

Information on Agencies or Organizations Currently Implementing this Strategy Virginia and Pennsylvania are among the states that have established strategic goals to educate both passenger vehicle and truck drivers on hazardous driving behaviors around other vehicles. Virginia’s public information on Share the Road is available at www.dmv.state.va.us/webdoc/general/safety/motorcarrier. Pennsylvania has developed a Share the Road brochure that may be distributed at rest areas or through other means. It is available from Gary Modi, Chief, Safety Management Division, PENNDOT Bureau of Highway Safety and Traffic Engineering (717-783-1190, gmodi@state.pa.us). There are no known jurisdictions reporting No-Zone violations directly to the media for wider dissemination of Share the Road information to the public. The Colorado Weld County initiative, a multiagency program consisting of problem assessment, enforcement, engineering, and education strategies, is described in a state profile in Appendix 2. Points of contact in Colorado include Patricia Olsgard, Colorado Motor Carriers Association (303-433-3375, patti@cmca.com), and Stephanie Olson, CDOT Highway & Traffic Safety Division (303-757-9465, stephanie.olson@dot.state.co.us). The Colorado initiative is further documented in Appendix 2. SECTION V—DESCRIPTIONS OF STRATEGIES V-32 EXHIBIT V-8 (Continued) Strategy Attributes for Promulgating Share the Road Information Through Print and Electronic Media Attribute Description Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes Materials are available in the public domain to provide a basis for press releases and articles on Share the Road concepts. See: http://www.sharetheroadsafely.org/. It would be wise to contact the State Attorney General’s office to ensure that there would be no legal consequences of media reporting of this information. Widespread implementation is likely to take years. There were some newspaper editors who adamantly refused to report alcohol or safety belt use in crashes. Only after other newspapers were doing it routinely did such publications follow suit. Because Share the Road information is readily available on the Internet, there should not be significant costs associated with the production of materials. However, there will be a need for someone to work with the media to inform them and encourage them to address this topic. Someone, possibly in the Governor’s Highway Safety Office, will need to become familiar with the Share the Road concept and materials and devise a plan for working with the media. No legislation is required. None identified.

SECTION V—DESCRIPTIONS OF STRATEGIES Objective 12.1 D—Improve Maintenance of Heavy Trucks Vehicle maintenance is one of the most fundamental activities of commercial vehicle fleet safety management. The FMCSA and the states have various regulations and enforcement programs in place to ensure that vehicles have properly functioning equipment. Unfortunately, roadside inspection vehicle out-of-service rates for mechanical problems are quite high – generally in the range of 20 percent to 30 percent (Blower, 2002). Various studies of truck crash characteristics and causation have addressed the question of the degree to which truck mechanical problems contribute to crash involvement. Of heavy trucks involved in fatal crashes, about 3 percent are reported to have brake defects, and 1 percent are reported to have tire defects (Blower, 2002). Preliminary data from the Large-Truck Crash Causation Study indicate that about 4 percent of the sampled crashes involved a critical truck vehicle factor, such as a defective component (Craft and Blower, 2003b). On the other hand, in-depth investigations performed by Michigan’s Fatal Accident Complaint Team (FACT) have indicated that as many as 55 percent of trucks involved in fatal crashes have at least one vehicle mechanical defect and that about half of these would be sufficient to place the vehicle out of service (Blower, 2002). The extent to which vehicle mechanical defects constitute a direct causal or severity-increasing factor is difficult to assess. Nevertheless, Blower (2002) concludes that truck brake, tire, and other mechanical defects contribute “substantially” to truck crashes. In a survey of commercial truck and bus fleet safety managers and other experts, Knipling et al. (2003) found that the problems of neglect of vehicle maintenance or driver failure to inspect vehicles were not highly rated as safety problems relative to other problems. Ironically, though, fleet safety managers in the same survey rated regularly scheduled vehicle inspection and maintenance to be the most effective safety management practice of 28 practices included in the survey. Strategy 12.1 D1: Increase and Strengthen Truck Maintenance Programs and Inspection Performance General Description State truck inspection programs are largely supported by the Motor Carrier Safety Assistance Program (MCSAP). MCSAP is a federal grant program that provides financial assistance to states to reduce the number and severity of commercial vehicle crashes and incidents. The program promotes consistent, uniform, and effective commercial motor vehicle safety programs. The program helps ensure that safety defects, driver deficiencies, and unsafe motor carrier practices are detected and corrected before they result in, or contribute to, crashes. Key requirements for states to receive annual MCSAP funding include adoption and enforcement of state laws that are compatible with the Federal Motor Carrier Safety Regulations (FMCSRs) and the completion of a Commercial Vehicle Safety Plan (CVSP). Major MCSAP program elements include driver/vehicle inspections, traffic enforcement, carrier compliance reviews, public education and awareness, and data collection. The FMCSA Web site contains more information on MCSAP at http://www.fmcsa.dot.gov/ safetyprogs/mcsap.htm. V-33

FMCSR § 396.17 requires annual safety inspections of commercial vehicles, including both tractors and trailers, and roadside inspection programs are a continuous state activity to achieve adequate vehicle maintenance. Proper vehicle maintenance and a clean inspection record are considered fundamental by safety-conscious fleets. This includes compliance with federal and state requirements for pretrip, posttrip, and annual vehicle inspections. Many safety-conscious companies employ regular schedules for preventive maintenance (PM) and require their drivers to use checklists for pre- and posttrip inspections. In the safety management study by Knipling et al. (2003), “regularly scheduled vehicle inspection and maintenance” was practiced in more than 90 percent of the fleets surveyed and, as noted above, was rated by safety managers as the most effective safety management practice of 28 practices presented. Not all motor carriers voluntarily implement strong fleet maintenance programs, however. The state of Maryland has a program that meets the requirements of FMCSR § 396.17 but that also requires that carriers conduct and document an ongoing PM program for their vehicles. Enforcement officers in the state of Maryland may enter the premises of any motor carrier at any time during regular business hours to inspect equipment and also to review and copy records relating to the carrier’s PM program. This Maryland program to strengthen carrier PM programs has resulted in improved vehicle inspection performance both for vehicle inspections conducted at carrier sites and those conducted roadside. SECTION V—DESCRIPTIONS OF STRATEGIES V-34 EXHIBIT V-9 Strategy Attributes for Increasing and Strengthening Truck Inspection Programs Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Mechanically unsafe trucks operating on public highways, including those engaged in both Interstate and intrastate operations. The requirement of comprehensive carrier PM programs, in addition to annual vehicle inspections, is intended to force carriers to act proactively and systematically to ensure the mechanical safety of their vehicles. Potentially, this additional requirement will reduce mechanical defects, particularly relating to brakes and tires. The Maryland Division of State Documents publishes and distributes a PM Handbook that states the relevant Code of Maryland regulations and also extensive and very detailed inspection and other PM procedures. The effectiveness of the program is limited by the number of enforcement officers available to conduct the carrier-based inspections and PM program reviews. Measures of program effectiveness would include number of PM program reviews conducted, vehicle pass-fail rates at the carrier site, and roadside inspection out-of- service rates. If a statewide program is implemented, state vehicle out-of-service rates can be compared with levels prior to the program or with national data. For example, in Maryland, the 2003 large-truck OOS rate is 17.6 percent, compared with the national rate of 25.4 percent. None identified. The program is limited by its budget and staff allocations.

