[Federal Register: May 2, 2000 (Volume 65, Number 85)] [Proposed Rules] [Page 25539-25611] From the Federal Register Online via GPO Access [wais.access.gpo.gov] [DOCID:fr02my00-26] [[Page 25539]] ----------------------------------------------------------------------- Part II Department of Transportation ----------------------------------------------------------------------- Federal Motor Carrier Safety Administration ----------------------------------------------------------------------- 49 CFR Parts 350, et al. Hours of Service of Drivers; Driver Rest and Sleep for Safe Operations; Proposed Rule [[Page 25540]] ----------------------------------------------------------------------- DEPARTMENT OF TRANSPORTATION Federal Motor Carrier Safety Administration 49 CFR Parts 350, 390, 394, 395, and 398 [Docket No. FMCSA-97-2350; formerly FHWA-97-2350 and MC-96-28] RIN 2126-AA23 Hours of Service of Drivers; Driver Rest and Sleep for Safe Operations AGENCY: Federal Motor Carrier Safety Administration (FMCSA), DOT. ACTION: Notice of proposed rulemaking (NPRM); request for comments. ----------------------------------------------------------------------- SUMMARY: The FMCSA is proposing to revise its hours-of-service (HOS) regulations to require motor carriers to provide drivers with better opportunities to obtain sleep, and thereby reduce the risk of drivers operating commercial motor vehicles (CMVs) while drowsy, tired, or fatigued to reduce crashes involving these drivers. This action is necessary because the FMCSA estimates that 755 fatalities and 19,705 injuries occur each year on the Nation's roads because of drowsy, tired, or fatigued CMV drivers. The regulations proposed in this document would: First, revert to a 24-hour daily cycle, and a 7-day weekly cycle. Second, adjust the work-rest requirements for various types of operations. Third, emphasize rest. Require for long-haul and regional drivers a period of 10 consecutive hours off duty within each 24-hour cycle, and two hours of additional time off in each 14-hour work period within each 24-hour cycle. Fourth, require weekends, or their functional equivalent, to include at a minimum a rest period that includes two consecutive periods from 11:00 p.m. to 7:00 a.m. Fifth, require the use of electronic on-board recorders (EOBRs) in CMVs used by drivers in long-haul and regional operations. DATES: You must submit your comments to this NPRM no later than July 31, 2000; however, late comments will be considered to the extent practicable. ADDRESSES: Signed, written comments should refer to the docket number appearing at the top of this document and must be submitted to the Docket Clerk, U.S. DOT Dockets, Room PL-401, 400 Seventh Street, SW., Washington, DC 20590-0001. Written comments and data may also be submitted electronically by using the submission form at http:// dmses.dot.gov/submit/BlankDSS.asp. All comments received will be available for examination at the above address between 9 a.m. and 5 p.m., e.t., Monday through Friday, except Federal holidays. Those desiring notification of receipt of comments must include a self- addressed, stamped envelope or postcard. FOR FURTHER INFORMATION CONTACT: For information on the proposed rule: Mr. David Miller or Ms. Deborah Freund, Federal Motor Carrier Safety Administration, (202) 366-1790, and Mr. Charles Medalen, Office of the Chief Counsel, Federal Highway Administration, (202) 366-1354. For information on the public hearings: Mr. Stan Hamilton, Federal Motor Carrier Safety Administration, (202) 366-0665. For information about submitting comments and data electronically: DMS Web staff at Mail.Dockets@tasc.dot.gov, Department of Transportation, 400 Seventh Street, SW., Washington, DC 20590-0001. SUPPLEMENTARY INFORMATION: Table of Contents I. Electronic Availability of This NPRM II. Introduction III. The Safety Problem IV. A Brief History of the Hours-of-Service Rules A. The ICC's Original Rules B. Immediate Changes to the HOS Rules C. 1962 Amendments D. Exemptions E. Developments in the 1970's and 1980's V. Comments to the ANPRM VI. FMCSA Response to Comments and Research Cited VII. HOS Regulation Development Process A. Research Findings 1. The workday should be more regular: Maintenance of circadian rhythm 2. The driver should be afforded more opportunity for daily and weekly sleep 3. Driving hours in any duty shift should generally not exceed 12 hours 4. The time of day when driving is performed should be considered 5. Non-compliance by drivers and motor carriers increases the potential for adverse safety outcomes B. Guiding Principles for Regulatory Improvement C. Types of Motor Carrier Operations D. Regulatory Options E. The Expert Panel F. Recordkeeping Requirements 1. Time Records 2. Electronic On-Board Recorders (EOBRs) G. Supporting Document Requirements 1. 1998 Notice of Proposed Rulemaking 2. Comments to Docket FHWA-98-3706 (Supporting Documents) 3. FMCSA's Response to the Comments on the Supporting Documents NPRM 4. Modified Supporting Documents Proposal H. Revised Regulatory Options I. Benefits and Costs 1. Crash Reduction 2. Paperwork Reduction 3. Total Benefits 4. Quantitative Costs 5. Small Business Costs 6. Qualitative Impacts 7. Benefits and Costs Combined J. The Option Selected to Propose VIII. Additional Petitions Received IX. Implementation X. Additional Proceedings XI. Section-by-Section Evaluation XII. Rulemaking Analysis and Notices Regulatory Identification Number Motor Carrier Safety Act Executive Order 12866 (Regulatory Planning and Review) and DOT Regulatory Policies and Procedures Regulatory Flexibility Act Unfunded Mandates Reform Act of 1995 and Executive Order 12875 (Enhancing the Intergovernmental Partnership) Paperwork Reduction Act National Environmental Policy Act Executive Order 12988 (Civil Justice Reform) Executive Order 13045 (Protection of Children) Executive Order 12630 (Taking of Private Property) Executive Order 13132 (Federalism Assessment) Executive Order 12372 (Intergovernmental Review) I. Electronic Availability of This NPRM Internet users may access this NPRM and all comments received by the U.S. DOT Dockets, Room PL-401, by using the universal resource locator (URL): http://dms.dot.gov. It is available 24 hours each day, 365 days each year. Please follow the instructions online for more information and help. An electronic copy of this document may be downloaded by using a computer, modem, and suitable communications software from the Government Printing Office's Electronic Bulletin Board Service at (202)512-1661. Internet users may reach the Office of the Federal Register's home page at: http://www.nara.gov/fedreg and the Government Printing Office's web page at: http://www.access.gpo.gov/nara. Internet users may also find this NPRM at the FMCSA's Motor Carrier Regulatory Information Service (MCREGIS) website for proposed rules at: http://mchs.fhwa.dot.gov/rulesregs/fmcsr/rulemakings.htm#proposedrule. II. Introduction There is general consensus that modifications to current HOS regulations would substantially improve CMV safety by reducing the fatigue factor in CMV-involved crashes. There is evidence that many crashes occur as a result of CMV driver error, that driver error is often the result of inattention, [[Page 25541]] that inattention can often be the result of fatigue, that the fatigue which causes inattention is often related to sleep deprivation, and that sleep deprivation is often related to working conditions of drivers. This proposal would make the HOS regulations more effective by requiring motor carriers and drivers to adhere to the following six standards and enforcing them: 1. Promote scheduling, dispatching, and operating practices minimizing the use of tired, inattentive drivers. 2. Make available for each driver a consecutive minimum off-duty period of time each workday and workweek for the purpose of obtaining restorative sleep. 3. Make available for each driver an additional minimum off-duty period of time each workday, during the workday or afterwards, to allow a driver to tend to personal necessities and rest at the driver's discretion. 4. Empower the driver to accept or refuse dispatch or continuation of a trip based upon the driver's assessment of his/her alertness level. 5. Enhance motor carriers' and CMV drivers' knowledge and use of safety techniques, devices, and practices that avoid driver impairment due to lack of sleep. 6. Require the use of automated time EOBR technology to monitor the work-rest cycles of long-haul and regional drivers and compliance with the rules, as well as encourage the use of technology for other drivers. The basic HOS rules have been in effect in their current form since 1962, and controversial for even longer than that. The rules allow CMV drivers in interstate commerce to drive up to 10 hours after 8 consecutive hours off duty. After being on duty for 15 hours, a driver may not drive without taking another 8 consecutive hours off duty. Weekly limits provide that if a motor carrier does not operate CMVs every day of the week, its drivers may not drive after being on duty 60 hours in 7 consecutive days; if the carrier operates CMVs every day, its drivers may not drive after being on duty 70 hours in 8 consecutive days (49 CFR 395.3). Although the charge is sometimes made that the FMCSA is pursuing an ill-conceived ``one size fits all'' policy toward the highly diversified motor carrier industry, the HOS rules in fact include a number of exceptions for specific situations or operations (see Sec. 395.1). It has become increasingly clear, however, that a complete reevaluation of the HOS rules is needed. America's transportation system has changed significantly since 1962, and even more fundamentally since the late 1930's, when the Interstate Commerce Commission (ICC) adopted the first HOS rules. Long-haul truckers in the 1930's could average only 25 miles per hour (mph)--the top speed was 40 mph--and the best daily run was about 250 miles (11 M.C.C. 203). The construction of the Interstate Highway System has contributed to significantly higher traffic speeds and volumes. Trucking, once a relatively minor adjunct to the railroads, has become the dominant form of transportation for most commodities. Much of the nation's truck traffic moves on the Interstates and other high-speed roads, sometimes for very long distances. Increased exposure to the risk of accidents follows automatically from annual increases in the number of trucks and other vehicles on the road and in total vehicle miles of travel (VMT). The high volume and speed of traffic on the Interstates and many other roads require a high level of driver alertness, for the sheer mass of a truck can make it deadly when accidents occur. Of course, trucks also operate in local or regional environments, often in heavy traffic, and drivers are required to perform an ever-wider range of duties. The results of scientific research into fatigue causation, sleep, circadian rhythms, night work, and other matters were unavailable decades ago when the HOS rules were formulated. Many people have indicated their concern over driver fatigue, and their concomitant belief that the present HOS regulations do not adequately ensure that drivers are rested. Driver fatigue was voted the number one safety concern of the Federal Highway Administration's (FHWA) 1995 Truck and Bus Safety Summit, a meeting of over 200 drivers, motor carrier representatives, government officials, and safety advocates. The National Transportation Safety Board (NTSB) has also asked the FMCSA to investigate driver fatigue. On June 1, 1999, the NTSB, recognizing that fatigue is an issue which affects all transportation modes, issued the following recommendation to the Department of Transportation: Require the modal administrations to modify the appropriate Codes of Federal Regulations to establish scientifically based hours-of-service regulations that set limits on hours of service, provide predictable work and rest schedules, and consider circadian rhythms and human sleep and rest requirements. Seek Congressional authority, if necessary, for the modal administrations to establish these regulations. The FMCSA had already devoted several years' of work toward the development of this NPRM at the time the NTSB issued its recommendation. The FMCSA believes that the revised HOS rules proposed today will reduce the acute and cumulative fatigue which appears to beset many drivers and thus prevent a significant number of crashes and fatalities, while limiting major compliance costs on those segments of the motor carrier industry that have the lowest fatigue-related CMV crashes and focusing the major compliance costs on those segments with the highest fatigue-related CMV crashes. The FMCSA's jurisdiction over the HOS regulations for motor carriers and drivers is specified in Table 1. Motor carriers and drivers are subject to applicable State motor vehicle and highway safety laws and regulations, regardless of whether the motor carriers or drivers are subject to any or all of the FMCSRs. In October, 1999, the Secretary of Transportation rescinded the authority previously delegated to the Federal Highway Administrator to perform motor carrier functions and operations, and to carry out the duties and powers related to motor carrier safety, that are statutorily vested in the Secretary. That authority was redelegated to the Director of the Office of Motor Carrier Safety (OMCS), a new office within the Department (see, 64 FR 56270, October 19, 1999, and 64 FR 58356, October 29, 1999). The OMCS had previously been the FHWA's Office of Motor Carriers (OMC). The Motor Carrier Safety Improvement Act of 1999 established the Federal Motor Carrier Safety Administration (FMCSA) as a new operating administration within the Department of Transportation, effective January 1, 2000 (Public Law 106--159, 113 Stat. 1748, December 9, 1999). The Secretary therefore rescinded the motor carrier authority delegated to the Director of the OMCS and redelegated it to the Administrator of the FMCSA (65 FR 220, January 4, 2000). The staff previously assigned to the FHWA's OMC, and then to the OMCS, are now assigned to the FMCSA. The motor carrier functions of the FHWA's Resource Center's and Division (i.e., State) Offices have been transferred without change to the FMCSA Resource Centers and FMCSA Division Offices, respectively. For the time being, all phone numbers and addresses are unchanged. Similarly, rulemaking activities begun under the auspices of the FHWA and continued under the OMCS will be completed by the FMCSA. The recent redelegations do not affect the validity of the November 5, 1996, Advance Notice of Proposed [[Page 25542]] Rulemaking (ANPRM) in this proceeding (61 FR 57252). All comments to that docket have been transferred to the new FMCSA docket and have been considered in preparing this document. The NPRM has been under development since the June 30, 1997, ANPRM docket closing date. Although the FMCSA has attempted to remove all non-relevant present- tense references to the FHWA and the OMCS, any that remain should be considered references to the FMCSA. Table 1.