activated signals plus gates, non-train-activated signals, etc.? If this type of data breakdown is not available, what type of data breakdown do you have in this area? Answer. As mentioned in the answer to question 2(b) the relative significance of the many causes is not yet well understood. It is not possible for us to provide such a breakdown regarding causes.

As to accident types, the data can be subdivided and accumulated in many ways: time of day, day of week, weather, who struck whom, available warning devices, type of train, type of vehicle, train and vehicle speeds, locales, observed motorist actions, directions of travel, number of tracks, highway lanes, etc. Please refer to FRA's Rail-Highway Crossing Accident/Incident and Inventory Bulletin, No. 1 for Calendar Year 1978, a copy of which is attached. The following is extracted from Table 23:

1978 RAIL-HIGHWAY ACCIDENTS/INCIDENTS AT PUBLIC CROSSINGS INVOLVING MOTOR VEHICLES

Question. Would you please provide us how much money and manpower was allocated to railroad grade crossing accidents by the Federal Railroad Administration and the Federal Highway Administration in the last four fiscal years? Answer.

FRA'S MONEY AND MANPOWER TO RAIL-HIGHWAY CROSSING PROBLEMS

The Federal Highway Administration rail-highway crossing safety programs are administered through the state highway agencies in the same manner as all Federal-aid highway projects. Funds are annually apportioned to the states who select and construct projects. The following table shows by fiscal year the amount of Federal-air highway funds, all fund classes, authorized for crossing safety improve

ments:

During the period October 1, 1978, through November 30, 1979, FHWA authorized $269.4 million Federal-aid highway funds for the construction of grade separations. This expenditure for a 14-month period indicates an increased rate of activity in the

grade separation programs by the states. During the last four fiscal years, 631 new grade separations have been completed, 645 grade separations reconstucted, and active warning devices installed and completed at 3,143 at-grade crossings for a total cost of over $836 million in Federal funds. Preliminary engineering work at many more crossings was actively underway during this period. During the same 4-year period FHWA expended $1,183,000 in related rail-highway crossing research, plus $185,000 for developing and conducting training courses related to crossing safety. It is estimated that approximately 60 man-years of FHWA time is annually devoted to administering the Federal-aid rail-highway crossing program. The time and resources expended by the states, local authorities, and railroads in the development and construction of projects is substantial.

Question. How much is requested for this area this year separately by the Federal Highway Administration and the Federal Railroad Administration?

Answer. FRA's fiscal year 1981 request includes $673,000 and six person-years which is planned for rail-highway program efforts.

Regarding the FHWA, the Surface Transportation Assistance Act of 1978 authorized from the Highway Trust Fund additional sums to continue the rail-highway crossing categorical safety program and regular Federal-aid highway program, which are used to fund improvements at crossings, through fiscal year 1982. In addition, trust funds are also available to continue FHWA's current crossing safety program through fiscal year 1982.

Question. The General Accounting Office, in their report, "Railroad Crossing Safety-At What Price?", argued that the formula for funding states in the rail crossing area should take "need" into account and should not be allocated on a merely population or geographical basis. How exactly does the formula operate at the present time in this area?

Answer. In accordance with the Highay Safety Act of 1978, Sec. 203, and the United States Code, Title 23; Highways, Sec., 104, the following allocation proportions are used:

FOR SEC. 203 FUNDS (RAIL-HIGHWAY CROSSINGS)

25 percent.-One-third in the ratio which the area of each state bears to the total area of all the states.

One-third in the ratio which the population of rural areas of each state bears to the total population of rural areas of all states.

One-third in the ratio which the mileage of rural delivery and intercity mail routes where service is performed by motor vehicles in each state bears to the total of such mileage in all states.

25 percent. In the ratio which the population in urban areas, or parts thereof, in each state bears to the total population in such urban areas in all states.

50 percent. In the ratio that total rail-highway crossings in each state bears to the total of such crossings in all states.