SECTION V—DESCRIPTIONS OF STRATEGIES Information on Agencies or Organizations Currently Implementing this Strategy More information on the Maryland PM program can be obtained from Administrative Officer Malcolm Rote, Maryland State Police (410-694-6116, mrote@mdsp.org). Strategy 12.1 D2: Conduct Postcrash Inspections to Identify Major Problems and Problem Conditions General Description Extensive data can be compiled on trucks and other vehicles involved in crashes, including detailed information on the tractor, the trailer, and the cargo. Information on the truck driver should include physical condition, training and experience, recent sleep history, use of drugs and medications, hours of service, company policies, trip origin and destination, restraint use, and motor carrier characteristics. Additional data can be compiled on the crash itself, including detailed information on truck condition and damage. FMCSA and NHTSA have a joint program underway to collect crash reconstructions of a large representative sample of serious heavy-truck crashes (Craft and Blower, 2002). Analyses of the collected data may reveal specific vehicle problems that appeared to be contributing to fatal crash risk (Blower, 2002). Also of interest is whether other factors, e.g., some commodities, routes, or types of carriers, result in a higher risk of crashes and should be examined for possible modifications. Michigan initiated, but later discontinued, a state program to require investigations and vehicle inspections of trucks involved in fatal crashes. This program was called the Fatal Accident Complaint Team (FACT). Pennsylvania enacted legislation in 2001 (Section 4704 of the Pennsylvania Vehicle Code) requiring that all trucks in fatal crashes undergo an MCSAP inspection. Such state programs, over time, can compile sufficient state-specific data to enhance the overall truck safety efforts. V-35 EXHIBIT V-9 (Continued) Strategy Attributes for Increasing and Strengthening Truck Inspection Programs Attribute Description Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes Available resources are a primary implementation factor. Proportional to the number of enforcement officers and carrier visits. Enforcement officers must be thoroughly trained on the details of the inspection and PM program procedures. The Maryland program is mandated by legislation specified in the Code of Maryland Regulations Title 23 (Vehicle Laws) Subtitle 3 (Preventive Maintenance Program) and Title 11 (DOT) Subtitle 22 (Motor Vehicle Administration – Preventive Maintenance Program). None.

SECTION V—DESCRIPTIONS OF STRATEGIES V-36 EXHIBIT V-10 Strategy Attributes for Conducting Postcrash Inspections to Identify Major Problems and Problem Carriers Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties Appropriate Measures and Data Associated Needs Policy and program personnel in public agencies that will benefit from details regarding unsafe trucks operating on public highways. This is an experimental strategy. No evaluations have been performed to provide estimates of effectiveness. Any consideration of this strategy should involve pilot testing, with carefully designed and executed evaluations. Support and cooperation of local enforcement agencies are essential for program success. It is these personnel who must make initial notification of a crash. Without voluntary notification, project investigators will not be able to respond, resulting in an incomplete database. In Michigan, local enforcement, at the outset, viewed the state as interfering in their jurisdiction. But they came to appreciate and support the program as it became evident that the investigators were not interfering with their activities and in many cases could contribute to their own safety programs. The Pennsylvania program was mandated in June 2001 in Section 4704(c)(2) of the state vehicle code. The measure specifies that “a qualified Commonwealth employee as designated by the department” shall perform inspections of all large trucks and transit vehicles involved in fatal crashes, as well as their drivers, before vehicle or driver is allowed to continue operations. Possible initial reluctance on the part of local enforcement, as well as reluctance on the part of state investigators to take on additional responsibilities, may have to be overcome. It is anticipated that experience with the program, if it is operated properly, will result in gaining their wholehearted support. Motor carriers may also be “lukewarm” to the program, but the better-run carriers may become supporters if it results in pressure on their competitors who cut corners on maintenance and operations. In Michigan, there was difficulty documenting program benefits. There was insufficient demand for study data and findings by government agencies, safety advocacy groups, researchers, and the trucking industry to justify its continuation (Powers, 2002). When conducting the investigations, Investigators need to distinguish between vehicle defects that preceded the crash and those that resulted from the crash. Also, distinction should be made between defects that contributed to the crash and those that are present but are not related to the crash. Process measures would include the number of trained investigators in the field and the proportion of fatal truck crashes successfully investigated, providing usable data. Since this strategy is directed at a support system (information management), it will not be feasible to relate the effectiveness of the program directly to crashes, but findings from analyses of the data should provide feedback to the state to determine how vehicle inspection programs could be modified to address particular problems detected and to identify motor carriers for compliance reviews. Problem stretches of roadway can also be identified. Surrogate measures may be employed, such as the change in the incidence of unsafe trucks operating on the highway. A protocol for dealing with lawyers and motor carriers needs to be developed. In the case of FACT, because data were collected by government employees, they had to be made available upon request. However, lawyers were usually unable to interpret the data provided. The primary purpose of FACT was to achieve greater understanding of fatal truck crashes and how they might be prevented, not to support or refute the positions of any disputants.

SECTION V—DESCRIPTIONS OF STRATEGIES Information on Agencies or Organizations Currently Implementing this Strategy Michigan conducted this strategy for about 4 years but then discontinued it due to its expense, a need to redirect human resources participating in FACT, lack of demand for the data, and the consideration that FMCSA was launching a similar and broader study. There are no current plans to reinstate the program. A point of contact in Michigan is Capt. Robert R. Powers, Michigan State Police Motor Carrier Division (517-336-6447, powersr@state.mi.us). Additional information is available in Appendix 3. Pennsylvania has enacted legislation that requires all trucks in fatal crashes to undergo an MCSAP inspection. The point of contact in Pennsylvania is Dan Smyser, Chief, Motor Carrier Division, PENNDOT (717-787-7445, smyser@state.pa.us). See Appendix 4 for further details. V-37 EXHIBIT V-10 (Continued) Strategy Attributes for Conducting Postcrash Inspections to Identify Major Problems and Problem Carriers Attribute Description Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes As noted above, the cooperation of local enforcement agencies, as well as motor carriers, is essential for program success. These participants need to be informed well ahead of time about the program and its purpose. The potential value to these stakeholders should be highlighted. The program in Michigan was developed and implemented in about 6 months. Time was required to develop the data forms, and care was taken to ensure that they were compatible with other data systems that were relevant, e.g., the national Trucks in Fatal Accidents (TIFA) file. Following implementation, revisions were made, so that it took about 1 year before the entire system was consistently collecting useful data. Because Michigan ranked in the top 10 states for number of fatal truck crashes, the federal government provided $100,000 to be used to address the problem. Costs were not a major problem, in that no new personnel were required. The program operated by expanding the duties of existing personnel, both local and state. However, the additional responsibilities were not significant, and eventually program participants became program enthusiasts. The major program cost was for data analysis. Motor carrier crash investigators will have to be trained to use the new report forms in most cases. This will be the major training requirement. In addition, local enforcement should be notified of the fatal crash occurrence. In Pennsylvania, personnel from many organizations may be called upon to perform the inspections. This includes the Pennsylvania State Police, the Public Utility Commission, PENNDOT, and selected local police departments. Individuals performing the inspections must be trained and certified. Responsibility for performing the inspection is determined based upon who initially responds to the crash (i.e., state or local police). No legislation is required, although, as noted, the Pennsylvania program is mandated and supported by legislation. None identified.