--Applicability of FMCSA Hours of Service of Drivers Rulemaking ------------------------------------------------------------------------ In interstate In intrastate If you operate a: commerce commerce ------------------------------------------------------------------------ CMV--A motor vehicle(s) that has You must comply You are not any of the following four with all FMCSA subject to the characteristics: HOS \2\ FMCSA HOS. You 1. A GVW, GVWR or GCWR \1\ of at requirements and may be subject least 4,537 kilograms (10,001 are subject to to proposals pounds); or proposals made made in this 2. Is designed or used to transport in this NPRM. NPRM, if your more than 8 passengers, including State or local the driver, for compensation; or government 3. Is designed or used to transport adopts final more than 15 passengers, including rules based on the driver, and is not used to these proposals transport passengers for in order to compensation; or participate in 4. Is used to transport hazardous the Motor materials in quantities requiring Carrier Safety the vehicle to be marked or Assistance placarded under the Hazardous Program, 49 CFR Materials Regulations (49 CFR part part 350. 172, subparts D & F). ------------------------------------------------------------------------ \1\ GVW, GVWR, and GCWR are acronyms for Gross Vehicle Weight, Gross Vehicle Weight Rating, and Gross Combination Weight Rating, respectively. See 49 CFR 390.5. \2\ As noted in Sec. 390.3(f) and specifically provided elsewhere in the FMCSRs, the following six categories of CMVs and drivers operating these CMVs are exempt from the FMCSRs, in whole or in part: (1) The occasional transportation of personal goods by individuals not for compensation (such as moving your own household) are exempt from all the FMCSRs. (2) The transportation of children and/or school personnel from home to school and from school to home are only subject to the driver CDL and alcohol and controlled substance requirements of 49 CFR Parts 382 and 383. (3) Transportation performed by the Federal government, a State, any political subdivision of a State, or an agency established under an agreement between States that has been approved by the U.S. Congress are only subject to the driver CDL and alcohol and controlled substance requirements of 49 CFR Parts 382 and 383. (4) CMVs used wholly on private property not open to public travel (such as yard hostlers and yard tractors in a motor carrier's terminal) are exempt from all the FMCSRs. (5) Fire trucks, ambulances, and rescue vehicles involved in emergency related operations, unless a State exempts driver of such vehicles (See 49 CFR 382.103(d) and 383.3(d)), are only subject to the driver CDL and alcohol and controlled substance requirements of 49 CFR Parts 382 and 383. (6) Transportation of human corpses and sick and injured persons. (7) The operation of CMVs designed to transport less than 16 passengers, including the driver, until March 6, 2000. BILLING CODE 4910-22-P [[Page 25543]] [GRAPHIC] [TIFF OMITTED] TP02MY00.000 [[Page 25544]] [GRAPHIC] [TIFF OMITTED] TP02MY00.001 BILLING CODE 4910-22-C [[Page 25545]] III. The Safety Problem While nearly everyone who has studied the current HOS rules agrees that they could and should be improved, it is difficult to reach consensus on alternatives because the extent and nature of the problem are unclear. This was acknowledged in a June 1, 1999, letter from Jim Hall, Chairman of the NTSB, to DOT Secretary Rodney E. Slater. Chairman Hall said, among other things: Fatigue has remained a significant factor in transportation accidents since the Safety Board's 1989 recommendations were issued. Although generally accepted as a factor in transportation accidents, the exact number of accidents due to fatigue is difficult to determine and likely to be underestimated. The difficulty in determining the incidence of fatigue-related accidents is due, at least in part, to the difficulty in identifying fatigue as a causal or contributing factor in accidents. There is no comparable chemical test for identifying the presence of fatigue as there is for identifying the presence of drugs or alcohol; hence, it is often difficult to conclude unequivocally that fatigue was a causal or contributing factor in an accident. In most instances, one or more indirect or circumstantial pieces of evidence are used to make the case that fatigue was a factor in the accidents. This evidence includes witness statements, hours worked and slept in the previous few days, the time at which the accident occurred, the regularity or irregularity of the operator's schedule, or the operator's admission that he fell asleep or was impaired by fatigue. Despite the difficulty in identifying fatigue as a causal factor, estimates of the number of accidents involving fatigue have been made for the different modes of transportation; the estimates vary from very little involvement to as high as about one-third of all accidents. Although the data are not available to statistically determine the incidence of fatigue, the transportation industry has recognized that fatigue is a major factor in accidents. Further, the Safety Board's in-depth investigations have clearly demonstrated that fatigue is a major factor in transportation accidents. The objective of this proposal is to reduce the number of fatigue- related truck and motorcoach crashes. The overall benefit therefore will depend on the effect this proposal will have on reducing the current number of these crashes. For purposes of this proposal, the FMCSA has assumed that bus drivers operate in ways similar to truck drivers. The FMCSA requests comments about the accuracy of this assumption. Although there are research studies in the docket concerning the performance of bus drivers suffering from fatigue, the FMCSA could find no research studies on which to distribute bus drivers subject to FMCSA jurisdiction among different operations for the safety and benefit-cost analyses in the docket. There are significant differences in published estimates of the number and proportion of fatigue-related CMV crashes. Much of this results from the differing analytical approaches used, particularly differences in the set of crashes analyzed. Generally speaking, these studies can be divided into two classes: those relying on large-scale accident data files, and those based on more intensive analysis of a smaller number of crashes. The FMCSA and the National Highway Traffic Safety Administration (NHTSA) have conducted several fatigue-related CMV crash studies using large-scale data bases, primarily the Fatality Analysis Reporting System (FARS) and the General Estimates System (GES). These databases, which are managed by NHTSA, are based largely (but not exclusively) on police accident reports (PARs). Most police accident forms contain a field for driver contributing factor, and generally among the choices are driver fatigue, drowsiness, or asleep at the wheel. In most analyses, crashes in which one of these fields is checked are classified as fatigue-related. Crash analysts frequently criticize use of PARs for fatigue analysis, as they believe that PARs understate the true extent of fatigue. Police face a number of difficulties in determining whether fatigue contributed to a crash. First, the responding officer's primary concern is assisting crash victims and restoring the flow of traffic. Investigating the causes of the crash is often a second (or lower) level concern. Second, few police officers are trained in crash reconstruction and, consequently, they are unable to conduct a detailed investigation of the physical and mechanical evidence. Also, it is unclear what should be counted as a fatigue-related crash. Clearly all crashes where fatigue is cited should be included, but there are other crashes where fatigue may play a less direct role. Crashes involving inattention, distraction, or other driver failures may be related to fatigue, as fatigued drivers are more prone to various types of mental error. These errors are major causal factors in crashes. The ``Tri-Level Study of the Causes of Traffic Accidents,'' Treat et al. (1979), is perhaps the most in-depth study ever performed in the United States on crash causation. This was principally a study of automobile drivers and their crashes. It found that ``recognition failure'' was involved in 56 percent of the crash cases analyzed. ``Recognition failure'' in this study meant: (1) Improper lookout, including faulty visual surveillance and ``looked but did not see;'' (2) inattention, including preoccupation with competing thoughts; (3) internal or (4) external distractions, including attention to competing events, activities, and objects in and out of the vehicle. While driver-drowsiness/fatigue was found to be a certain or probable factor in 2 percent of the cases, 23 percent involved faulty visual surveillance, 15 percent involved inattention, and 13 percent involved distraction. The FMCSA believes the study is generally applicable to CMV drivers and their crashes because the agency believes both CMV and automobile drivers are susceptible to driving related problems associated with visual surveillance, inattention, and distraction. More recent studies have also found high levels of inattention and distraction. In a study of nearly 700 Crashworthiness Data System (CDS) and GES crashes, Najm et al. (1995) determined that recognition errors were the primary causes of 45 percent of the cases studied, compared to 3.7 percent primarily due to driver drowsiness. General Motors scientists reviewed over 1,000 PARs from the State of Michigan, and reported that 17 percent were attributable to ``daydreaming'' and 18 percent to improper lookout, with just 1 percent due to ``dozing.'' Deering (May 17, 1994). A recent study by the United States Coast Guard also suggests that direct measurement of fatigue may understate its true extent, U.S. Coast Guard (1997)(MSC 68/INF.11). Coast Guard researchers developed a ``fatigue index,'' based on the number of fatigue symptoms reported, and the number of hours worked and slept in the 24 hours prior to the incident. Using this formula increased the percentage of critical vessel cases categorized as fatigue related from 1.2 percent to 16 percent. Critical vessel cases involve significant property damage or the loss of the vessel. Critical casualty cases involve personnel injuries. For critical casualty cases, the fatigue index resulted in an adjustment increase from 1.3 to 33 percent. These reports indicate the need to be more expansive and inclusive when defining fatigue-related incidents, as well as the likelihood that fatigue statistics based solely on accident reports underestimate the true extent of the fatigue problem. Fatigue increases the likelihood that a driver will not pay sufficient attention to driving or commit other mental errors. As discussed above, in-depth studies of crashes have found that inattention and other mental lapses [[Page 25546]] contribute to as much as 50 percent of all crashes. While fatigue may not be involved in all these crashes, it clearly contributes to some of them. The agency tentatively estimates that 15 percent of all truck- involved fatal crashes are ``fatigue-relevant,'' that is, fatigue is either a primary or secondary factor. This includes the 4.5 percent of fatal crashes where fatigue is directly cited, and another 10.5 percent where it contributes to other mental lapses, which then result in a crash. The FMCSA conducted some sensitivity analysis to determine the impact of different baseline levels of fatigue. Table 2 shows the FMCSA's estimate of the number of fatigue-related crashes by operational type. Table 2.--Estimated Annual Number of Fatigue-Related CMV Fatalities and Injuries ---------------------------------------------------------------------------------------------------------------- Local Other Long haul Unknown Total ---------------------------------------------------------------------------------------------------------------- Fatalities..................................... 61 166 480 49 755 Injuries....................................... 1,581 4,319 12,532 1,273 19,705 Fatal crashes.................................. ........... ........... ........... ........... 645 Injury crashes................................. ........... ........... ........... ........... 13,519 Total...................................... 1,642 4,485 13,012 1,322 20,460 ---------------------------------------------------------------- ---------------------------------------------------------------------------------------------------------------- Long-haul operations account for two-thirds of all fatalities, excluding those for which the length of haul was unknown. Only one half of one percent of fatalities and injuries occur in a crash with a truck whose driver has been reported driving 12 or more hours, although, as discussed above, the true figure is likely to be higher. Other research also indicates that local drivers are less likely to be involved in a fatigue-related crash than long-haul drivers. An analysis of the fatal crash rate and mileage figures from the University of Michigan Transportation Research Institute's (UMTRI) Trucks Involved in Fatal Accidents (TIFA) database and the Bureau of the Census' 1992 Truck Inventory and Use Survey (TIUS) shows a dramatic difference in the crash experience of local and other trucks. Local, single-unit straight trucks had an average of 0.0022 fatigue-related fatal involvements per 1000 registered trucks. The comparable figure for long-haul tractor-trailers was 0.0781, approximately 35 times greater. On a per-mile basis, long-haul trucks were almost 20 times more likely to be involved in a fatigue-related crash. Massie et al. (1997). Time of Day Without reliable exposure data disaggregated by operational type, it is difficult for the FMCSA to confirm that truck travel matches the crash distribution. The UMTRI analyzed the relationship between the risk of fatigue given a fatal accident involvement and the risk of fatigue per VMT, by truck body type and trip type, using VMT data from the TIUS. This analysis, which is partially reprinted in Appendix B to the Preliminary Regulatory Evaluation (PRE) that is in the docket and incorporated by reference in this NPRM, suggests that the relative risk of fatigue given a fatal accident involvement is a good predictor of the risk of fatigue per VMT. This is important because exposure data are not available for many relevant variables (such as time of day and hours driving). Chart 1 shows the distribution of trucks involved in fatal crashes. It generally mirrors truck traffic, with a mid-afternoon peak when the data show truck travel to be the highest. Chart 2 shows the distribution of fatigue involvement, which is quite different than overall involvement in crashes. Fatigue peaks between 4 a.m. and 6 a.m. Chart 3 combines data from both Chart 1 and 2 to show the relative risk of a fatigue involvement, given that a fatal crash occurs. Chart 3 closely resembles Chart 2, indicating that the incidence of fatigue is more variable than that of crashes. Relative risk looks the same for both long-haul and other trucks. Chart 4 shows that both types of operations have peak relative risks between 4 a.m. and 6 a.m. There is no significant difference between the time of day distribution of straight and combination trucks. Hours Driving Chart 5 shows the relative risk of a fatigue-related fatal crash by the number of hours of driving. Data on hours driving up to 1992 came from phone interviews and from the FHWA's form MCS-50T accident reports. Motor carriers involved in certain accidents were required to complete these forms up to 1992. Since elimination of the requirement to file MCS-50T accident reports in 1993, data on hours driving come entirely from phone interviews by UMTRI researchers. The interview source is the owner of the truck, so the agency expects some under- reporting for hours above the current limits. About one quarter of all respondents refused to answer this question, much higher than the percent missing for any other question. Nonetheless, the data clearly show the impact of extra hours driving on the likelihood of fatigue being cited