The 50 percent is a consideration added in the 1978 Act as a result of the referenced GAO report. (See pertinent Committee comments: Senate Committee on Environment and Public Works-Senate Report No. 833, May 15, 1978, pages 10-11.) This action, along with the opening of the program to all public crossings versus just those on Federal-Aid Systems, has gone a long way towards placing the monies where problems are greatest.

Question. Do you believe that factors of need could be built more effectively into the formula?

Answer. The changes in the apportionments formula from the Highway Safety Act of 1978 and the expansion of the program to include all public crossings have substantially improved the States' ability to more effectively address the rail grade crossing problem. At this time we do not see the need for further modifications of the apportionment formula.

Question. If so, how do you believe this need factors should be dealt with the formula?

Answer. Although we do not see the need for changes to the formula at this time, there may be a need to reassess the formula as the program progresses. The FHWA currently has two efforts underway which may result in the promulgation of warrants for rail-highway crossings and/or of new regulations applicable to program management. Either or both of these may be useful in defining further program needs.

Question. The General Accounting Office stated that the Department of Transportation had under development computer models to provide States an analytical model as to how to prioritize the various needs for upgrading grade crossing problems. What is the status of this analysis?

Answer. The Department has several active research projects which involve development of analytical models which may prove useful in helping States develop priorities in their railroad-highway crossing safety programs. Reports are being prepared on these research efforts and should be available later this year.

It should be noted that none of the analytical models developed include all factors a State would consider in establishing priorities. The models therefore can only serve as an additional tool or aid a State may use in setting priorities.

Question. If it is completed, could it be provided to the Committee?

Answer. Copies of the research reports, as they become available for distribution, will be provided to the Committee.

Question. Is there any requirement that the States utilize this model in making decisions as to where to focus their expenditures in the rail crossing area?

Answer. No. As previously noted, the model does not include all factors a State might use in setting priorities and mandatory use of the models will not be required.

Question. What requirements, if any, are there to focus State funding on upgrading the most dangerous grade crossings?

Answer. The Department's current regulations (23 CFR 924) implementing Section 203 of the Highway Safety Act of 1973, as amended, require the States to initiate a process of establishing priorities for implementing railroad-highway crossing projects. This priority process focuses Section 203 funds on the most dangerous crossings.

Question. What requirements, if any, are there to require the States to use specific types of safety equipment at specified grade crossing?

Answer. Part VIII of the Manual on Uniform Traffic Control Devices (MUTCD) prescribes the national standards for traffic control systems at crossings. The MUTCD requires, as a minimum, the installation of reflectorized crossbucks, advance warning signs, and pavement markings on paved surfaces. Criteria for the installation of other types of traffic contol systems are not set forth in the MUTCD. However, FHWA regulations (23 CFR 646) require the use of automatic gates with flashing lights under certain conditions when a Federal-aid highway project is involved.

In general, each State highway agency has the responsibility to investigate, analyze, and propose measures to reduce hazards at crossings. Such safety measures may include a variety of improvements in addition to or in lieu of active warning devices.

Question. If there are no such requirements, do you believe that regulations should be changed to mandate such requirements on the States?

Answer. The selection of projects and nature of improvements by the States have been effective in addressing the safety issues at grade crossings. It is not clear that significant added mandatory requirements would substantially improve effective

ness.

As noted above, certain mandatory requirements already exist. The FHWA is presently rewriting its regulations (23 CFR 646) dealing with railroad-highway projects and existing requirements dealing with traffic control systems for railroadhighway crossings on Federal-aid projects may be further strengthened or clarified. The Department is also considering the establishment of uniform nationwide criteria for the selection of various types of traffic control systems to be installed at all public railroad-highway crossings. An advance notice of proposed rulemaking was published by the FHWA on August 10, 1978. Work is continuing on this effort although recommended criteria have not yet been fully developed.