Objective 12.1 E—Identify and Correct Unsafe Roadway Infrastructure and Operational Characteristics General Description The physical and operational characteristics of large trucks often place them near the safety limits imposed by highway geometric design and the traffic environment (Harwood et al., 2003a and 2003b). Thus, roadway features such as lane width, upgrades, downgrades, horizontal curves, and interchange ramps may be associated with heightened safety concerns relating to large trucks as compared with smaller, lighter vehicles. Although AASHTO (2001) roadway design guidelines, including design speeds, are based on a consideration of the designs of various large vehicles (i.e., large buses and various large-truck configurations), margins for driver error are far less than they are for smaller, more maneuverable vehicles. There are roadway sections that are characterized by large numbers of heavy-truck crashes. Clearly, the volume of heavy-truck traffic is an important factor here, but it is still worthwhile to examine the roadway characteristics to determine what, if anything, might be done to reduce the toll. Most of the strategies related here are designed to either impact the speed of the truck or overcome the loss of control due to excessive speed. Several agencies are beginning to look at the safety impacts of lane restrictions for trucks or exclusive truck lanes, in addition to the historical use of this strategy to improve traffic operations. Appendix 9 discusses this idea. Studies of vehicle highway speeds in North America indicate that the majority of drivers of both light and heavy vehicles consistently exceed posted speed limits (Tardif, 2003; NHTSA, 1991). Both light- and heavy-vehicle drivers exceed speed limits, but average heavy-vehicle speeds are typically slightly lower—on the order of 2 to 5 mph. Moreover, the percentage of truck drivers engaging in extreme speeding (e.g., 80+ mph) is far less than that of light- vehicle drivers. Excessive speed still constitutes a major risk factor for large trucks, however. Possible interventions include improved signing to alert truck drivers as well as other drivers of the hazardous nature of a section of highway; using devices being developed under the Intelligent Transportation Systems (ITS) program to provide real-time feedback to truck drivers exceeding the safe speed; providing pull-offs at the top of the grade to enable drivers to prepare for maneuvering the grade safely; installing arrester beds (truck escape ramps) where appropriate; and installing median rumble strips or barriers in areas where crashes occur because trucks and/or passenger vehicles run off the left side of the road. In the case of barriers, most are not designed to contain heavy vehicles, although some jurisdictions have constructed large, heavy- duty guardrails at some high-risk locations to accommodate heavy trucks (Harwood et al., 2003a). Interchange areas are also potentially hazardous locations for trucks. Ramps with sharp curves and inadequate acceleration or deceleration lanes can be high-crash-risk locations. Some of these proposed interventions will affect truck crashes somewhat indirectly by modifying the behavior of other drivers. For example, head-on collisions, both fatal and nonfatal, are much more likely to be precipitated by the other driver crossing the center line or median, rather than the large-truck driver (Blower, 1999). Therefore, median rumble strips or improved medians should have a greater effect on these drivers, where inattention or drowsy driving is a major factor, but the changes should reduce head-on collisions with trucks. SECTION V—DESCRIPTIONS OF STRATEGIES V-38

SECTION V—DESCRIPTIONS OF STRATEGIES Many of the strategies that are applicable here are covered in other guides in this series: • 15.1 Run-Off-Road Crashes • 15.2 Crashes on Highway Curves • 18.1 Head-On Crashes Some of the strategies that are applicable to these problems would involve a major reconstruction effort. Since the orientation of the AASHTO guides is toward low-cost, short- term solutions, the most extensive and costly countermeasures are not discussed here. Strategy 12.1 E1: Identify and Treat Truck Crash Roadway Segments—Signing It is not always possible to make major changes in highway configurations, even when it is found that the current configuration may create problems for some drivers and vehicles. However, specific segments of highway that are identified as sites with a disproportionate occurrence of truck crashes can be treated to inform drivers of the hazards so that they can modify their driving accordingly. Signs may be the traditional fixed type or be activated and changeable (e.g., advisory speed signs). The signing may also advise of the likelihood of targeted traffic enforcement. In general, it is considered advisable to combine the signing with an enforcement effort. Details regarding traffic enforcement are covered under Strategy 12.1 E3. The special case of rollover advisory warning signs is covered under Strategy 12.1 E2. Colorado’s Weld Country truck safety initiative, a prototype for future statewide programs, includes engineering, enforcement, and education strategies. A principal engineering initiative is improved signage for trucks, including (a) directional signage relating to geographic locations such as major shippers and receivers; (b) signage concerning specific hazardous roadway locations; and (c) signage for truck routes and hazardous materials corridors. With advancing technology, it appears that there will also be vehicle-based systems to provide in-cab advisories to commercial drivers when they are approaching high-crash locations. The U.S. DOT Intelligent Vehicle Initiative (www.its.dot.gov/ivi/ivi.htm) includes a major program of R&D focusing on commercial vehicle operations (CVO). A current study is the “Generation 0” field operational test of several safety technologies installed on Mack trucks. The three systems being tested are a lane tracking system, an in-cab advisory of potentially hazardous locations, and an automatic collision notification system (which automatically calls local EMS following a crash impact). Of interest here is the advisory system, which employs a global positioning system (GPS) both to identify locations and for positioning in the truck. Based on state crash data files and other roadway features (e.g., sharp curves, high-wind areas, tight exit ramps, narrow bridges, and recurring congestion), 500 high-truck-crash locations were identified in 12 states. For the study, these locations are termed “advisory sites” rather than “hazardous locations” because the latter might expose states to undue liability. Moreover, there is no control for truck traffic volume at these sites, so they are not necessarily the sites of greatest risk (this limitation applies to almost all current attempts to identify high-risk highway locations). The in-cab advisory (10 words or less specifying the nature of the potential hazard) is displayed on a small cathode ray tube (CRT) screen mounted on the truck instrument panel. The system is directionally sensitive (i.e., trucks must be approaching from the travel direction associated with high-crash potential) and provides the advisory about 1 mile before the critical location. This is a good example of V-39

SECTION V—DESCRIPTIONS OF STRATEGIES V-40 how vehicle-based technologies may be used to complement infrastructure signage and other countermeasures. Of course, fundamental limitations of such vehicle-based systems are that they take years to be deployed in the vehicle fleet, and their implementation depends mainly on truck buyers ordering them at the time of vehicle purchase. EXHIBIT V-11 Strategy Attributes for Identifying and Treating Truck Crash Roadway Segments—Signing Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Principally, the target is drivers of large trucks who may be traveling in a manner that is unsafe for the segment of roadway. Secondarily, the target is other drivers, who may be reached with the message to also use extra caution. The strategy is considered experimental and, therefore, should be initially implemented on a pilot basis to allow evaluations to be conducted to ensure that the strategy is cost-effective. The separate effect of signing on crashes involving large trucks traveling in an unsafe manner has not been determined. Experiments with this are being evaluated by the Pennsylvania DOT. However, since these are often implemented in conjunction with a selective enforcement program, the effect of the signing will be difficult, if not impossible, to separate. A key to success is having a process and system to identify the locations considered especially hazardous for trucks. Success also requires the cooperation and support of state and local enforcement, which must provide intensified enforcement in these areas. There should be public information and education (PI&E) efforts to increase general awareness of the truck safety problem and increase the likelihood that the signed information will be heeded. Enforcement personnel need to be available for dedicated efforts at signage locations to reinforce their effectiveness. This strategy should not be viewed as a permanent substitute for highway engineering improvements, but rather should be considered an interim effort until such time as the highway itself can be modified. Process measures include the number and type of signs that are placed, including the number of corridors and ramps treated (i.e., information versus interactive signs installed). Impact measures would include number of truck crashes, particularly the target crashes of the specific intervention(s), e.g., reduction in rollovers in response to installation of interactive truck rollover signing. The ultimate measure of effectiveness is whether heavy-truck crashes decline. Surrogate safety measures include unsafe driving maneuvers, truck and other vehicle speeds, speed differentials between trucks and other vehicles, and the difference between the operating speed of trucks and the design speed of a geometric element. Signing needs to be maintained, and the interactive signs need to be monitored to ensure that they are performing appropriately. Although signing can be implemented without the involvement of other agencies, the support of state and local enforcement is essential. In addition, it is always wise to keep key stakeholders informed of proposed changes. Key legislators, the state trucking association, truck drivers, organizations representing the driving public (e.g., AAA), and the insurance industry are candidate stakeholders to consider informing.