in a crash. Not surprisingly, risk increases with time driven. Approximately 20 percent of the fatal crashes per year where fatigue is coded as a factor involve the driver being behind the wheel for 13 or more hours. There also appears to be a slight increase in the risk of fatigue- related crashes at 5 hours. This is difficult to discern in the following chart, but becomes apparent when looking at risks for long- haul drivers. Jovanis et al. (1991) found a similar pattern in their examination of crashes from one long-haul carrier, including both a bump at 5 hours and a more dramatic and consistent increase in crash risk after 8 hours. Lin et al. (1993). Lin noted a limitation in their analyses, and provide a caveat to the estimates of the odds ratios in the last driving hour category--a large number of non-crash trips are completed during the 8th or 9th hour of driving, but the authors' ``assumed failure time,'' defined as the expected time of involvement in an accident, would occur after this trip completion time. As noted above, long-haul trucks are involved in about 67 percent of all fatigue-related truck crashes. These vehicles also have a greater relative risk of fatigue involvement for almost any given number of hours driving. Chart 5 would not appreciably change if vehicles were broken down by trip type, except when hours driving exceed 11, where the small numbers of crashes yield some extremely high relative risk values. The distribution of crashes by vehicle type is not so clear-cut. Two-thirds of all trucks involved in fatal crashes between 1991 and 1996 were combination vehicles, including both tractor semi- [[Page 25547]] trailers and straight trucks pulling a trailer. These vehicles were also involved in four-fifths of all fatigue-involved fatal crashes, only slightly higher than the percentage of all fatal truck crashes. This suggests that truck body type is a good proxy for predicting fatigue. Long-haul combination vehicles account for about half of all fatal CMV crashes, but three-fourths of all trucks in fatigue-involved fatal crashes. Straight trucks in long-haul operations are more likely to be fatigue involved; although they represent just 7 percent of trucks involved in fatal crashes, they account for 14 percent of fatigue- involved trucks. The relative risk for drivers of these vehicles is almost 2, while it is closer to 1.5 for drivers of combination vehicles in long-haul operations. This over-representation may be partly due to drivers of straight trucks being unaccustomed to the rigors of long- haul operations. Injury Crashes Data on non-fatal crashes are even more limited than for fatals. All the factors militating towards under reporting of fatigue in fatal crashes are even more prevalent in non-fatals. In addition, because the best estimate of the number of non-fatal truck-involved crashes is based on a sample rather than a census (as is the case with fatal crashes), the FMCSA is not able to segregate these crashes by carrier operational type. Therefore, for this analysis, the agency uses the ratio of all injury crashes to fatal crashes as a proxy for fatigue- related injury crashes. The agency has also estimated that injury crashes follow the same patterns as fatal crashes, with overall crashes higher in the afternoon and fatigue-related crashes peaking between 4:00 a.m. and 5:00 a.m. To evaluate the consistency between fatal and injury crashes, the agency examined injury-crash data from Texas. The agency chose to review Texas for a number of reasons. First, it typically ranks among the highest in terms of fatal truck crashes, ensuring that the agency would have a large sample to analyze. Second, Texas reports a high proportion of fatigue in fatal truck crashes, which suggests the State is better at reporting fatigue. This analysis shows that injury crashes generally mirror the fatal crash distribution by time of day. No data are available on injury crashes by hours driving. Based on Texas' reports and the analysis of the general mirroring of injury crashes to fatal crash distribution, the FMCSA has determined that the distribution of injury crashes, both overall and fatigue-related, would follow the pattern exhibited by fatal crashes. The complete analysis has been placed in the docket. This analysis does not include property damage only (PDO) crashes. We were not able to find any reliable information on PDO crashes by trip distance, hours driving, or time of day. The FMCSA also believes that fatigue-related crashes tend to be more severe than non-fatigue- related crashes, so the number of fatigue-related PDO crashes is probably small. In any case, the damage from PDO crashes, whether fatigue related or not, is, by definition, minimal. In an analysis in another rulemaking, the FMCSA estimated that the average truck involved PDO crash costs society between $5,000 and $10,000. RIN (Regulatory Identification Number) 2125-AD27, Docket FHWA-1997-2222. To the extent that there are a sizable number of PDO crashes which would be affected by this proposal, overall benefits would be greater than the agency's estimate. See the PRE in the docket for a complete discussion of injury crashes. See the section headed ``XII. Rulemaking Analysis and Notices'' for more information about RINs. The FMCSA invites comment on any aspect of the safety problem, the data and estimates used by the FMCSA, and the conclusions reached as a result of the analyses. Please provide with your comments all data, studies, and reports you rely upon that you believe the FMCSA should use. IV. A Brief History of the Hours-of-Service Rules The following is a brief history of the HOS rules. The docket contains a more complete discussion of the rules and their relationship to the Fair Labor Standards Act (FLSA) and implementing regulations. A. The ICC's Original Rules The Motor Carrier Act of 1935 (MCA) (Public Law 74-255, 49 Stat. 543, August 9, 1935), in addition to authorizing far-reaching economic regulation of the trucking industry, directed the ICC to establish qualifications and maximum hours of service for drivers working for private and for-hire interstate property carriers and for-hire interstate passenger carriers. This authority is now codified at 49 U.S.C. 31502. The ICC published its safety proposals, including HOS limits, on July 8, 1936 (1 FR 738). In preparing its draft rules, the Commission examined all State HOS laws and regulations and solicited input from motor carriers and drivers. A formal rulemaking action, including hearings, was conducted before Division 5 of the ICC. (Hearings: 1 FR 1015, August 7, 1936; 1 FR 1510, October 2, 1936; 1 FR 2161, December 18, 1936.) On December 29, 1937, the ICC promulgated its final HOS regulations (effective July 1, 1938), along with detailed findings and explanations (3 M.C.C. 665, 3 FR 7, January 4, 1938). Concerning drivers' need for off-duty time, the ICC found: It is obvious that a man cannot work efficiently or be a safe driver if he does not have an opportunity for approximately 8 hours sleep in 24. It is a matter of simple arithmetic that if a man works 16 hours per day he does not have the opportunity to secure 8 hours sleep. Allowance must be made for eating, dressing, getting to and from work, and the enjoyment of the ordinary recreations. 3 M.C.C. 665, at 673 (1937). Under the regulations adopted by the ICC, motor carriers could not permit or require drivers to be on duty more than 15 hours out of 24; drivers were thus allowed at least 9 hours off duty every day. The limit was designed to give them an opportunity for a minimum of 8 hours of sleep. Within the 15-hour on-duty period, the ICC set a 12-hour maximum daily work period for drivers. Work was defined as ``loading, unloading, driving, handling freight, preparing reports, preparing vehicles for service, or performing any other duty pertaining to the transportation of passengers or property.'' The ICC intended the 3-hour difference between 15 hours on duty and 12 hours of work to be used for meals and rest breaks. The Commission also set a weekly on-duty limit of 60 hours in any 7 consecutive days or 70 hours in 8 consecutive days. B. Immediate Changes to HOS Rules Within a short time, however, representatives of organized labor (including the American Federation of Labor, the Teamsters, and the Machinists) petitioned for a stay of the original regulations. A few motor carriers made a similar request. The ICC agreed, and oral arguments were heard again. Labor wanted HOS limits of 8 hours per day and 48 hours per week. The ICC commented that: [T]here was no statistical or other information which would enable [us] to say definitely how long a driver can safely work. However, at the argument before us, the labor representatives particularly stressed the 15-hour limitation, contrasting such a tour of duty with the 8-hour day which is now so generally recognized as the normal standard for workers. The evidence before us clearly does not suffice to enable us to conclude that a duty period as low as 8 hours in 24 is required in the interest of safety. We may call attention, as did the division, to the contrast between factory operations, generally sustained in character, and the operation of busses and trucks, generally characterized by frequent stops for refreshments, gas, or rest, [[Page 25548]] or because of conditions encountered in highway and street traffic. The monotony or nervous and physical strain of driving such vehicles is alleviated by these breaks in the periods devoted to driving, and the period of actual work is considerably below the period on duty. 6 M.C.C. 557, at 561 (July 12, 1938) The Commission ultimately decided to change the 12-hour work limit in 24 hours to a 10-hour driving limit in 24 hours. Motor carriers were required to give drivers 8, rather than 9, consecutive hours off duty each day. That meant drivers could be kept on duty as much as 16 hours out of 24; the specific daily on-duty limit was rescinded. The 60- and 70-hour limits were unchanged (3 FR 1875, July 28, 1938). The ICC remarked that these rules were somewhat less flexible than the original HOS regulations, but considerably more flexible than the standards requested by organized labor. ``[A]s the great bulk of the trucking operations covered by these regulations are conducted on a 6-day basis,'' the report said, ``the practical effect of the weekly limitation is to provide a 10-hour day.'' 6 M.C.C. 562 (1938). The Commission reiterated a similar point elsewhere in its report on the amended rules. [I]t was strongly urged upon us [the ICC] that the daily and weekly maxima prescribed by division 5 would make it difficult to negotiate contracts for shorter hours, or for unorganized labor to hold the hours it has. It was said that already carriers have used the regulations prescribed in the prior report as a means of lengthening hours. Considerations other than those with which we may properly deal enter here, though we look with distinct disfavor on carriers or others who use regulations premised on safety as a means of defeating employees' efforts to improve their economic status. It is questionable, however, whether the practice has been or will be a serious one. The fact that we hereinafter prescribe 60 hours on duty as the weekly maximum should not interfere with the negotiations by organized labor of contracts providing for shorter hours. 6 M.C.C. 557, at 560 (July 12, 1938). The ICC's hope that the HOS rules would not be used to lengthen drivers' hours has not been borne out. That is a matter of some importance, since the FLSA, which generally required overtime pay after July, 1945 for more than 40 hours of work per week, included an exemption for motor carriers subject to the ICC's regulations. Amended only slightly, the exemption remains in effect today at 29 U.S.C. 213(b): (b) Maximum hours requirements: The provisions of section 207 of this title [Maximum hours] shall not apply with respect to-- (1) any employee with respect to whom the Secretary of Transportation has power to establish qualifications and maximum hours of service pursuant to the provisions of section 31502 of title 49 * * *. The result is that truckers engaged in interstate commerce work some of the longest hours known in this country, without the opportunity for time-and-a-half overtime pay beyond the 40th hour. On October 22, 1938 (3 FR 2533), the U.S. Department of Labor's (DOL) Wage and Hour Division (WHD) published regulations (29 CFR part 516) implementing section 11(c) of the FLSA (29 U.S.C. 211(c)). These regulations require motor carriers subject to the minimum wage provisions of the Act--which includes carriers subject to the jurisdiction of the ICC (and now the FMCSA)--to make, keep, and preserve time records showing when drivers start and stop work, as well as the total number of hours they work per day and week. C. 1962 Amendments The first, and in fact the only, fundamental change to the HOS rules since the late 1930's occurred in 1962 (89 M.C.C. 19, March 29, 1962, 27 FR 3553, April 13, 1962). For reasons it never explained clearly, the ICC retained the 8-hour off-duty requirement and the 10- hour driving limit, but dropped the 24-hour limit. This had profound effects. For example, a driver who came on duty and started driving at 12:01 a.m. Monday would have to stop driving at 10:00 a.m. If the driver then took 8 hours off duty, he or she could drive again from 6:00 p.m. to midnight, for a total of 16 hours on Monday. The previous rule would have limited the driver to a total of 10 hours driving time in any 24-hour period. The unintended consequences of this change are described below. D. Exemptions The 1938 revisions to the HOS rules were barely in place when the first request for an exception was filed. Several industry associations argued that the 10-hour driving limit should be extended when bad weather made it impossible to complete a normal run in 10 hours. The ICC allowed an extra 2 hours for ``unfavorable weather conditions.'' See 11 M.C.C. 203, January 27, 1939, 4 FR 475, January 31, 1939. This exception, slightly modified, is still available (49 CFR 395.1(b)(1) Adverse driving conditions). Other requests followed over the years; the exceptions granted by the ICC are codified at 49 CFR 395.1(b)(2) Emergency conditions, (c) Driver-salesperson, (d) Oilfield operations, (e)100 air-mile radius driver, (f) Retail store deliveries, (g) Sleeper berths, (h) State of Alaska, (i) State of Hawaii, and (j) Travel time. On July 30, 1991 (57 FR 33638), the FHWA published a final rule exempting motor carriers and drivers from most of the Federal Motor Carrier Safety Regulations (FMCSRs), including the HOS rules, while providing emergency relief during a declared regional or local emergency, and more limited relief for tow truck drivers responding to a police request to move wrecked or disabled vehicles (49 CFR 390.23). The emergency relief provision required drivers who had been on duty more than 60 hours in 7 days, or 70 hours in 8, while providing direct assistance to emergency relief efforts, to take 24 hours off duty before returning to normal driving in interstate commerce. On August 19, 1992 (57 FR 37504), the agency published a proposal to permit drivers to begin anew any on-duty period of 60 hours in 7 days or 70 hours in 8 days upon taking 24 consecutive hours off duty. The FHWA made this proposal to provide opportunities for improved efficiency in operations consistent with highway safety. On February 3, 1993 (58 FR 6937), the FHWA withdrew the proposal, having received virtually no substantive responses to the critical questions asked. Most of the 68,000 comments offered opinions unsupported by