SECTION V—DESCRIPTIONS OF STRATEGIES Information on Agencies or Organizations Currently Implementing this Strategy The Pennsylvania Department of Transportation (PENNDOT) Bureau of Highway Safety and Traffic Engineering is currently pursuing this strategy to identify highway stretches characterized by large numbers of truck crashes and to implement this strategy to address the problem. See Appendix 5 for further details. Information on the Intelligent Vehicle Initiative (IVI) CVO “Generation 0”test of in-cab advisories of high-truck-crash locations (as well as information on other IVI commercial vehicle R&D) can be obtained from the FMCSA IVI CVO platform manager, Tim Johnson (202-385-2362, tim.johnson@fmcsa.dot.gov), or from the IVI Web site (www.its.dot.gov/ivi/ivi.htm). Strategy 12.1 E2: Install Interactive Truck Rollover Signing General Description Large trucks have high centers of gravity, especially when their trailers are loaded. This physical characteristic renders them much more vulnerable than smaller vehicles to rollover on curves. Moreover, it appears that vehicle height has an effect on drivers’ perception of speed; greater heights are associated with lower perceived vehicle speeds (Rudin-Brown, 2004). Further, drivers of tractor-semi-trailers cannot sense the level of lateral acceleration experienced by their trailers because of the articulation between the tractor and the trailer. Interstate and other freeway exit ramp curves can be dangerous locations for tractor-trailers because the driver must perceive the point at which to begin braking and the amount of braking needed to safely slow from full freeway speeds to a much lower speed to negotiate the ramp curve. In a survey of state DOTs, Harwood et al. (2003a) reported that 74 percent of responding states indicated that they had safety problems at such locations and that 57 percent of them employ special warning signs for trucks at their highest-risk locations. A smaller percentage (31 percent) employs advisory speed limits for trucks at these locations. V-41 EXHIBIT V-11 (Continued) Strategy Attributes for Identifying and Treating Truck Crash Roadway Segments—Signing Attribute Description Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes State crash data must be analyzed and field visits made to candidate sites to identify the specific locations of truck crashes, as well as the apparent causes, so that this strategy is applied where it is likely to have some impact. Costs include the original crash data analyses, the signing on selected corridors, and the enforcement required to maintain effectiveness. For details on the latter, see Strategy 12.1 E3. Personnel will be needed to conduct the detailed crash analyses, and state or local police will be needed to conduct enforcement efforts. Therefore, training on the special dynamics and needs for truck safety should be included in training for this type of personnel. None. None identified.

Highway ramps or curves that experience a high incidence of truck rollovers can be identified from state highway crash files. Interactive signs that include highway detectors can be installed at identified locations (Bushman and Lindsay, 2002). A typical “intelligent” interactive system includes sensors for both weight (i.e., weigh-in-motion) and speed and a display to flash a warning sign for trucks that are assessed to be at rollover risk. These systems may vary in complexity and cost; the simplest systems measure only vehicle speed or height (thus identifying a large truck), but more sophisticated systems can measure multiple vehicle parameters, including speed, height, and weight to calculate rollover risk more accurately and thus provide more targeted warnings (Harwood et al., 2003a). Warning displays may be programmed with a single warning or changeable messages for different situations (Bushman and Lindsay, 2002). If properly designed, installed, and maintained, these interactive systems can result in significantly decreased truck speeds on ramps and resulting decreased crash risks (Harwood et al., 2003a). Exhibit V-12 provides a schematic of an interactive truck rollover advisory as well as a picture of a dynamic display. A similar principle can be applied to downhill speed advisories for trucks (Bushman and Lindsay, 2002). Just as truck drivers may have difficulty sensing unsafe speeds on curves, they may not appreciate the risks associated with high downhill speeds under some conditions. The layout of such truck downhill speed advisory systems is analogous to the warning system on curves. Vehicle weight is an important factor in determining loss-of- control risk, so optimal systems include weigh-in-motion capabilities. Maximum safe speeds are calculated based on a predetermined formula based on truck weights, speeds, roadway gradients, and truck braking capabilities. Each truck may be given a specific message conveying an advisory speed for that vehicle. An alternative solution to infrastructure-based warning systems is a vehicle-based system to prevent rollovers. It will likely be 10 to 20 years before the system penetrates the majority of the heavy-truck fleet. Further information on this may be found in Appendix 7. Another ITS concept is to equip heavy vehicles with embedded roadway maps indicating highway curve locations and a positioning system to determine vehicle location in relation to highway curves. Such a system would provide a warning to drivers if they entered a curve at excessive speed. Information on Agencies or Organizations Currently Implementing this Strategy The Pennsylvania Department of Transportation is currently pursuing the vehicle-highway interactive strategy at some Interstate exit ramps and other high-rollover-risk locations. See Appendix 6 for further details. SECTION V—DESCRIPTIONS OF STRATEGIES V-42

SECTION V—DESCRIPTIONS OF STRATEGIES V-43 EXHIBIT V-12 Schematic of an Interactive Truck Rollover Advisory System and Photograph of Dynamic Display Provided by the PENNDOT Bureau of Highway Safety and Traffic Engineering.

SECTION V—DESCRIPTIONS OF STRATEGIES V-44 EXHIBIT V-13 Strategy Attributes for Installing Interactive Truck Rollover Signing Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes This strategy targets trucks operating at unsafe speeds on highway ramps or curves identified as having a high incidence of truck rollovers. It is anticipated that truck drivers will respond appropriately to an interactive sign warning them that they are exceeding a safe speed for the highway location. Harwood et al. (2003a) reviewed implementation cases in a number of states, including California, Texas, Missouri, Virginia, Maryland, and Pennsylvania. Most cases resulted in significantly decreased truck speeds. Crash reductions were observed also, though they were less likely to be statistically significant because of their small numbers. A key to success is having a system to identify the locations considered especially hazardous for trucks. Moreover, targeted speed enforcement at these locations will reinforce the effectiveness of both conventional and interactive signing. The interactive signing must be calibrated for each location. If signs are slowing trucks well below safe speeds, the signs will lose their effectiveness as drivers discover the discrepancy. Baker et al. (2001) found that incorporating dynamic measures of vehicle weight into the sensor suite increases system accuracy and reduces false alarms. Costs may be significant; see below. Process measures include the number and type of signs placed and the number of locations treated. Impact measures include measures of actual rollover crashes during an extended period of time prior to sign installation and following installation. Surrogates included speeds entering the ramps or curves and whether and how much speeds are reduced in response to the signing. None identified. None identified. The process of identifying and evaluating candidate sites, as well as the design and installation of the system, will require significant time. Warning threshold speeds must be calibrated for each site and monitored to ensure that they are not too high (resulting in false negatives, i.e., speeding trucks not receiving speed warnings) or too low (resulting in false positives, i.e., too many trucks receiving speed warnings). Implementation could take a year. Costs vary with the capabilities and sophistication of the systems employed. In Pennsylvania the cost for installing sensors to detect truck speed and provide feedback to truck drivers has been approximately $210,000 per location. Highway engineers may require training or consultation on various aspects of high-risk location identification, system design, installation, maintenance, and evaluation. None required, although it is always wise to keep key legislators informed for such programs. None identified.

SECTION V—DESCRIPTIONS OF STRATEGIES Strategy 12.1 E3: Modify Speed Limits and Increase Enforcement to Reduce Truck and Other Vehicle Speeds General Description An obvious rationale for reducing posted speed limits is to reduce average speeds, with resulting reductions in crash incidence and/or average crash severity (Stuster et al., 1998), although relationships between posted speeds and prevailing operating speeds are not always strong (Fitzpatrick et al., 2003). Existing speed limits may be set too high for heavy trucks given their operational limitations such as longer stopping distances and greater vulnerability to rollover on curves (Harwood et al., 2003a). In addition to increased enforcement (as discussed above), it may be necessary to reduce existing truck speed limits. If speed limits are reduced in response to a high incidence of truck crashes, should they be reduced for trucks only or for all vehicles? A rationale for differential speed limits is that trucks have much longer stopping distances than do light vehicles and have other speed- related risks such as rollover at lower speeds and vulnerability to loss of control in cross winds. Thus, trucks should maintain lower highway speeds. On the other hand, the advocates of uniform speed limits contend that differential truck-car speed limits increase vehicle speed variance on the roadway, resulting in more truck-car conflicts and potential for more rear-end and passing-related crashes. In a review of highway/heavy-vehicle interaction, Harwood et al. (2003a) reported that about one-third of the states employ differential speed limits for large trucks at some locations, either for particular classes of roadway (e.g., rural freeways) or for selected roadways. However, Harwood and others (e.g., Garber and Gadiraju, 1992) have found that reducing speed limits for trucks only (i.e., having differential speed limits) does not reliably reduce crashes, although it may change the distribution of various crash types. Garber et al. (2003) compared the safety effects of uniform car-truck speed limits with differential speed limits on rural Interstate highways. A surprising finding was that, overall, vehicle speeds have not been significantly affected by the type of speed policy. Moreover, statewide changes in rural Interstate speed policy from uniform to differential, or vice-versa, have not resulted in consistent or significant changes in crash rates, including crash rates for rear-end crashes specifically. Situations where differential speed limits may be advisable include curves and steep downgrades where it may be critical that trucks drive at a slower speed to avoid rollovers, brake failure, or disastrous potential runaway problems. Pennsylvania has a program to reduce the truck speed limit for trucks with a gross weight generally over 26,000 pounds, based on an engineering and traffic study. Based upon recommendations from the trucking industry, Pennsylvania uses a “hazardous grade speed limit” that is consistent with the speed at which these trucks climb the hill in the opposite direction. Care must be taken to follow standards when establishing speed limits. The Manual on Uniform Traffic Control Devices (MUTCD) specifies that speed limits should be established “After an engineering study has been made in accordance with established traffic engineering practices” (Section 2B.13 Speed Limit Sign). It further specifies the following: At least once every 5 years, States and local agencies should reevaluate nonstatutory speed limits on segments of their roadways that have undergone a significant change V-45