empirical (or even anecdotal) material. Except in general terms, there were no discussions of potential impacts on highway safety. The agency needed to know if 24 hours is sufficient time for a driver to obtain the rest necessary to resume driving safely after accumulating 60 to 70 on-duty hours in as short a period as 4.25 days. This question remained unanswered until the completion of Canadian studies (O'Neill, T. et al. (1999) and Smiley, A. & Heslegrave, R. (1997)). The agency discusses these studies later in this document. Citing the waiver authority enacted as part of the Motor Carrier Safety Act of 1984 (MCSA) (Public Law 98-554, October 30, 1984)(Sec. 206(f), 98 Stat., at 2835), representatives of many industries either filed petitions for waivers of the HOS regulations or contacted the agency about the possibility. Among the carriers requesting exemptions were those involved in utility operations (electricity, natural gas, television), farming and farm supplies, construction, drilling of blast holes for rock quarries, highway traffic marking, custom harvesting, ground water drilling, and transporters of produce, cement, and ready- mix concrete. The FHWA granted none of these requests [[Page 25549]] because the proponents were unable to provide the agency with sufficient data to show that the waiver would meet the statutory test: (1) Not contrary to the public interest and (2) consistent with the safe operation of CMVs. Section 345 of the National Highway System Designation Act of 1995 (Public Law 104-59, 109 Stat. 568, 613) (NHS Act) created a statutory exemption from all of the HOS provisions for individuals transporting crops and farm supplies during planting and harvesting seasons, and a more limited exemption (from the 60- and 70-hour rules) for drivers of utility service vehicles, CMVs transporting ground water well drilling rigs, and construction materials and equipment. The FHWA, however, was authorized to conduct rulemaking on the advisability of each of these exemptions (except that for water well drilling rigs). If the agency determined that an exemption would not be in the public interest and would have a significant adverse impact on the safety of CMVs, the exemption could be blocked before it went into effect, modified or revoked. The NHS exemptions apply only to drivers and motor carriers operating in interstate commerce, and the Act specifically denied preemptive effect to any of the exemptions. The States, therefore, are free to adopt or reject any of the HOS exemptions, whether for interstate or intrastate commerce, without jeopardizing their eligibility for Motor Carrier Safety Assistance Program (MCSAP) funding. The FHWA adopted all of the required exemptions on April 3, 1996 (61 FR 14677) (see 49 CFR 395.1(k) Agricultural operations, (l) Ground water well drilling operations, (m) Construction materials and equipment, and (n) Utility service vehicles), but deferred until a future date any rulemaking action to consider modifying or revoking them. While this notice was being prepared, the FHWA received a petition from the Advocates for Highway and Auto Safety (AHAS) seeking rulemaking to reevaluate the 1996 exemptions. By this NPRM, the agency is granting the AHAS petition, which has been placed in the docket. The agency will also be consolidating the Regulatory Identification Number (RIN) 2126-AA29 (formerly 2125-AE09) into this action. See the section headed ``XII. Rulemaking Analysis and Notices'' for more information about RINs. The FMCSA is proposing to modify the agricultural exemption and revoke all but one of the other NHS Act exemptions. Because Sec. 345 does not authorize rulemaking to reevaluate the exemption for ground water well drilling rigs, it will remain in effect. The utility and construction exemptions, however, would be revoked, and those operations would be required to comply with the proposed ``weekend'' provisions of the HOS rules for Type 4 or 5 operations, described in detail below. E. Developments in the 1970's and 1980's In 1970, when Congress authorized funds to be spent on research, the FHWA, which had then assumed the safety responsibilities previously exercised by the ICC, initiated inquiries into drivers' hours of service and fatigue. The research and subsequent rulemaking continued for the next ten years. On February 12, 1976, the FHWA published an ANPRM that discussed three possible options for regulatory revisions (41 FR 6275). Because the agency did not receive sufficient information to determine whether the HOS should be amended, it published a second ANPRM on May 22, 1978 (43 FR 21905) inviting comments on three different regulatory options. After reviewing the comments to the second ANPRM, as well as transcripts from six public hearings that generated more than 9,000 pages of testimony from 1,200 interested parties, the FHWA determined that none of the proposed regulatory options could be supported in its entirety. The FHWA then developed three new proposed HOS options and carried out detailed cost- effectiveness and other regulatory analyses as required by Executive Order 12044 (1978). Option I would have reduced the 10-hour driving limit to 8 hours, among other things. Option II would have eliminated the differences in time allowances for driving time, on-duty time, and duty tour by specifying each have a limit of 12 hours, among other things. Option III would have maintained the status-quo regulations, with one exception: it would have banned driving between midnight and 6:00 a.m. The cost to society of Option I was estimated to be $11.496 billion and its benefits to society to be $450 million. The cost of Option II was estimated to be $10.642 billion and its benefits $450 million. The cost of Option III would have been $11.019 billion and its benefits $63 million. The rulemaking was terminated on September 3, 1981. V. Comments to the ANPRM The FHWA published an ANPRM to amend the HOS regulations on November 5, 1996 (61 FR 57252). This action was mandated by section 408 of the ICC Termination Act (Public Law 104-88, December 29, 1995, 109 Stat. 803, 958). The FHWA nearing completion of several research projects and sought the results of other relevant research to consider in this effort. The FHWA requested comments on the current HOS regulations and sought assistance in locating any other relevant information, including research, operational tests, or pilot regulatory programs conducted anywhere in the world, that could be used in developing revised HOS rules for CMV drivers. The agency received 1,650 comments in response to the ANPRM. The strongest support for amending the rules came from truck drivers, although no demand for major increase or decrease in hours emerged from their comments. Many drivers commented they wanted to work fewer hours without loss in annual incomes, and many other drivers wanted to drive longer hours and earn higher annual incomes. The only major theme that most agreed on was that they would like to see the agency prohibit them from loading and unloading cargo. The specific concerns or issues raised by the commenters who discussed technical or economic issues are addressed in the following sections. The respondents represented 13 advocacy groups, 3 consultants to the industry, 32 individuals, including concerned citizens and spouses of drivers, 1 insurance company, 4 labor unions, 3 law firms, 1,159 motor carriers (including owner/operators and drivers operating for motor carriers), 7 motor carriers of passengers, 49 trade associations (including 1 motor coach association), 1 truck driving school, 3 universities (including specific research departments of various institutions), 7 federal government agencies or representatives (including Senators Robb, Abraham, Specter, and Helms, Representative Cliff Stearns, and the NTSB), 4 state government agencies, 1 local government, and 33 other entities (including respondents of indeterminate affiliation). Most respondents submitted comments that did not provide relevant scientific research or studies in support of their comments. The FMCSA considered additional comments submitted after the June 30, 1997, closing date. Following are general comments on information contained in the docket. 1. In September 1997, after the June 1997 closing date, the docket contained 572 responses to the 69 specific questions asked in the ANPRM. 2. Of the 572 question-specific responses, 127 included submissions of, [[Page 25550]] or references to, a specific report, study, or survey. 3. Of the 1,650 comments analyzed, 731 were form letters to the docket. There were two different form letters; and none of these letters provided any relevant scientific research or studies to support the comments. The first form letter, received from 152 motor carriers in the construction industry, indicated that driver productivity is constrained by weather, the time of the year, and daylight hours. Furthermore, these carriers emphasized that they almost always operate within a localized geographical area, and concluded that drivers in the construction industry be exempt from the HOS regulations that govern over-the-road drivers. The FMCSA consolidated the figures for these 152 carriers and found they operated 5,614 single-unit, straight trucks and 1,734 truck tractors. See Table 3. Table 3.--152 Construction Industry Comments ---------------------------------------------------------------------------------------------------------------- Comment Minimum Maximum Average per carrier ---------------------------------------------------------------------------------------------------------------- Typical Trip Miles................... 2 miles................ 700 miles.............. 43.55 miles. Typical Trip Time.................... 0.02 hours............. 12 hours............... 1.89 hours. Total Trips (over 3 years)........... 0 trips claimed........ 1,000,000 trips........ 101,107 trips. Total Miles (over 3 years)........... 60 miles............... 21,744,000 miles....... 2,232,038.89 miles. Accidents (over 3 years)............. 0 accidents............ 58 accidents........... 3.31 accidents. ---------------------------------------------------------------------------------------------------------------- A second form letter was submitted by drivers in the form of a survey. All respondents submitting these form letters indicated support for a change in the FMCSA HOS regulations. They also provided no relevant scientific research or studies to supported comments. Comments Identifying Potential Science-Based Alternatives All comments citing research studies as their basis for comments are discussed in this section. They are presented according to the regulatory elements they address. Off-Duty Time The Insurance Institute for Highway Safety (IIHS) recommended a minimum of 12 to 14 hours off duty per day, citing Wylie, C.D. et al. (1996) (the FHWA's Driver Fatigue and Alertness Study). The IIHS believes the rules must require a 14-hour minimum off-duty time after a driver has driven in the time period midnight through 6:00 a.m.; it bases this comment on Miller (1993). In a meeting on September 23, 1998, IIHS and National Safety Council officials came to the FHWA to discuss various HOS concerns. At that meeting, IIHS recommended a minimum 14-hour off-duty period, citing Lin et al. (1994, 1993), Jones and Stein (1987, 1989), Frith (1994), Saccomanno et al. (1996), NTSB (1995), and Summala and Mikkola (1994). IIHS provided copies of these studies and the FHWA/FMCSA filed them in the docket along with a summary of the meeting. The American Trucking Associations, Inc. (ATA) and Landstar Systems, Inc. support 10 hours per day off duty, also citing Wylie, C.D. et al. (1996). The International Brotherhood of Teamsters (IBT) recommends 9 hours off duty per day citing NTSB, (1995), while the National Sleep Foundation (NSF) presented a Department of the Army aviation study, Caldwell, J. (1997) and Belenky, G. (1994), suggesting that aviators obtain an average of 7 hours of sleep during a 24-hour period. ``Re-start'' Provisions (e.g., 24 Hour, 36 Hour, etc.) The ATA believes the FMCSA should implement a re-start provision and offered Woodward and Nelson (1974) and Krueger et al. (1987) to support its assertion that a 24-hour ``re-start'' after 36 to 48 hours of high workload is desirable. Sleeper-Berth Use The California Highway Patrol (CHP) believes Wylie et al. (1996) suggest that sleeping facilities, other than sleeper-berth equipment meeting the requirements of 49 CFR 393.76, would make enforcement difficult by requiring officers to determine whether alternate locations qualify as adequate sleep/rest locations. The AHAS believes short sleeper-berth periods should not be allowed to count toward minimum off-duty requirements and bases that comment on Wylie et al. (1996). AHAS recommends that crediting split sleeper-berth periods toward the 8-hour off-duty period be prohibited, also based on its interpretation of Wylie et al. (1996). The ATA and the Distribution and LTL Carriers Association (DLTLCA) support the continuation of sleeper-berth-type provisions, i.e., split sleep periods or dividing a consecutive off-duty period into two or more smaller periods with duty in between, citing Chiles (1968). The National Sleep Foundation (NSF) supports the continuation of sleeper- berth-type provisions, basing its position on work by Caldwell, J. (1997), Belenky, G. (1994), Bonnet (1994), and Dinges and Broughton (1989). The NTSB, citing research by Dinges (1989), and the IBT, citing NTSB (1995), however, oppose sleeper-berth-type off-duty hours. The Owner-Operator Independent Driver's Association (OOIDA) surveyed its members before it responded to the ANPRM. Its study, OOIDA (1997), shows 28 percent of its members split their off-duty time. A similar ATA survey, ATA (1997), shows that most driving teams split their sleep periods into 4 hours on duty and 4 hours off duty. The OOIDA survey shows that single drivers average 3 hours 52 minutes for the first sleeper berth period and 4 hours 20 minutes for the second. The ATA survey shows 5 percent of team drivers use the sleeper berth while the CMV is in operation. The OOIDA survey shows a figure of 11 percent for team drivers, and also reports that these team drivers obtain an average of 5 hours 38 minutes driving and 5 hours 14 minutes of rest in each off-duty sleeper-berth period. Rest Breaks or Napping The AHAS and the ATA recommend that the regulations require the practice of taking naps during long trips. AHAS cited Wylie et al. (1996) and ATA cited its own survey, ATA (1997). On-Duty Time The OOIDA believes findings in Wylie, et al. (1996) support its recommendation for the FMCSA to allow a maximum of 15 hours per day on duty. ROCOR Transportation favors between 14 and 15 hours maximum on duty time, and also cites Wylie, et al. (1996). The ATA and Landstar Systems, Inc. recommend 14 hours on duty time with no distinction between the driving and on-duty times and bases its recommendation on Wylie, et al. (1996) and Harris, et al. (1972). Driving Time The CHP believes Wylie et al. (1996) shows that 10 hours driving time is not [[Page 25551]] excessive if managed properly. Citizens for Reliable and Safe Highways (CRASH) recommends a maximum of 10 hours driving time in a 12-hour on- duty time period, citing Fuller (1983), Lin et al. (1993), and Jones and Stein (1987). In the meeting on September 23, 1998, discussed above under the subheading ``off duty time,'' IIHS also recommended a maximum 10-hour driving period, citing the same studies, Lin et al. (1994, 1993), Jones and Stein (1987, 1989), and three additional studies, Frith (1994), Saccomanno et al. (1996), NTSB (1995), and Summala and Mikkola (1994). An IIHS official also stated during that meeting that the IIHS could accept an increase in driving hours if the IIHS could be assured the drivers would be afforded more hours off duty. Weekly On-Duty Limits The OOIDA cites Wylie et al. (1996) as support for its assertion that the weekly on-duty limit be eliminated. Vallenduuk Tissingh Advocaten (VTA) of the Netherlands cited a European regulation and contended the FMCSA should adopt the European example of allowing 84 hours on duty while limiting driving up to 56 hours. The ATA believes Wylie, et al. (1996) and Mackie and Miller (1978) support the current 60- and 70-hour rules. Do Wage Methods Affect HOS? The AHAS favors elimination of the section 13(b) exemption to the Fair Labor Standards Act and cites NTSB (1995), NTSB (1990), and Braver et al. (1992) as the basis for its assertion. The NTSB believes a link between payment by the mile and HOS violations is shown by its 1995 research study. CRASH cites Beilock (1995) as support for its suggestions that the FMCSA should incorporate compensation systems allowing drivers to generate reasonable and adequate incomes while upholding the law. VTA cites University of Groningen research (1986) showing drivers in Europe who are paid monthly salaries have fewer HOS violations than drivers paid by distance traveled. On-Board Monitoring Devices The IIHS cites its own research study, IIHS (1995), as well as Stein (1994) and Gronemeyer (1994), as the basis for its recommending a requirement for on-board monitoring devices. The NSF cites Braver et al. (1992), NTSB (1990), and IIHS (1995) as the basis for requiring such devices. The ATA cites Penn and Schoen (1995) in its assertion that on-board monitoring devices should not be required. Six comments provided cost estimates, although only two of the six comments cited cost estimates based on studies, specifically IIHS (1995) and Penn and Schoen (1995). The other four cost estimates came from the Log Truckers Association of Montana, John Cheeseman Trucking, Inc., ROCOR Transportation, and Rockwell Transportation Electronics. ROCOR stated it had 2,000 units in use costing about $3,000 per unit. It estimated a higher cost for smaller motor carriers. Though the other commenters did not state the basis for their estimate, the FMCSA presumes they are based on past business decisions or marketing and sales information. Types of Motor Vehicles Included The OOIDA cites Wylie et al. (1996) in asserting that the driving time limits should be applicable only to motor vehicles over 4,537 kg (10,000 pounds). The OOIDA believes the FMCSA should not propose any new regulations defining ``driving time'' applicable to motor vehicles up to 4,537 kg, because it believes Wylie et al. (1996) found that hours of driving are not a strong or consistent predicator of observed fatigue. Adverse Weather CRASH does not believe a driver should be allowed to continue to drive an extra two hours when the driver encounters adverse weather conditions after being dispatched, citing Fuller (1983), Lin et al. (1993), and Stein and Jones (1987). The OOIDA, on the other hand, cites Wylie et al. (1996) as support for its belief that the FMCSA should retain the ``adverse weather rule'' provision allowing drivers an extra two hours because, in such circumstances, drivers sometimes have difficulty finding adequate parking. Bobtailing En Route The OOIDA believes that Wylie, C.D. et al. (1996) shows that drivers should be allowed to bobtail to motels and restaurants while off duty in the vicinity of en route stops. Bobtailing, in the context provided in question 17 of the ANPRM, means a single tractor CMV operating without a trailer coupled to it. (Bobtailing can also mean the operation of a straight truck in certain industries or parts of the country. The FHWA did not ask any specific questions regarding research data related to use of CMVs for travel to motels and restaurants while off duty in the vicinity of en route stops, although a regulatory interpretation currently allows the practice.) Provisions Relating to Circadian Rhythms The AHAS cites Harris (1977), Hamelin (1980, 1981, 1987), Mackie and Miller (1979), Hertz and Eastham (1986), and Moore-Ede et al. (1988) as the basis for its assertion that any new FMCSA regulations must do the following three things: 1. Prevent rapidly rotating shifts. 2. Avoid scheduling that results in a long driving day ending deep within the early-morning circadian trough. 3. Limit the number of consecutive hours that CMV drivers must drive if assigned to nocturnal driving. The IIHS, NTSB, and the NSF believe a circadian-based provision should be proposed, citing Moore-Ede et al. (1986), Gold et al. (1986), McDonald (1984), Tepas and Monk (1987), Caldwell (1997), and Akerstedt, T. (1995), but they offered no specific suggestions. Restriction On Early-Morning Driving Time The ATA cites Wylie et al. (1996) as the basis for its assertion that the FMCSA should not restrict early-morning driving time. The IIHS and NSF cite Rosa and Colligan (1988), Kogi and Ohta (1975), Caldwell (1997), and Akerstedt, T. (1995) to support their belief that early- morning driving time restrictions should be proposed. Exemptions, Variations, and Customizations The AHAS opposes all exemptions from HOS regulations that could allow longer daily driving hours, shorter off-duty rest periods, or other relief that fails to incorporate the same health and safety protections afforded long-haul drivers. The AHAS cited NTSB (1995) in support of its argument. The AHAS contends that the FMCSA has no research of record on the following topics: comparisons of long-haul versus short-haul fatigue-related crashes; sleep quality; and other considerations of alertness or performance deficiencies associated with differences in acute or chronic sleep deprivation, including differences identified from a comparison of long-haul and short-haul CMV driving. The Associated General Contractors of America (AGCA) submitted AGCA (1997) showing that construction industry drivers have short trips and a short seasonal operation period. The AGCA believes this study supports its contention that its member firms are not engaged in long- haul truck driving and should therefore be exempt from any HOS regulations. The National Ready Mixed Concrete Association (NRMCA) [[Page 25552]] believes that Wylie, et al. (1996) support that industry's request for an HOS exemption. The NRMCA also submitted its own survey (NRMCA (1997)) stating that drivers in its industry drive CMVs only 44 percent of their [working] time, that they average 22 miles per round trip and they drive an average of 21,024 miles per year. The Utility Solid Waste Activities Group and Virginia Electric and Power Company argue that the findings of Bowen, V. (1996) support the need for a utility industry exemption. They submitted both the study and a petition for an exemption. Consignor and Consignee Responsibility and Accountability The OOIDA cites OOIDA (1997) as the basis for its support for heavy fines for consignors and consignees responsible for drivers violating the regulations. Loading and Unloading Freight The AHAS cites Wylie et al. (1996) to show loading and unloading freight should be classified as on-duty time. It argues that eliminating such duties from a driver's responsibilities would justify reduction of total driving and duty time from 15 hours to a more reasonable period of time, such as 12 hours. The OOIDA cites OOIDA (1997) showing that 42 percent of responding drivers record loading and unloading time as off duty. The same survey shows 74 percent would record such time as on duty-not driving if the drivers were paid a reasonable amount for such services. Consignor and Consignee Delays The OOIDA (1997) survey shows 80 percent of responding drivers are not paid for time waiting while delayed by consignors and consignees. Seventy-three percent of responding drivers record consignor and consignee delays as off-duty time, while 66 percent would record such time as on duty-not driving if they were paid a reasonable amount for such delays. Performance-Based Regulation Feasibility The ATA believes the FMCSA should propose a program for HOS performance similar to the fatigue management demonstration project undertaken in Queensland, Australia and discussed in a paper authored by Gary Mahon of Queensland Transport. The ATA noted in its comments that the province of Alberta, Canada was also considering a fatigue management demonstration project. Alberta and Queensland have ongoing pilot programs within limited geographical areas. According to a discussion paper on the Queensland program that the FMCSA has received: The Queensland Fatigue Management Program (FMP) is designed to control all the factors that cause fatigue. The FMP takes into account more than just the number of hours spent driving. FMP operators enter into an accreditation agreement with Queensland Transport. This agreement sets out the conditions of the program and allows operators to operate outside the Truck Driving Hours Regulations. This is the incentive to take part in the program. The FMP Standards attempt to assist the operator with managing all of the factors that impact fatigue including: scheduling, rostering, driver health, workplace conditions, fitness for duty, time off, and management systems to operate the FMP etc. In order to operate under the FMP, the operator must develop and implement management systems and procedures that will allow them to meet the standards and to achieve the level of performance that is required. Enforcement occurs when operators fail to reach those standards and are disaccredited (sic) from the program.\1\ --------------------------------------------------------------------------- \1\ Abstract, Australian initiatives in managing fatigue in transportation, by Assoc. Prof. Laurence Hartley, Institute of Research in Safety & Transport, Murdoch University, South Street, Western Australia, Australia, in the Truck Safety Symposium 99, Steering Committee, June 3, 1999. The IIHS does not believe fitness-for-duty and other performance measurement devices have been adequately tested and validated. They suggest review of Williamson et al. (1996), Mausner and Braun (1974), and Lilienfeld and Stolley (1994) to avoid false negatives and false positives, while also accounting for environmental variables such as traffic congestion and weather conditions, before opting for performance requirements. The NSF also does not believe fitness-for-duty and other performance measurement devices are feasible and operationally practical at the present time. The NSF suggests review of Dinges (1995), Horne and Reyner (1995), Valley and Broughton (1983), and Dinges and Kribbs (1991) before requiring ``drowsy driving detection devices'' and that such devices may lead to a false sense of security for those drivers who would rely solely on them to detect declining alertness. The NSF believes one does not have to fall asleep to be a danger to oneself and others. VI. FMCSA Response to Comments and Research Cited One of the foremost goals of this rulemaking is to reduce the number of CMV drivers and others killed and injured in crashes. In formulating new rules, the FMCSA must consider persuasive evidence and reliable data. Most CMV drivers and other commenters did not have access to research concerning technical or economic aspects of fatigue, alertness, off-duty, on-duty, or driving time. The agency appreciates all the information provided by the responders, and the serious comments reflecting both experience and interpretations of the many studies, research reports, and surveys cited. It would not be useful to take issue with the information and comments provided in the abstract. Rather, the FMCSA reviewed these issues and comments in the context of developing its proposal. The remainder of this preamble will focus on those issues in relation to the development of the proposal. The agency, however, has not ignored the concerns of drivers and other commenters whose personal experiences have led them to support changes to the HOS regulations. Fatigue Management Programs or Plans Although the FMCSA is not proposing any regulations specifically dealing with Fatigue Management Programs or Plans, the agency is not ruling them out. There is still much to learn about such alternatives to prescriptive HOS regulations, and the FMCSA has been very interested in the experience of other countries in their implementation. Within the DOT, moreover, the Federal Railroad Administration is initiating fatigue management programs for selected railroads. As suggested by the IIHS and the NSF above under the discussion of Performance-based Regulation Feasibility, more study is needed before such alternatives should be incorporated in the regulation of commercial highway operations. The FMCSA continues to study the technology it believes is required to support such programs. Projects are presently under development to test at least two forms of technology in regular operations during this fiscal year. The agency believes that careful testing within the scope of pilot demonstration programs authorized under 49 CFR Part 381 (49 U.S.C. 31136[e]) would be the best way to investigate the feasibility of this promising approach to performance-based regulation in this area. VII. Regulation Development Process A. Research Findings In developing this section, the FMCSA has relied on a large body of research dealing with work, fatigue, alertness, sleep cycles and other related matters. The FMCSA reviewed over 150 research studies and other documents, many of which were referred to by docket commenters or provided as attachments to docket comments. Many [[Page 25553]] of the reviewed documents reported on research conducted on motor carriers and CMV drivers. Others, such as studies on shiftwork, sleep and performance, and the physiological nature of sleep, were judged relevant to the issue of CMV driver safety, even though they were conducted in other operational settings or in laboratory environments. It must be noted that the conclusions drawn from this research were done from the perspective of commercial highway transportation. Although there are industries seeking exemptions from the HOS rules citing research studies in the docket concerning the performance of their drivers, e.g., utility, construction, and motorcoach industries, the FMCSA could find no research studies showing these drivers do not suffer from fatigue, do not need sleep daily, and have fewer fatigue- related crashes than other truck and bus drivers subject to FMCSA jurisdiction. For example, as stated above, the Utility Solid Waste Activities Group and Virginia Electric and Power Company provided the Bowen (1996) study stating it supports the need for an utility industry exemption. Motor carriers participating in the study were a mix of FMCSA-regulated and non-regulated public utilities. The study used a sample of drivers who accumulated nearly five million miles of driving. Ninety crashes were reported (90 divided by 5 million miles = 18 crashes per million vehicle miles). The author relates (on page 15 of the report) ``Based on this limited exposure [1731 hours of driving time reported], it seems that after 80 hours on duty [in a 7 day period], the accident rate rises precipitously.'' The FMCSA notes that, although the Bowen study tracked exposure by both driving time and distance, there were no statistics presented aside from purely descriptive graphs and tables. No significance tests were discussed. No baseline data was presented. This concerns the FMCSA given the extremely high reported crash rate. The