in roadway characteristics or surrounding land use since the last review. . . . When a speed limit is to be posted, it should be within 10 km/h or 5 mph of the 85th- percentile speed of free-flowing traffic. Two types of Speed Limit signs may be used: one to designate passenger car speeds, including any nighttime information or minimum speed limit that might apply; and the other to show any special speed limits for trucks and other vehicles (MUTCD, 2003, http://mutcd.fhwa.dot.gov/kno-millennium.htm). The Institute of Transportation Engineers (ITE) has a draft advisory on speed zoning available at http://www.ibiblio.org/rdu/ite-szg.html. SECTION V—DESCRIPTIONS OF STRATEGIES V-46 EXHIBIT V-14 Strategy Attributes for Modifying Speed Limits and Increasing Enforcement to Reduce Truck and Other Vehicle Speeds Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues The target is truck drivers exceeding established speed limits. There have been variable results regarding the effect of speed limit changes on speeds. Also, there is no valid evidence that reducing speed limits for trucks will reduce large-truck crashes. The type of facility, as well as the initial and final speed limits will play important roles in the final result. To enhance the impact of this strategy, the state must work with the traffic law enforcement agencies and with the state motor carrier association to get the information out to truck drivers. It would also be helpful to actively involve local political representatives and the local media. It is especially important to involve trucking trade associations, companies, and drivers in the effort, so that the focus is on safety, rather than on identifying and punishing offenders. Care should be taken to avoid pressure to establish speed limits on a basis other than that specified by the MUTCD and to follow sound procedures of the type established by ITE. Process measures would include number of locations, or miles of road, for which speed limits have been changed and hours (and officer-hours) of targeted speed enforcement. Impact measures would include changes in speed-related crashes in the affected areas. Changes in truck and passenger vehicle speed in the affected areas may be used as surrogate measures, but caution should be taken, since the connection to impacts on crashes is tenuous at best. Almost any enforcement program will be enhanced by a coordinated public information and education (PI&E) program. Change of speed limit is usually difficult to achieve and sometimes politically charged. It is advisable to involve all stakeholders in decision processes from an early point. Enforcement and engineering agencies should work in close cooperation.

SECTION V—DESCRIPTIONS OF STRATEGIES Information on Agencies or Organizations Currently Implementing this Strategy In 1996, Colorado raised its Interstate highway speed limit to 75 mph. Since that time, traffic volumes have dramatically increased in some areas of the state. On I-25 north of Denver to Fort Collins, the combination of increased speeds and increasing traffic volume raised the question of whether the speed limit should be reduced back to 65 mph along this section of Interstate. A number of truck drivers and carrier safety managers have expressed support for such a speed limit reduction, and the strategy was included in a recent Colorado truck safety initiative. However, opposition by commuters and other travelers in the area was anticipated, and the matter is unresolved at this writing. Objective 12.1 F—Improve and Enhance Truck Safety Data Strategy 12.1 F1: Increase the Timeliness, Accuracy, and Completeness of Truck Safety Data General Description Trucks cross state lines much more often than other vehicle traffic does. Averaging almost 65,000 miles annually, combination trucks travel through many jurisdictions and consequently can incur violations in multiple districts. A primary purpose of the CDL is to limit a driver to holding a single license and to establish a reporting system that compiles a single record incorporating data from all jurisdictions where infractions and/or crashes occur. Because of trucks’ speed and distance covered, for data to be useful, they must be V-47 EXHIBIT V-14 (Continued) Strategy Attributes for Modifying Speed Limits and Increasing Enforcement to Reduce Truck and Other Vehicle Speeds Attribute Description Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes Achieving approval of a change of speed limit, including obtaining the studies that may be necessary, could result in an extended implementation period. Otherwise, once problem areas are identified, implementation should take no more than a few months. Strategies targeted at reductions in posted speeds should not be costly, but increased enforcement in affected areas will be. Because truck crashes, both fatal and nonfatal, occur most frequently during daytime and on weekdays, enforcement should be most effective during these times. No special training should be required, as this strategy is a standard part of engineering and enforcement agency operations. It may be necessary to pass legislation in some cases where speed limits for trucks are to be changed on a systemwide basis. None identified.

SECTION V—DESCRIPTIONS OF STRATEGIES V-48 State Fiber System JudicialDPS DOTLaw Enforcement Vehicle Federal Databases TraCS Law Enforcement Database Accidents Driver RecordsCommercial Vehicle Inspections DOT Databases Courts DPS Criminal Reporting Database DPS = department of public safety. available rapidly and in complete and accurate form. Technology is available today to enable such data entry and accessibility, but few jurisdictions are currently utilizing it. The state of Iowa, in partnership with the U.S. DOT, has developed “TraCS”: Traffic and Criminal Software (see http://www.iowadot.org/natmodel/index.htm) under a national model program for demonstrating the use of new technologies for improving data collection and analysis. The system allows law enforcement officers and others to collect, validate, print, and receive information in the vehicle using a notebook or pen-based computer. This information can be transferred to central databases for reporting, analyses, and retrieval. TraCS has reduced time requirements for data collection and entry, increased accuracy, and made safety data rapidly available for analysis and action. It is being licensed to numerous other states for various applications. For a map of the partner states, see http://www.dot.state.ia.us/natmodel/otherstates.htm, and for further details, see http://www.dot.state.ia.us/natmodel/letpublication.pdf. TraCS was not developed solely for truck safety applications. Nevertheless, one major component of TraCS is the Vehicle Safety Inspection System (VSIS), which is an alternative to a U.S. DOT–developed system called ASPEN. Both systems facilitate the collection and reporting of data from truck inspections. An advantage of TraCS is that it is easily linked to other police data systems such as systems for accident and citation reporting. In Iowa, truck enforcement officers writing truck inspection reports can, with one mouse click, also file their citation reports. This capability greatly expedites truck safety enforcement. Exhibit V-13a illustrates conceptually the electronic data flow of TraCS. EXHIBIT V-13A Electronic Data Flow