data was not analyzed to assess time-of-day effects. Bowen did not report the highway classes where the travel took place and this factor was not used to discriminate among portions of the data. The FMCSA believes that Bowen's study does not support that the utility industry's request for an exemption from the HOS rules. It also must be noted that the conclusions drawn from all the research may not be consistently applicable throughout the transportation modes. The FHWA convened a panel of representatives of several agencies within the DOT to review the draft proposal from the perspective of the transportation modes regulated by those agencies. Although those representatives took no issue with any of the conclusions drawn from the research cited so far as truck transportation was concerned, they noted that in some cases, due to the difference in operating conditions and required adherence to various laws and treaties, the conclusions may not translate directly to other modes. Some examples are offered in the discussions that follow. Studies germane to this NPRM and relied upon by the FMCSA are discussed in An Annotated Literature Review Relating to Proposed Revisions to the Hours-of-Service Regulation for Commercial Motor Vehicle Drivers, Freund, D.M., Office of Motor Carrier Safety, November 1999, Publication No. DOT-MC-99-129. That review, as well as a copy of each research study it discusses, has been placed in the docket. The review sets the stage for the problem assessment with chapters covering research on the contribution of operator fatigue to highway vehicle crashes, in general, and the CMV driver fatigue contribution to CMV crashes, in particular. The review then focuses on research into the causes of drowsiness--lack of sufficient, quality sleep--and includes a chapter on the effects of sleep deprivation on alertness and performance, followed by one on naps as a sleep deprivation countermeasure. The review moves forward to address general considerations concerning HOS regulations, working conditions, and regulatory compliance before focusing on setting of schedules, shift rotation, and the special case of multi-day shifts. The final three chapters address outcomes of hours-of-service pilot tests and waivers, operational and performance models, and technological approaches to CMV driver alertness management. As O'Neill and his co-authors of ``Understanding Fatigue and Alert Driving,'' a training course developed by the ATA in partnership with the FHWA, point out, ``Fatigue has several causes: (from) inadequate rest, sleep loss and/or disrupted sleep; from stress; from displaced biological [circadian] rhythms, excessive physical activity such as driving or loading [cargo], or from excessive mental or cognitive work.'' (ATA, p. 8). The term ``circadian'' comes from the Latin words circa dies, or ``about a day,'' i.e., 24 hours. Circadian rhythms become displaced as a result of schedule irregularity that affects the time when people sleep. Adverse effects of sleep deprivation can occur when the opportunity to take sleep is curtailed, when people try to obtain sleep during periods of the day when their systems are in a more-active physiological state (such as during the mid-morning and early evening), or when environmental conditions are not conducive to obtaining sleep. The adverse effects include slower reaction times, poor and variable responses, deterioration of judgment, vigilance, and attention, and alertness. Loss of sleep can also produce subjective feelings of tiredness, loss of motivation, and deterioration of mood. (ATA, p. 7). On the other side of the coin, long working hours (especially when the work demands vigilance and concentration) can contribute to fatigue and cause people (by their own choice, or as directed by others) to reduce the time they take for sleep. The converse of long duty hours is a shorter period of time remaining in a 24-hour period for sleep. The maintenance of schedule regularity was an element of the original HOS regulations issued by the ICC. Until 1962, the HOS regulations limited driving and on-duty time in a 24-hour period. Although the on-duty time limit of 15 hours and the 8-consecutive-hour off-duty period were set in 1941, it is quite possible the actual off- duty period may have been slightly longer to comply with that 24-hour period. Thus, the 1962 rule change that introduced the requirement for driving and on-duty periods to be separated by an 8-hour off-duty period may have had two unintended effects: (1) It allowed drivers to be placed on a schedule that was irregular from a circadian standpoint and (2) it decreased the actual daily off-duty time provided to them. As several researchers point out, there is a dual predicament with night workers: not only are they required to perform tasks during the time of day they are least able to from physiological and cognitive standpoints, they must sleep during the time of day their bodies are least receptive to it. Recommendations From the Research The research reviewed during the FMCSA's process of developing this NPRM suggests five main areas of consideration applicable to CMV drivers: (1) The work day should be more regular; (2) drivers should be afforded more opportunity for daily and weekly sleep; (3) working hours, including hours spent driving in any duty shift, should be limited to no more than 12 hours; (4) the time of day when [[Page 25554]] driving is performed or, conversely, when sleep may be obtained, should be considered; and (5) non-compliance by drivers and motor carriers increases the potential for adverse safety outcomes. Although not using identical terms, most of the comments also suggested these areas for prime consideration. 1. The Work Day Should Be More Regular: Maintenance of Circadian Rhythm It has been well established that the hours of the day and night are not equivalent from the perspective of human alertness and safe, efficient, and productive performance of workplace tasks, Wylie, et. al. (1996); Brown, I.D. (1994); Campbell, K. (1988); Folkard, S. (1997); Hildebrandt, G., Rohmert, W., & Rutenfranz, J. (1974); Wylie, D. (1998); Akerstedt, T. (1991); Hildebrandt, G. (1976); Rutenfranz, J., Knauth, P., & Colquhoun, W. (1976); Vidacek, S. et al. (1986); Williamson, A.M. & Sanderson, J.W. (1986); O'Neill, et al. (1999); Akerstedt, T. (1997). Humans are biologically programmed to operate on a daily cycle of just over 24 hours. The cycles of daylight and darkness act as synchronizers (see Duffy, J.F. et al. (1996)). If people suddenly shift their wake-sleep cycles (e.g., when traveling across time zones), they must adjust to the new ones and become re- synchronized. This takes about one day per time zone crossed. Many work environments must be staffed on a 24-hour basis, so workers are scheduled in shifts. Shiftwork can introduce another problem. A nightshift worker, required to sleep during periods of higher physiological activity and to be awake during periods of lower activity, may have difficulty adjusting to an inverted wake-sleep schedule and can accumulate a sleep debt that can seriously affect the level of performance and safety. Even when a consistent schedule is established and wake-sleep patterns are stabilized, it is generally recognized that physiological and performance levels reach the low point of their cycles in the hours after midnight and in the early to mid-afternoon. Therefore, night workers are most susceptible to the dual predicament mentioned above. Unless the night shift worker is able to obtain sufficient restorative sleep on a regular basis, the risk of substandard and potentially unsafe performance substantially increases. 2. The Driver Should Be Afforded More Opportunity for Daily and Weekly Sleep Daily sleep. Each driver should have an opportunity for eight consecutive hours of uninterrupted sleep every day. The current rules require a minimum of eight consecutive hours off. Many motor carriers do not provide drivers more than the minimum 8 hours off duty, although the present regulations certainly allow them to do so, and many drivers accept tight schedules without objection. These drivers may have to commute home, eat one or two meals, care for family members, bathe, get physical exercise, and conduct other personal activities, all within their 8-hour off-duty period. The current rules also allow the 8-hour off-duty period to be split into two periods when rest is taken in a sleeper berth. One of the two periods may be as short as 2 hours. To afford the driver an opportunity to obtain a minimum period of 8 hours to sleep, the research shows that the off-duty periods need to be increased. Nine hours off duty was originally required in 1937. For various reasons, organized labor objected to most of the original regulations, and upon further deliberation, the ICC reduced the 9-hour off-duty period to 8 hours in each 24 hours. 6 M.C.C. 557, July 12, 1938. Several studies strongly suggest the FMCSA should require an even longer consecutive off-duty period than the 9 hours the ICC required in its original 1937 HOS regulations. To provide additional off-duty periods within the 24-hour cycle for necessary personal activities and rest, docket comments and research strongly suggest the need for total off-duty periods from 10 to 16 hours. Studies in aviation (Gander, et al. (1991)), rail (Thomas, et al. (1997), Moore-Ede et al. (1996)), and maritime environments (U.S. Coast Guard Report No. CG-D-06-97, U.S. Coast Guard (1997)(MCS 68/INF.11)) illustrate the same point. Studies of truck drivers, including Lin et al. (1993) and McCartt, et al. (1995), point specifically to increased crash risk and recollections of increased drowsiness or sleepiness after fewer than nine hours off- duty. Studies performed in laboratory settings, as well as studies assessing operational situations, explore the relationships between the sleep obtained and subsequent performance (Dinges, D.F. & Kribbs, N.B. (1991); Bonnet, M.H. & Arand, D.L. (1995); Belenky, G. et al. (1994); Dinges, D.F. et al. (1997); Pilcher, J.J., & Hufcutt, A.I. (1996); Belenky, G. et al. (1987). The results of the studies can be summarized simply: a person who is sleepy is prone to perform more poorly on tasks requiring vigilance and decisionmaking than a person who is alert. The time when sleep is taken is important as well because sleep fragmentation can be a byproduct of the timing or the quality of the sleep environment (Bonnet, M.H. (1994); Roehrs, T., Zorick, F., & Roth, T. (1994); Mitler, M.M. et al. (1997); Wylie, D. (1998)). It is virtually impossible to get an adequate amount of sleep when time for commuting, meals, personal errands, and family/social life is subtracted from an 8-hour off duty period, as the ICC found in 1937. Wylie et al. (1996), for example, showed that drivers in the study obtained nearly 2 hours less sleep per principal sleep period than their stated ``ideal'' (5.2 hours versus 7.2 hours). However, many of them did not manage their off-duty time efficiently or effectively to obtain sufficient sleep. All commuting, meals, personal hygiene, social interaction within the study setting, the study protocol itself, and sleep had to fit into their off-duty periods. The U.S. and Canadian drivers participating in that study operated under schedules set up to allow driving up to the maximum time periods permitted under U.S. or Canadian regulations. The drivers returned to regular work-reporting locations at the end of a shift. The elapsed time between beginning and ending a shift included many ancillary duties and other activities in addition to driving so that time available for sleep was generally limited to 8 hours. Participants who drove a regular 10-hour daytime schedule every day spent 5.8 hours in bed and 5.4 hours asleep. Study drivers who ran a regular 13-hour schedule starting in the daytime spent 5.5 hours in bed and 5.1 hours asleep. This was about 2 hours less than the drivers would have preferred to sleep. The time-in-bed similarities between the 13-hour and 10-hour daytime drivers was likely due primarily to their proximity to the sleep center--the 13-hour drivers had to commute less than 10 minutes from their home terminal to the sleep laboratory and 10-hour drivers had to commute between 20 to 30 minutes. (All times cited are for the principal sleep periods, and do not include the naps that some drivers took during their work shifts.) Also, the drivers in both of these daytime-driving groups were able to obtain their principal sleep during optimal times of the day, starting in late evening and ending in the early morning. Other studies have found that the amount of sleep obtained by CMV drivers is variable and often short. Arnold, P. et al. (1996), interviewed over 700 CMV drivers in the state of Western Australia, which has no formal HOS regulations. Of the drivers interviewed, about 5 percent reported having no sleep on one day during the prior week, 12.5 percent reported [[Page 25555]] obtaining less than 4 hours of sleep one or more work days in the prior week, and about 30 percent reported obtaining less than 6 hours of sleep on at least one work day. Prior to commencing their current trips, about two-thirds of drivers had between 6 and 10 hours of sleep, but about 20 percent had less than 6 hours of sleep (pp. 27-28). VanOuwerkerk, F. (1988) in a study based on interviews with 650 international European Economic Community (EEC) drivers, noted that drivers reported a median sleep time of 6.7 hours and a median rest period of 7 hours. They reported that the ``minimum rest time [reduction from 11 hours to eight hours not more than two times per week, as permitted under the current EEC Council Directive] has become the rule'' as far as both drivers and enforcement officials were concerned. In their survey of 511 medium-and long-distance truck drivers in the United States, Abrams, C., Shultz, T., & Wylie, C.D. (1997), found no statistically significant differences in the stated rest needs among the categories of drivers (owner-operator, company driver, regular route, irregular route, solo, team): on an average day, a driver reported needing an average of 7 hours of sleep. There was a slight difference between union and non-union drivers; the former reported needing about 31 minutes less sleep. Just over 90 percent of the drivers reported that they usually used a sleeper berth while on the road. Almost three-fourths of the drivers reported taking their sleep in a single period, spending eight to nine hours in the berth. Just over two-thirds of the drivers who split their sleeper berth period reported usually spending 4 to 5 hours in the berth during one period. The time of day when sleep is taken can affect how long the sleep period lasts. Grandjean, E. (1982), cited several surveys of European shiftworkers. ``It appeared that daytime sleep was distinctly shorter than night sleep the workers took on their rest day. The average length of sleep in the daytime was six hours, whereas on the rest day the average varied between six and twelve hours, with longer sleep on the second of the two rest days than on the first'' (pp. 248-249). Grandjean cites Lille's term ``sleep debt'' to describe the longer sleep periods the nightshift workers took on their two between-shift rest days and noted that a single night's sleep was apparently insufficient to ``pay it back.'' Other modal administrations within the U.S. DOT have also found similar findings with respect to their regulated industries. An aviation study documented how commercial flight crews organize their sleep during layovers on long-haul (trans-oceanic) trips involving the crossing of multiple time zones. Duty