SECTION V—DESCRIPTIONS OF STRATEGIES V-49 EXHIBIT V-14 Strategy Attributes for Increasing the Timeliness, Accuracy, and Completeness of Truck Safety Data Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties Appropriate Measures and Data This strategy targets law enforcement and members of the traffic safety system who are making planning, operational, and maintenance decisions that affect large trucks. This strategy addresses a support system. It is not possible to directly relate the effectiveness of this strategy to large-truck crashes, due to the many intervening steps between provision of data and impact on truck operations. However, the improvement of data systems reduces paperwork at every stage of information processing and increases accuracy and timeliness of data. Iowa, using the TraCS system, has reduced the time in which incident information is electronically available to the state to as little as 1 day. They have also eliminated duplicate data entry by agency staff, significantly reduced errors on reports, and reduced time for completing reports. The time to make commercial vehicle inspection data available for analysis has decreased from 100 days to 6 days. The time between the occurrence of a crash and the availability of the crash data has decreased from 12-18 months to as little as 8 hours. Mobile data units in patrol cars, barcode technology, and “smart maps” on which officers can indicate incident location combine to reduce data entry time while accuracy increases. TraCS agencies without mobile laptops may still use TraCS on desktop computers at their agency. In this case, an officer must make notes in the field and bring them to the office to be entered by either the officer or a clerk. The advantages of creating a local database and being able to electronically transmit to the state provide enough benefits for many agencies to use this approach. Agencies using desktop set- ups are limited to reports that don’t require a driver signature, unless drivers are brought into the office. This is sometimes the procedure in small towns. Inter-agency coordination and cooperation are required for successful implementation. In Iowa, the Department of Transportation (Motor Vehicles Division and Information Technology Division), the Department of Public Safety (State Patrol, Governor’s Traffic Safety Bureau, Data Services Bureau), the Department of Natural Resources, multiple U.S. DOT agencies, and multiple county agencies are involved. More jurisdictions are joining the initiative as time and funding allow. The value of the system is enhanced as more jurisdictions participate, in that the breadth of information increases. Because agencies differ in their equipment, the software needs to be usable on a wide variety of platforms. The TraCS system may require the purchase of new equipment, such as upgraded computers, barcode scanners, and printers. And new equipment often means new staff training requirements. While traditional systems are cumbersome, they are also familiar. New systems must overcome the inevitable institutional inertia that is inherent in bureaucracies. Process measures would include the number and proportion of agencies using the improved system. Operational impact measures include the time required to complete reports, changes in reporting errors, and time from incident to availability of information for retrieval. Personnel productivity measures will also indicate the impact of this strategy on operations. Impact on truck safety would be difficult to measure, but improved speed in identifying drivers who should lose licensure may serve as a proxy measure. (continued on next page)

Information on Agencies or Organizations Currently Implementing this Strategy Specific agencies, in addition to the Iowa DOT, that are involved in testing and implementing the national model effort may be found at http://www.iacptechnology.org/Programs/ NationalModel.htm. Information may also be obtained from Mary Jensen, TraCS Program Manager, Motor Vehicle Division, Iowa DOT (515-237-3235, mary.jensen@dot.state.ia.us). SECTION V—DESCRIPTIONS OF STRATEGIES V-50 EXHIBIT V-14 (Continued) Strategy Attributes for Increasing the Timeliness, Accuracy, and Completeness of Truck Safety Data Attribute Description Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes None identified. The extent to which data system improvements are effective depends on the breadth of their use. If only the state police have such a system, then incidents coming under the jurisdiction of counties, towns, and cities will not be included and will have all the disadvantages of traditional systems. Inter-agency cooperation is essential for achieving optimal benefits from improved data systems. The program in Iowa began in 1994 with only eight enforcement agencies participating. Over the years it has expanded to include more than 200 agencies. In addition to enforcement agencies, current participants include the state DOT, division of motor vehicles, state and local engineering units, ITS, and the judicial system. Types of data handled include crash reports, citation data, drunk driving arrest and report forms, commercial motor vehicle citations, and arrest and complaint forms for criminal activities. This strategy represents a major change in data collection and processing. It will require time for gaining approval, careful planning, software development, training, acquisition of new equipment, trial implementations, and final installation. It is anticipated that at least 18 months to 2 years will be required to carry through this process. As noted, the TraCS system may require up-front investment in hardware and software, as well as in training. Although the system may increase productivity and improve accuracy and timeliness, funds for the initial capital investment may be required for the system to be installed. New TraCS states need to create their forms using the Software Development Kit (SDK) provided with TraCS. TraCS and the SDK are provided free of charge, but there are development and deployment costs. Although TraCS works on a variety of computers and printers, devices that meet all requirements (performance, features, and durability) are usually expensive. The cost of the units varies as a function of the quality of the hardware used. Some agencies already have suitable laptops in their vehicles. Other agencies may use TraCS on a desktop at their office; newer computers can handle TraCS nicely. However, older computers may need to be upgraded or replaced. Personnel need to be trained on how to use the new equipment. Report forms, driver licenses, vehicle registrations, and other relevant documents need to be barcoded. Ordinarily, no new legislation is required. However, some states may need legislation to allow for the use of digital signatures. None identified.

SECTION V—DESCRIPTIONS OF STRATEGIES Objective 12.1 G—Promote Industry Safety Initiatives Strategy 12.1 G1: Perform Safety Consultations with Carrier Safety Management General Description Enforcement strategies are intended to ensure that all motor carriers and drivers comply with certain fundamental safety requirements, ranging from driver medical condition and other qualifications, driver hours-of-service compliance, vehicle condition, highway speed compliance, and compliance with other carrier operational and traffic regulations. Federal, state, and local enforcement activities lay the foundation for ensuring a safe industry and for identifying and punishing noncompliant carriers and drivers. However, punishment is not the only way to stimulate safety-related changes in the motor carrier industry. Educational approaches complement enforcement and indeed can address safety practices not related to compliance. As such, they are potentially much more comprehensive and are naturally more likely to be received positively by carriers. Development of a cooperative relationship between government and industry on the subject of motor carrier safety is strongly valued by most state motor carrier safety officials (Patten, 2001). Regulatory compliance by motor carriers and their drivers is perhaps best viewed as an essential prerequisite for safe operations. However, compliance per se is probably not sufficient to ensure safe commercial vehicle operations. Active carrier safety management, addressing areas and practices beyond compliance, is necessary to achieve and sustain high operational fleet safety (Corsi and Barnard, 2003; Knipling et al., 2003; Stock, 2001, American Trucking Associations Foundation, 1999). There are a number of initiatives that states and other levels of government can take to provide safety education and consultation to the motor carrier industry. Often these initiatives are most effective if they are conducted in partnership with industry, e.g., state motor carrier trade associations. Potential activities include distribution of safety-related publications (brochures, manuals, bulletins, etc.), seminars and workshops for fleet safety managers, seminars or other special training for drivers on topics such as defensive driving, “circuit rider” visits to motor carriers to provide free or low-cost safety consultation, volunteer mentoring for new or problem carriers by established safe carriers, nonpunitive compliance reviews, and advisory warning letters sent to problem carriers before any punitive actions are taken. Patten (2001) reviews a number of different types of state-sponsored education activities for motor carriers, the number of states using each activity (according to survey data), and state officials’ ratings of their safety effectiveness. For example, 33 states conduct seminars, classes, or conferences, and 97 percent of the responding officials from these states rate the programs as effective or very effective. The number of motor carriers contacted through these programs averaged 68 per state annually, but varied widely across states from just a few to 5,000. Also addressed in the Patten report are state motor carrier association education and information dissemination activities. The Tennessee Department of Safety has an Alternative Commercial Enforcement Strategies (ACES) program that provides compliance-related information to fleets in a nonthreatening V-51

manner. In ACES, specially trained officers visit fleets using an advisory rather than an enforcement approach. The officers provide as much information as possible to help fleets to be more proactive in avoiding safety and compliance problems. Training services provided range from demonstrating vehicle inspection procedures to reviewing compliance paperwork requirements to training new drivers. Later visits may be enforcement oriented, but the initial visit is advisory and permits fleet operators to improve their practices. During 1999, the Tennessee ACES group made more than 700 industry contacts and also visited more than 2,000 schools to provide truck safety information and education to the public. In addition, Share the Road information was provided at a number of large public events such as parades, sporting events, and festivals. The award-winning Tennessee ACES unit is composed of 12 sergeants/officers statewide. ACES is based on the concept of community-oriented policing where the “community” is the commercial vehicle industry of Tennessee. Companies visited are offered a variety of educational services free of charge, including • Hours of service classes, • Medical qualifications for commercial vehicles, • Safe driving tips, • Share the Road/No Zone program, • Assistance in establishing drug/alcohol testing programs, • Assistance in establishing vehicle maintenance programs, • Assistance in establishing record-keeping requirements, • Assistance in establishing file maintenance for drivers and vehicles, • Informational Level 1 inspections, • Programs for private carriers of passengers, • Conducting new driver programs/orientation, • Classes for hazardous materials regulations, • Training on conducting pre- and posttrip inspections, and • Educational contacts and compliance reviews. An innovative approach to improving compliance, originating in the state of New York, is the “compliance letter.” Instead of issuing a citation to carriers or conducting a full compliance review, the state may simply require that problem carriers write a letter to the state, stating that they are aware of the regulation(s) in question and current deficiencies in their operations and describing their plans to get into full compliance. Otherwise, these fleets receive no punishment at this stage. The state has found that this nonpunitive exercise often gets the attention of fleet management and motivates them to upgrade their safety and compliance practices proactively, prior to experiencing any major fines or other sanctions. Colorado’s Circuit Rider program is an industry-based initiative to provide free consultation to fleets on their safety compliance and management practices. The program, supported by a NHTSA 402 grant and managed by the Colorado Motor Carrier Association, employs veteran carrier safety managers who travel around the state visiting motor carriers that have requested the consultation. A major attraction of the Circuit Rider program to participating SECTION V—DESCRIPTIONS OF STRATEGIES V-52