periods on the trips that were studied in Gander et al. (1991) averaged 10.3 hours. The duty periods alternated with layovers averaging 24.8 hours, and the crew members typically took their sleep in two periods. The authors noted that the sleep/wake patterns were complex, with an average pattern of sleep and wakefulness of 19 hours awake/5.7 hours asleep/7.4 hours awake/5.8 hours asleep. The flight crews also reported naps on the flightdeck and during their off-duty periods. The authors stated: ``This study clearly documents that, in scheduled commercial long-haul operations, there are physiologically and environmentally determined preferred sleep times within a layover. The actual time available for sleep is thus less than the scheduled rest period.'' (p. 1) A railroad study compared two schedules, one involving fast- backward rotations, the other, slow-backward rotations. The Thomas et al. (1997) study participants received a crew call two hours prior to going on-duty, as railroad crews do generally in their normal operational workdays. Under the fast-rotating schedule, the engineers' sleep duration averaged 4.6 hours during a 9.3-consecutive-hour off- duty period. Under the slow-rotating schedule, the average sleep duration was 6.1 hours during a 12-consecutive-hour off-duty period. (The minimum off-duty time required was 8 consecutive hours if the engineer had less than 12 hours on duty.) Locomotive engineers participating in a survey (Moore-Ede et al. (1996)) were asked several questions concerning sleep needs and sleep obtained. The following information is drawn from Questions 67 through 70 of the Stage II Volunteer Survey, Overall Totals. To ``feel alert and well rested,'' 10.8 percent stated they needed 5 or fewer hours; 35.1 percent, 6 hours of sleep; 29.7 percent, 7 hours; and 24.3 percent, 8 or more hours. On average, the engineers said their actual sleep taken on days they worked averaged: fewer than 5 hours, 21.6 percent; 5 hours, 16.2 percent; 6 hours, 21.6 percent; 7 hours, 29.7 percent; and 8 or more hours, 10.8 percent. On days off, the engineers stated they received: 7 hours, 37.8 percent; 8 or more hours, 43.2 percent; all other amounts 6 hours or less, 18.9 percent. During vacations, sleep times averaged: 35.1 percent, 7 hours; 51.4 percent, 8 or more hours; all other amounts 6 hours of less, 13.5 percent. A study was conducted to quantify the nature and extent of fatigue in mariners in the U.S. Coast Guard's ``Fatigue and Alertness in Merchant Marine Personnel: A Field Study of Work and Sleep Patterns,'' Report No. CG-D-06-97. Data in the form of work and sleep logs were collected for periods from 10 to 30 days from 141 mariners on eight different ships involved in U.S. Pacific intercoastal trade. Among the findings: average sleep duration at home was 7.9 hours while average sleep duration at sea was 6.8 hours; watchstanders slept less and rated their sleep to be of lower quality compared to command personnel or day-workers; and the work days of watchstanders are longer than for command and day-work personnel. The researchers also analyzed what they termed ``critical fatigue indicators'': the proportion of 24-hour periods during which total sleep was less than four hours, the proportion of alertness self-assessments of 3 or fewer recorded in the sleep log, and the proportion of sleep latencies (time between going to bed and falling asleep) of five minutes or less as recorded in the sleep log. Watchstanders fared worse than command or daywork personnel, and watchstanders on the 4-8 watch had a considerably higher incidence of sleep durations under 4 hours in a 24-hour period. The report concluded: ``The data clearly portray that risk factors for fatigue are present in maritime work schedules. The solution to this problem involves providing the opportunity for a longer continuous rest period, and motivating mariners to take advantage of that rest period for a single, uninterrupted sleep. A review of shipboard operational practices may identify various means to provide longer continuous rest periods and other approaches to fatigue reduction.'' (p. 6) Weekly sleep. For weekly off-duty periods, the research indicates that to negate the effect of accumulated week-long sleep deprivation and restore alertness to the human body it is necessary to have at least two consecutive nights off-duty that include the periods from midnight to 6:00 a.m. For long-haul CMV drivers, this ``weekend'' (i.e., a period to permit recovery from cumulative fatigue, not necessarily falling on a Saturday and Sunday) should be up to 56 hours long, but could be reduced to 32 hours as long as that period included two nights covering two periods from midnight to 6:00 a.m. The research suggests that drivers may need even more nights off duty if they have a severe sleep deficit. This may be a good example where the science would not change, but [[Page 25556]] where the same provisions cannot be provided in another transportation mode without creating another safety problem. It may not be possible, for instance, to have an ocean ship in a port for each ``weekend'' recovery period, and everyone could not be provided two consecutive nights off each week while the ship is away from port. CMV traffic is much more flexible. In his 1987 study of European heavy-goods-vehicle drivers, Hamelin suggests that the risk of an accident might be more dependent on cumulative fatigue than on the short periods of work time prior to a crash noted on the accident reports (the forms required an entry of time since the last stop, rather than the time since the beginning of a trip). Hamelin computed accident risk rates according to the time of day, the time on task, and whether the drivers were in the ``Transport'' or ``Other'' occupational categories. For all drivers, the risk rate was nearly double (1.82 times baseline) after 11 hours of work. Hamelin found for the category labeled ``Other'' drivers, who had worked over 11 hours, it was over four times as high during the period 8:00 p.m. and 7:00 a.m. In a study of driving patterns used by a U.S. less-than-truckload motor carrier, Jovanis, P. et al. (1991) also noted increased crash risk associated with long driving times over two or more days of a week. The Society of Automotive Engineers, Inc.'s ``Truck and Bus Industry Glossary,'' February 1988, No. SP-88/732, defines ``less-than- truckload (LTL)'' as ``a quantity of freight less than that required for the application of a truckload [TL] rate.'' Jovanis and his co- authors noted, ``The two patterns with the highest risk of an accident were those that contained heavy driving during the prior three days and consisted of driving from 3:00 p.m. to 3:00 a.m. (Pattern 1) and from 10:00 p.m. to 10:00 a.m. (Pattern 8). The lowest risk was associated with driving from 8:00 p.m. to 6:00 a.m. but with limited driving on the prior three days.'' (p. 27) Wylie et al. (1996) found that although some of the performance data did not show a clear-cut relationship to driving time (time on task), drivers' self-ratings did correlate significantly with time since the start of the trip and with the cumulative number of trips. ``Thus, the self-ratings were not very good indicators of drowsiness, but they may have been indicative of increasing stress or compensatory effort that signaled fatigue or loss of alertness.'' (p. 5-11). Smiley, A. & Heslegrave, R. (1997) cited several scientific studies dealing with recovery time as a portion of their review of scientific literature on rest and recovery requirements. The context of the review was an assessment of the potential adequacy of a 36-hour cumulative- fatigue-recovery provision that had been proposed by motor carrier industry groups to Transport Canada. Smiley and Heslegrave cited a 1967 study by Lille suggesting that a single day off was insufficient for night workers to recover after a sleep debt accumulated over five days. Other studies they cited indicated a preference, in terms of recovery, for a three-day rest period compared to a two-day period after three 12-hour night shifts. One example was a study (Hildebrandt et al. (1974)) that illustrated this advantage of two days and three days off, compared to one day off, in operator performance (locomotive engineers' missed multiple in-cab warning signals that resulted in automatic braking being triggered) and a 1994 literature review indicating that two nights of recovery sleep are usually sufficient to allow near full recovery after extended periods of sleep loss. Smiley and Heslegrave concluded that, ``nevertheless, although the available research is sparse, it is sufficient to raise concerns about a 36 hour reset that would allow drivers to accumulate up to 92 hours on-duty within a seven-day period, particularly for night driving.'' (p. 14) O'Neill, T. et al. (1999) studied drivers on long (14-hour) daytime duty schedules in a driving simulator. The drivers did not appear to have accumulated significant sleep loss during the study but their amount of measured sleep increased and their sleep latency decreased on their first off-duty days. The research conditions tested found ``the effectiveness of a full two nights and one day off (that is, `Friday night' to `Sunday morning' as a minimum safe restart period--about 32 hours off duty) under the conditions tested.'' (p. 48) Griffin, G., et al. (1992) assessed potential outcomes of the FHWA's August 19, 1992 (57 FR 37504) NPRM, concerning HOS for CMV drivers, termed the ``24-hour restart'' provision. As discussed above in today's NPRM, the FHWA withdrew the proposal on February 3, 1993. The authors noted, ``The implications for safety are the most difficult to determine . . . a search for secondary accident data that would be useful in addressing the implications that the 24-hour rule would have on safety was made at the state and federal level. No data was identified that would be statistically valid . . .'' (p. 37) Rosekind, M.R. (1997) in a major aviation conference presentation, advised listeners to consider cumulative fatigue effects. ``It is important to maintain an optimal sleep opportunity every 24 hours and also address the potential for cumulative effects. Therefore, appropriate recovery time should be allowed per week (days or rolling hours). Scientific studies show that two nights of recovery sleep are typically needed to resume baseline levels of sleep structure and waking performance and alertness.'' (p. 7.6). Second, the driver should be afforded more opportunity for daily and weekly sleep. 3. Driving in Any Duty Shift Should Generally Not Exceed 12 Hours The research suggests that performance degrades and crash risk increases markedly after the 12th hour of any duty time during a work shift (Hamelin (1987); Brown (1994); Campbell (1988); Rosa and Bonnet (1993); Rosa (1991); Rosa et al. (1989); Harris and Mackie (1972); Mackie and Miller (1978); U.S. Army (1983); Transportation Research and Marketing (1985)). The studies by Campbell, Brown, and Transportation Research and Marketing focused on CMV crash cases; the other studies were performed in actual workplaces (industrial plants, in the case of Rosa's studies) and over-the-road CMV operations (U.S. Army and Mackie and Miller). Some recent research has suggested that naps can improve performance later in work cycles. The use of naps was not a factor explored in these earlier studies. This research finding, however, might not apply very well to other transportation modes, where duty needs may encompass more than is contemplated in the duty of driving CMVs. Such other duty needs and considerations might include: the restricted environment of airplanes, locomotives, and ships; the length of trips away from airports, harbors, and depots; and on-board airplane, locomotive, and vessel needs and emergencies. 4. The Time of Day When Driving Is Performed Should Be Considered Research suggests that there is a higher risk of fatigue-induced single-vehicle accidents at night, especially between midnight and 6:00 a.m. For example, the following eight studies are representative of the research. Hamelin (1987) computed accident risk rates among drivers of heavy-goods vehicles according to the time of day, the time on task, and whether the drivers were in the ``Transport'' or ``Other'' occupational categories. For ``Transport'' and ``Other'' categories combined, the accident risk rate was [[Page 25557]] nearly double (1.82 times the baseline) after 11 hours of work. However, for the category ``Other Branch'' drivers who had worked over 11 hours, it was over four times as high during the period 8:00 p.m. and 7:00 a.m. Blower & Campbell (1998) reported that about 20 percent of all fatal crashes and fatalities and 10 percent of all injuries involving a long-haul truck (tractor pulling at least one trailer) occur between midnight and 6:00 a.m. Crashes at night tend to be more severe, with about 435 injuries per thousand crashes between midnight and 6:00 a.m., compared with 320 injuries per thousand for the remainder of the day. There are about three times as many fatalities per thousand crashes from midnight to 6:00 a.m. Using exposure data classifying night as 9:00 p.m. to 6:00 a.m., truck travel during that period is associated with a relative risk about twice that of the rest of the day. Kecklund and A kerstedt (1995) examined data for all accidents involving an injury and all truck accidents on Swedish motorways for the period 1987-1991. Risk ratios were computed against a baseline time period of 8:00 a.m. to 4:00 p.m. Single-vehicle accidents where alcohol was not a factor peaked at 4:00 a.m. at 13 times the baseline level. The risk of a fatal accident was 35 times the baseline level at 4:00 a.m.; severe (27 times) and minor (19 times) injury accidents also peaked at that time. Risk ratios for overtaking-vehicle and oncoming- vehicle accidents also peaked at 4:00 a.m., but the ratios were considerably lower (3-4 times). For trucks, single-vehicle accidents between 3:00 a.m. and 5:00 a.m. peaked at 3.8 times the baseline risk. Hildebrandt, G., et al. (1974) examined one month's records of approximately 15,000 locomotive engineers employed by the German Federal Railways. They reviewed records covering 2,238 automatic braking incidents and nearly 20,000 second-level warning signals. The authors studied the relative frequency of second-level warning signals and of the occurrence of automatic braking by time of day and by length of shift and found peaks of about 125 percent of the daily average automatic braking incidents taking place around 3:00 a.m. and 1:00 p.m.