SECTION V—DESCRIPTIONS OF STRATEGIES fleets is that it is not related to enforcement and cannot result in punitive consequences to the carrier. Consultation with the fleet owner or safety manager might include the following: • Review of the carrier operation, including staffing levels, equipment, driver files, and insurance; • Review of the fleet’s approach to compliance with key Federal Motor Carrier Safety Regulations (FMCSRs), such as driver drug and alcohol testing and driver hours of service; • Advice on building a stronger safety program for the fleet; and • Providing information and tools to support the carrier’s safety management efforts, such as referrals to sources of information or consultation on specific safety practices. In addition to the direct consultation provided to individual fleets, the Colorado Circuit Rider program conducts safety workshops for motor carrier managers, drivers, and dispatchers. Topics include FMCSR compliance, drug and alcohol testing requirements and procedures, driver selection and hiring, driver performance evaluation, carrier safety management, and vehicle maintenance. See Appendix 8 for further details on this program. The Michigan Truck Safety Commission has established a nonprofit Michigan Center for Truck Safety (MCTS; www.truckingsafety.org) to provide free and low-cost training and consultation to truck drivers and carrier safety managers. The MCTS also manages public Share the Road education programs. Funding for the commission and the center comes from registration fees on heavy vehicles. Professional training includes driver coaching, “decision” driving courses (conducted on skid pads to teach drivers dynamic safety maneuvers such as pulling out of a jackknife), defensive driving, fatigue management, inspection training, load securement training, and safety manager training. There is also an annual Truck Exposition and Safety Symposium. The FMCSA plans a multimedia “Safety is Good Business” program to provide educational materials directly to fleets (FMCSA, 2001a). The program will cover a full range of safety- effective practices that fleet owners and managers can implement immediately or in short timeframes to reduce crashes. A central theme will be the high costs of crash involvement and the benefits of crash prevention. This program, now under development, will target new and small motor carriers. The educational and consultation outreach programs to the motor carrier industry can be viewed as somewhat analogous to FHWA-sponsored Local Technical Assistance Program (LTAP) and Circuit Training Assistance Program (CTAP) training. FHWA supports a network of 57 centers nationwide to provide workshops and consultation in a variety of topics relating to highway design and operation. In addition to course offerings, LTAP programs sponsor conferences and expositions, often attracting hundreds of state and local highway officials and contractors. Several of the motor carrier safety programs above have similarities to the LTAP/CTAP concept and have the potential to play a similar role in professional education and information dissemination. V-53

SECTION V—DESCRIPTIONS OF STRATEGIES V-54 EXHIBIT V-15 Strategy Attributes for Safety Consultation with Carrier Safety Management Attribute Description Technical Attributes Target Expected Effectiveness Keys to Success Potential Difficulties Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes The primary target is motor carrier fleet owners and safety directors. Various fleet safety management practices are among the most salient discriminators between safe and unsafe fleets (Corsi and Barnard, 2003). Improved carrier safety management can significantly improve fleet safety performance, particularly in new or small fleets without systematic safety programs. Most state motor carrier officials with industry education/consultation programs rate them as effective or very effective (Patten, 2001). One key for this strategy is to provide information and consultation in a nonthreatening way, that is, separately from enforcement activities. In terms of methods, there are a variety of innovative approaches for disseminating safety management information; multiple media, including print, should be used to provide information since many small carriers are relatively low-tech. Materials that are entirely Web based may not reach all desired carriers. Since numerous different activities are covered by this strategy, programs must select the most effective ones for disseminating information and changing fleet practices. Also, many carriers may not avail themselves of these consulting services because of fear of exposing unsafe or noncompliant operations to the authorities, in spite of assurances that they will not be vulnerable to enforcement action. Process measures would include the number of fleets receiving education or consultation and their aggregate size (e.g., number of drivers or trucks). The impact on truck safety would be measured by changes to fleet out-of-service, traffic violation, or crash statistics. Exposure variables would also be needed. A cooperative and collaborative relationship between government and industry will promote maximum effectiveness (Patten, 2001). Because these activities are not enforcement related, they may be viewed within some organizations as “icing on the cake” rather than as essential programs. Limitations of funding and/or personnel resources may prevent full implementation. Also, as noted, this strategy is greatly enhanced by government-industry collaboration. Funding and development times for educational materials and courses. Depends on the nature of the activities. Cost elements include personnel to manage and operate the program; costs of materials, equipment, and facilities; and costs experienced by fleets as they institute new programs and procedures. Individuals providing consultation to fleets must be highly qualified and credible to be regarded as authoritative by fleet safety managers. Ordinarily, no new legislation is required. However, some states may need legislation to fund or otherwise permit alternative enforcement strategies. None identified.

SECTION V—DESCRIPTIONS OF STRATEGIES Information on Agencies or Organizations Currently Implementing this Strategy Below are information sources for various safety consultation programs described above: • Tennessee ACES program: Captain Steve Binkley, Tennessee Department of Safety (615-687-2317, steve.binkley@state.tn.us). • New York Compliance Letters: Mark White, Manager, Permit Section, Traffic Engineering and Highway Safety Division, New York DOT (518-457-1795, mwhite@dot.state.ny.us). • Colorado Circuit Rider program: Patricia J. Olsgard, Director, Safety, Training, and Research, Colorado Motor Carriers Association (303-433-3375, x304, patti@cmca.com, see Appendix 8). • Michigan Center for Truck Safety: www.truckingsafety.org (800-682-4682). • FMCSA Safety Is Good Business Program: Tony Schafer, Safety Action Programs Division (202-366-2953, Anthony.schafer@fmcsa.dot.gov). Strategy 12.1 G2: Promote Development and Deployment of Truck Safety Technologies General Description Most of the truck safety strategies presented in this guide involve activities that are performed primarily by state motor carrier safety agencies, i.e., state DOTs or DMVs. However, many aspects of truck safety are influenced primarily by industry—the motor carrier transport industry and/or the truck manufacturing industry. One such safety element is vehicle safety design. Improved heavy-truck safety designs and technologies may help drivers (i.e., truck drivers or other drivers driving around trucks) avoid crashes or may be oriented toward improving occupant survivability during a crash. Manufacturers play a principal role in determining vehicle safety design by the standard and optional safety equipment installed on their vehicles. For example, electronic braking systems are relatively new technologies that are beginning to penetrate the new truck market. Anti-lock brakes are a more mature technology that was mandated by NHTSA (in Federal Motor Vehicle Safety Standard [FMVSS] 121) for all new trucks and trailers in 1996. The motor carrier industry also plays a principal role by selecting specific equipment for new vehicles—referred to as “spec-ing” the new vehicle. All new trucks must meet the FMVSSs (developed and enforced by NHTSA), but, beyond compliance with these standards, buyers of new trucks have considerable discretion in the safety-related features and components they select for their vehicles. Buyers may specify different engine performance specifications (e.g., related to maximum speeds and optimal fuel economy), different types of brakes, tires, mirrors, lighting and signaling configurations, and other components relevant to safe operations. In addition to basic safety-related components such as brakes and tires, various advanced technology collision avoidance systems have been developed and marketed. For example, the Eaton-VORAD (Vehicle On-board RADar) forward collision warning system is associated with a 35-percent reduction in truck-striking-rear-end crashes, according to the V-55

system’s Web site (truck.eaton.com/vorad). Other advanced technologies under development or marketed include adaptive cruise control, roll stability advisors and controllers, and lane departure warning systems (FMCSA, 2003b). Advanced on-board sensor systems can provide diagnostic monitoring of safety-critical components such as brakes and tires. These advanced technology devices may be selected for installation on new vehicles at the time of purchase or may be purchased for retrofit for vehicles already in the fleet. In addition to functioning as collision warning systems, advanced technologies can be used to monitor and modify commercial driver safety behavior. Such new or emerging technologies include adaptive cruise control, rollover detection and prevention systems, lane trackers and lane departure warnings, side sensing (proximity) devices, vehicle and cargo tracking systems, event data recorders (“black boxes”), and driver alertness monitoring (Roetting et al., 2003). A review of truck safety technologies commissioned by PENNDOT (Parsons Brinckerhoff Quade and Douglas, Inc., et al., 2002) describes technologies and safety practices relating to underride (or underrun) prevention, improved braking systems, prevention of postcrash fires, tire failure, driver fatigue, collision warning, electronic vehicle speed regulation, improved cab structural integrity during rollovers, advanced side mirror designs, and enhanced truck/trailer conspicuity. The study reviews new developments occurring in both North America and Europe. Some European truck safety design and operational standards are more stringent than those in the United States, such as standards for truck cab structural integrity during rollovers and a requirement for use of on-board recorders (tachographs) for driver hours-of-service verification (Hartman et al., 2000). In a survey, Knipling et al. (2003) found that most safety-conscious fleet safety managers carefully “spec” their new vehicles for basic safety equipment such as brakes, tires, mirrors, and conspicuity lighting, but that a minority currently order advanced technology safety devices such as forward radar obstacle detection systems. As noted in Section III, combination-unit trucks are, of all vehicle types, the vehicle type most likely to be associated with highly positive cost-benefits from the installation of enhanced safety equipment. Although combination-unit trucks (tractor-trailers) have relatively low crash rates per mile traveled, their high mileage exposures and the severity of their crashes combine to associate them with much greater costs for an average crash, average crash costs per year, and average vehicle life-cycle crash costs. Average crash costs over the operational life of a combination-unit truck are more than four times higher than most other vehicle types (Wang et al., 1999); see Exhibit V-16. For combination-unit trucks, there are greater benefits per vehicle and per investment dollar than for other vehicle types. Thus, they are generally the vehicle platform of choice for early cost-effective deployment of motor vehicle safety technologies (Wang et al., 1999). Single-unit large trucks (also called straight trucks) do not generally have such high life-cycle crash costs because their annual and lifetime mileage exposures are more similar to passenger vehicles than to long-haul tractor-trailers. Thus, for most vehicle-based safety technologies, combination-unit trucks are a much more attractive platform than are single-unit trucks. Truck vehicle safety technologies are not panaceas, however. Effective fleet deployments often require active vehicle maintenance and driver safety management. In a survey of SECTION V—DESCRIPTIONS OF STRATEGIES V-56

SECTION V—DESCRIPTIONS OF STRATEGIES motor carrier fleet safety managers, Knipling et al. (2003) found that specifying enhanced safety equipment on new vehicles was rated only average in effectiveness compared with other fleet safety management practices, and that the use of advanced technology collision avoidance systems (e.g., forward/rear obstacle detection) was rated below average. Of course, these systems are likely to become more popular and highly regarded as the technology is advanced in the coming years. Depending upon the technology, driver acceptance and proper use of the equipment may be an issue. A study by Penn + Schoen Associates, Inc. (1995), found that commercial drivers were often skeptical of on-board technologies that they had not yet used and were especially wary of technologies perceived as invasions of privacy (e.g., monitoring systems) or as diminishing the role of driver judgment (e.g., driver advisory or warning systems). Driver resistance must be overcome if the full promise of these technologies is to be realized (Roetting et al., 2003). V-57 EXHIBIT V-16 Per-Vehicle Life-Cycle Crash Costs for Passenger Cars (PCs), Light Trucks/Vans (LT/Vs), Combination-Unit Trucks (CUTs), and Single-Unit Trucks (SUTs) Source: Wang et al., 1999 0 10000 20000 30000 40000 50000 60000 70000 80000 PCs LT/Vs CUTs SUTs Motor Vehicle Type Li fe -C yc le Cr as h Co st s ($) EXHIBIT V-17 Strategy Attributes for Promoting Development and Deployment of Truck Safety Technologies Attribute Description Technical Attributes Target Expected Effectiveness This strategy targets truck manufacturers, motor carrier fleet owners, and safety directors. Effectiveness depends upon the safety device installed. Different technologies address different safety problems and target crashes, so comparisons are difficult. (continued on next page)

SECTION V—DESCRIPTIONS OF STRATEGIES V-58 EXHIBIT V-17 (Continued) Strategy Attributes for Promoting Development and Deployment of Truck Safety Technologies Attribute Description Keys to Success Potential Difficulties Appropriate Measures and Data Associated Needs Organizational and Institutional Attributes Organizational, Institutional, and Policy Issues Issues Affecting Implementation Time Costs Involved Training and Other Personnel Needs Legislative Needs Other Key Attributes Because this strategy is focused upon new technologies, their impact on fleet safety is initially going to be considered experimental. As noted above, there is often resistance to new technologies. Except for cases involving mandated technologies or safety designs, pilot studies are advisable before full implementation. This will allow fleet operators a better basis to judge their cost-effectiveness, as well as to provide experience for drivers to assess the reality of their concerns. Other factors being equal, safety design enhancements to combination-unit trucks are likely to have greater per-vehicle benefits than similar enhancements to single-unit large trucks because of the high life-cycle crash costs associated with combination- unit trucks. Success will be proportional to market penetration. Some vehicle-based, heavy-truck safety technologies require active fleet management monitoring and management for maximum benefits. Reliability and maintenance problems, driver nonacceptance, and driver misuse of devices (e.g., coming to rely on collision warning systems) are among the problems that may occur. Process measures would include the number of large trucks equipped with various safety technologies or improved safety designs. Outcome measures would include effects on heavy-truck, out-of-service rates and crashes. Measures of exposure will also be needed. Because of the gradual market penetration process, rapid changes in these outcome statistics are not likely. Active management monitoring and evaluation of effects on driver safety behavior are needed. Initially, this should be through a pilot test program. Some devices, such as anti-lock brakes and enhanced conspicuity lighting on trailers, have been mandated by NHTSA. Many others are not mandatory and are purchased at the discretion of fleet owners and managers. Complete penetration of new vehicle-based technologies in the heavy-truck fleet is likely to take two decades or more. The average operational life of large trucks is nearly 15 years, and most new safety technologies are ordered and installed as new vehicles are purchased. Costs depend upon the technology involved. In general, costs will decrease over time as a particular technology matures, manufacturing methods improve, and economies of scale are experienced. Cost elements include capital, operating and maintenance, and training of the drivers and those who handle maintenance and repair. Some devices require driver training, or at least a brief orientation. For example, anti- lock brakes are effective only if drivers learn how to use them properly. In addition, training will be needed for those who maintain and repair devices. Ordinarily, legislation is not required. However, some safety design changes and technologies are mandated by NHTSA in the FMVSSs. None identified.

SECTION V—DESCRIPTIONS OF STRATEGIES Information on Agencies or Organizations Currently Implementing this Strategy Among the organizations involved nationally in promoting the use of truck safety technologies are NHTSA, FMCSA, the U.S. DOT Intelligent Transportation System Joint Program Office, the Truck Manufacturers Association, SAE (http://heavyduty.sae.org/), and the American Trucking Associations Truck Maintenance Council. A Truck Manufacturers Guide Web site (www.cojoweb.com/truck_manuf.html) contains links to various heavy-truck and trailer manufacturers and equipment vendors. The Transportation Research Board has a number of committees active in commercial vehicle design and other motor carrier safety issues. These include the Committee on Motor Vehicle Size and Weight (AT055) and the Committee on Truck and Bus Safety Research (ANB70). Information on these and other truck transportation committees can be found at http://www.trb.org/directory/comm_homepages.asp. V-59