-2:00 p.m. They found peaks in the activation of the acoustic signal at around 3:00 a.m. and 3:00 p.m. The authors concluded there exists a 12-hour period of variation in vigilance, superimposed upon the 24-hour circadian period. Jovanis, P., et al. (1991) studied several driving schedule patterns in a less-than-truckload motor carrier's ``pony express'' operation. In this type of operation, drivers make an outbound trip from the home terminal to another terminal, drop their trailer, pick up another trailer, and return to the home terminal at the end of a work shift. Cumulative driving hours ranged from 7.8 to 8.4 daily, while the drivers' daily cycle of on duty and off duty ranged from 22.3 to 23 hours, indicating some circadian disruption. The highest daily risk was found to occur from 4:00 a.m. to 6:00 a.m. Elevated risk was found from midnight to 8:00 a.m. According to the authors, ``The two patterns with the highest risk of an accident were those that contained heavy driving during the prior three days and consisted of driving from 3:00 p.m. to 3:00 a.m. (Pattern 1) and from 10:00 p.m. to 10:00 a.m. (Pattern 8). The lowest risk was associated with driving from 8:00 p.m. to 6:00 a.m. but with limited driving on the prior three days.'' (p. 27) Dinges, D.F. et al. (1997) studied young male adults under conditions of partial sleep deprivation (average of 5 hours of sleep obtained per night). They reported that the only statistically reliable subjective measure of time-of-day effects (out of several measurements used) was the Stanford Sleepiness Scale. However, four of the five objective performance measures showed statistically significant variations by time of day (tests were administered at 10:00 a.m., 4:00 p.m., and 10:00 p.m.). One measure of vigilance and reaction time (psychomotor vigilance task) also showed significant variation across the days of sleep restriction. Dinges and his coauthors reported the effects of sleep restriction on performance measures appeared to level off between the second and fifth day of sleep restriction; for subjective assessments, between the second and sixth day. The authors believed the subjects were undergoing an ``adaptation'' to sleep restriction following an initial shift to higher daytime sleepiness levels, and that this change may have resulted from changes to sleep itself during those periods (although EEG was not monitored). However, by the sixth to seventh day performance was deteriorating and subjective sleepiness was increasing again. The authors noted the linear nature of the trends indicated the performance changes are cumulative, and they did not level off. Wylie, et al. (1996), in their operational study of CMV drivers on fixed routes under one of four types of schedules, noted that the strongest and most consistent factor influencing driver fatigue and alertness was the time of day. Compared to driving at other times, night driving (midnight to 6:00 a.m.) was associated with more video observations of driver drowsiness, poorer lane-tracking, and worse results on tests of mental performance. Lin, et al. (1993), studying schedules of less-than-truckload CMV drivers, assessed consecutive driving time, multi-day driving patterns over a seven-day period that included time of day of driving and days driving within that period, driver age, driving experience, and hours off-duty prior to the trip. Out of ten driving patterns, they found the driving pattern with the lowest risk was ``Pattern 2,'' a highly regular schedule with on-duty times generally spanning the period 6:00 a.m. to about 2:00 p.m. and off-duty times generally spanning the period 6:00 p.m. to 4:00 a.m. Risk in six other schedule patterns that included night and very early morning driving, morning and evening rush-hour driving, and very infrequent scheduled driving had computed crash risk about 1.5 times as high (p. 5). When off-duty hours were assessed, the risk following off-duty periods less than or equal to 9 hours was 1.4 times higher. When driving time was the category of interest, Lin et al. (1993) found there were no statistically significant differences among the first four hours but the ratio increased from that time until the last driving hour. Lin and his collaborators noted a limitation in their analysis, and provide a caveat to the estimates of the odds ratios in the last driving hour category: a large number of non-crash trips are completed during the 8th or 9th hour of driving, but the authors' ``assumed failure time,'' defined as the expected time of involvement in an accident, would occur after this trip completion time. (p. 5) A limit, however, on the driving of CMVs between midnight and 6:00 a.m. would not necessarily result in a reduction in CMV-involved crashes. The benefit-cost analysis prepared for this NPRM and filed in the docket, as well as the section with the subheading ``Benefits and Costs'' later in this NPRM, discuss the issues, and additional research findings related to the benefit-cost, of a night-restriction option that the agency analyzed in more detail. The proposed HOS regulations will address in three ways this issue of higher crash risks associated with night driving. First, the amount of consecutive hours off-duty proposed is longer than the current regulatory minimum. The objective is to allow drivers to have the opportunity to sleep longer during their off-duty periods. Second, the regulatory proposal would return to the pre-1962 24-hour clock. The circadian disruption permitted under the current regulations [[Page 25558]] (if drivers or motor carriers scheduled duty cycles to touch the edges of the current regulatory envelope of 10 hours of driving and 8 hours off-duty), which also adversely affects sleep quality and quantity, would be mitigated. Third, the regulations would require drivers to be given a consecutive off-duty period spanning at least two periods between midnight and 6:00 a.m. in every seven consecutive days to obtain restorative sleep. This has two benefits: it provides the opportunity for sleep during circadian-optimal times and it prevents drivers from being required to work more than five consecutive night shifts. 5. Non-Compliance by Drivers and Motor Carriers Increases the Potential for Adverse Safety Outcomes While drivers who drive to the maximum number of hours allowed and rest to the minimum number of hours required by the HOS rules may be fatigued, the situation of drivers who are not in compliance is undeniably worse. Whatever the limitations of, for example, 5 to 6 hours of interrupted sleep, it is clearly more restorative than little or no sleep as reviewed in the discussion of the studies for daily sleep under VII.A.2. above. Unfortunately, many drivers violate the HOS regulations. Braver et al. (1992) interviewed over 1,200 drivers at truck stops, truck inspections stations, and agricultural inspection stations in the early 1990s. Based on the drivers' responses, the authors classified 73 percent of the drivers as hours-of-service violators. A 1995 survey of over-the-road drivers in New York State, McCartt et al. (1997), found that over one-third reported driving more than 60 hours in a typical week, and a similar percent reported working 70 hours in a typical week. A more recent survey, Belzer et al. (1999), performed by the University of Michigan's Trucking Industry Program (UMTIP) corroborated this high violation rate. Belzer and his collaborators surveyed over 800 mostly over-the-road drivers at a number of truck stops in the Midwest in 1997. Only 16 percent of drivers surveyed reported that log books were generally accurate, and 56 percent stated that they had worked more hours than recorded in their records of duty status (RODS) in the last 30 days. Over-the-road drivers and owner-operators were most likely to report working more hours than recorded in the RODS, whereas local drivers were least likely. The average number of hours worked was over 63 in the previous 7 days. Twenty-five percent of drivers reported working at least 75 hours in the last 7 days, and 10 percent reported working more than 90 hours. The UMTIP also reported that the average annual mileage was found to be 112,765, where local drivers at the 10th percentile reported driving 25,000 miles annually and long-haul drivers at the 90th percentile reported driving 170,000 miles annually. Assuming drivers have two weeks of annual vacation, that leaves 351 days per year potentially available for driving. Since the current rules prohibit drivers from driving after being on duty more than 70 hours in eight consecutive days, drivers have 3,071.25 on-duty hours during which they could possibly drive [351 days divided by 8 days = 43.875 periods per year x 70 on-duty hours for driving per period]. If one assumes that a driver was able to average 50 mph for every on- duty hour all year long, he or she could drive a maximum of 153,562 miles per year [3071.25 hours x 50 mph]. A 50 mph average for an entire year is highly unrealistic, yet the Belzer survey showed that the 90th percentile of long-haul drivers surveyed covered 170,000 miles per year. The only reasonable conclusions are that these drivers grossly violated the hours-of-service limits, significantly overestimated their annual mileage, or spent part or all of the year in team operations and counted the distances they traveled during on-duty not driving and sleeper-berth periods. Drivers who comply with the HOS regulations may not be adequately rested, and the significant percentage who do not comply are probably not rested. B. FMCSA Principles for Regulatory Improvement The FMCSA determined that, based on the motor carrier and highway research and operational characteristics of the industry, it had to design regulations that incorporated the following requirements. Increase the 18-hour on-duty/off-duty work cycle to a normal 24-hour work cycle. Increase time off to allow sufficient time for 7 to 8 hours of sleep. Require mandatory ``weekend'' recovery periods of at least two nights of recovery sleep to resume baseline levels of sleep structure and waking performance and alertness. Address the effects of operations between midnight and 6:00 a.m. by requiring off-duty periods that enable restorative sleep by including two consecutive periods between these hours. Allow ``weekends'' of sufficient length to ensure safety and provide adequate protection for driver health and safety. Increase operational flexibility by offering a menu of HOS options customized to different major or distinct operational segments while still maintaining an appropriate level of safety. The FMCSA believes these requirements will significantly reduce fatigue problems related to sleep deprivation, if drivers and motor carriers adhere to them. The FMCSA recognizes, however, that these rule changes do not eliminate the potential problem the ICC described in 1937, namely: We have no control over the manner in which a driver may spend his time off duty, although some of his spare time activities may tire him as much as any work would do. We can only emphasize, by this comment, the responsibility which is the driver's own to assure himself of adequate rest and sleep, in the time available for this purpose, to insure safety of his driving, and likewise the employer's responsibility to see that his drivers report for work in fit condition. 3 M.C.C. 665, at 689. Drivers must still manage their off-duty time if these, or indeed any, proposals are to be effective. Some drivers may continue to push themselves to drive more hours than this proposal allows in order to earn more money. Others may perform non-driving jobs during their off- duty time; have long commutes to and from home; or engage in other pursuits that interfere with their obligation to obtain the proper sleep and be prepared to operate safely. Under this proposal, all time spent in any work must be counted as on-duty time, since all work can either induce fatigue or deprive the driver of sleep. C. Types of Motor Carrier Operations The motor carrier industry is diverse. A motor carrier is any person who uses a commercial motor vehicle, regardless of ownership, to transport passengers, property, or the vehicle itself for any purpose. Motor freight and passenger transportation differ in many respects from other modes of transportation. Other modes may have different means of tracking hours of service of their critical employees because of the operating characteristics and more structured environment. Many motor carriers are statutorily exempt from FMCSA jurisdiction (e.g., most of those in intrastate commerce) or exempted by regulation (e.g., Federal, State, or local government vehicles (Sec. 390.3(f)). The FMCSA grouped motor carriers subject to its jurisdiction into the following 5 types whose substantial [[Page 25559]] operational differences might warrant more individualized consideration than the current HOS rules allow: Type 1--Long haul. These drivers are away from their normal work reporting location and home for more than three days at a time; in total, they are away from home for a large part of the year. Their primary task is driving, although they may well engage in other activities, particularly loading and unloading cargo (or helping passengers and moving baggage, if the driver operates a motor coach). Type 1 drivers may have regular or irregular wake-sleep cycles, depending upon the requirements placed upon them by their employers, their clients (if they are independent owner-operators), their personal preferences, or a combination of all three. Many of these drivers use sleeper berths, but some sleep in hotels, motels, company sleeping quarters, truck stops, or other accommodations. Type 1 drivers have the highest accident exposure (based on distance traveled) of all driver categories, usually over 161,000 kilometers (km) (100,000 miles) per year (the benefit cost analysis uses 114,000 miles); team drivers may have twice that amount. They also have the least regular wake and sleep cycles, which often includes daytime off-duty periods when they must obtain sleep. These drivers are the least likely to be subject to frequent direct monitoring by their employing motor carriers, that ultimately are responsible for managing the driver's work/rest schedules. Type 2--Regional. These operations are similar to Type 1, except that drivers are away from their home base only three or fewer days at a time. For example, a Type 2 driver might report for work at 7:00 a.m. Monday after a weekend off duty, leave on a trip requiring two overnights on the road, and return to his normal work reporting location by 9:00 p.m. Wednesday. Drivers for some large less-than- truckload carriers return to their normal work reporting location after only one night on the road. Type 2 drivers are generally able to take a larger proportion of their sleep periods in a familiar home environment. A Type 2 driver has a moderately high annual mileage-based accident exposure (from 120,000 to 161,000 km, or 75,000 to 100,000 miles) (the benefit cost analysis uses an interpolated estimate of 82,000 miles since the research the agency used did not directly address this group of drivers). These drivers, although often remote from a home base of operation, are more likely than Type 1 drivers to operate in more regularized schedules and to be subject to more frequent monitoring by their motor carriers. Type 3--Local split shift. Split-shift drivers spend most of their on-duty time driving, but most are local (or home-based), and their driving shifts are generally separated by several hours. A Type 3 driver might work in a commuter or tour motor coach operation, requiring on-duty periods, for example, from 5:00 a.m. to 9:00 a.m. and 4:00 p.m. to 8:00 p.m. To be considered in this category, the driver must be off-duty during the intervening hours (in the example, from 9:00 a.m. to 4:00 p.m.). The driver may deadhead to another location, be shuttled back to a base of operations, or merely be free to spend the time as he/she chooses at the location where his/her duty terminates. If the driver performs non-driving duties for the motor carrier, holds another job, or performs any other work, time spent in those activities must be counted as on-duty time, and may remove the driver from the Type 3 category. Type 3 drivers are different from Type 4 and 5 drivers because driving is the main part of their job, and because these drivers are not on duty more than 12 hours, although the end of their shift occurs more than 12 consecutive hours after they begin the workday. Type 3 drivers are fairly prevalent in the motor coach industry, but other operations would fall into the same category, e.g., commuter transportation, before-school/after-school-activity bus drivers, split-shift