Question. The National Transportation Safety Board has suggested that enhanced train conspicuity at night (retroflectors) is one means by which to assist in reducing accident, fatalities, and injuries. What is your view in regard to the utility of regulations to require retroflectors or other devices to make trains more visible in enhancing railroad grade crossing safety?

Answer. The FRA feels that a regulation or a regulatory action at this time would be premature. Over the years this suggestion has been put forth many times. Decision makers in the past have concluded that the potential utility of reflectors would be outweighed by the costs to place and maintain them. Some recent analyses, as well as developments in the efficiency of reflectors, has caused the FRA to consider the question again. No decision has yet been reached.

Question. It is our understanding that Canada requires retroflectors on its car fleet, and in fact, funds some of the cost. Is this correct?

Answer. The Rail Transport Committee (RTC) of the Canadian Transport Commission for twenty years has budgeted an annual total of $50,000 (cdn) for the installation of retroflectors. In recent years, only about half of this amount has been spent.

In all, about 150,000 cars have been equipped for a total estimated cost of $634,000. To date, no assessment of the program has been made. The feeling seems to be that the reflectors make only a marginal contribution to safety.

Question. Has any analysis been carried out as to whether such use of retroflectors might appreciably improve railroad grade crossing safety?

Answer. The analyses which have been done, did not withstand rigorous examination. The FRA is currently in the process of negotiating a study effort, which should be completed this year, to review work which has been done, to identify gaps and deficiencies, and to gather or outline procedures for obtaining information which is lacking. At that point, the FRA will decide on one of three courses: proceed to rulemaking, drop all further consideration, or establish a program to gather the data needed to make an informed, supportable decision. The Committee will be kept informed.

Question. The Office of Technology Assessment study on railroad safety stated, "One other possible problem with the pace of the program could be that the railroads are reluctant to install automatic systems because of the potential liability where the systems may not be fail-safe." Do you believe that there is any legitimacy to his hypothesis?

Answer. The pace of the program is no longer the problem it was when the OTA made their assessment. As of the end of fiscal year 79 more than 70 percent of all funds made available in the 1973, 1976 and 1978 Acts had been obligated. A much higher percentage of funds has been committed.

REGARDING LIABILITY AND FAIL-SAFE SIGNALS

Train activated rail-highway crossing warning devices are designed "fail-safe." This is, like all railroad signals, should any failure occur, the signal defaults to its most restrictive mode; that is, it goes on. For example, if a track circuit which detects the presence of a train is "shorted out" or broken, the signal activates. If the power goes off, the signal stands ready to operate from a backup power supply which will operate for from one to three days, depending on state law and the number of activations which occur. Even then, in the case of a gate when all power has failed, the gate will come down to its most restrictive mode for rail-highway crossing gates are held-up rather than driven down; part of the fail-safe design. Problems have occurred, though rarely, when power has been interrupted for extended periods, as a result of damage by vandals, or as a result of poor maintenance; but overall, the fail-safe designs currently in use have stood the test. The Department is not aware of any appreciable problems.

Liability is a serious economic problem. We have some 1975 data showing that these costs amounted to approximately $40 million per year. Our estimate now is that this probably exceeds $60 million. Even though the devices are "fail-safe" and liability does remain a problem, the pace of the program is picking up. If liability was impacting the program initially, we do not feel that its impact has any significance at this time. We do not feel the hypothesis has any current legitimacy. Question. If so, what, if anything, do you believe should be done about this problem?

Answer. Before leaving the issue of liability, we would like to stress that the Department has been concerned for some time about the impact, legitimacy and equity of rail-highway crossing liability costs on rail operations and economics, and even on program participation. The Transportation Systems Center, under joint FRA-FHWA sponsorship, has been looking into alternative means of managing, or otherwise bringing this problem under some kind of control or at least assisting the rail industry in coping with it. A final report titled: Rail-Highway Crossing Accident Liability Management Analysis is due from TSC on April 15, 1980, for review and approval. When this report is ready for distribution it will be made available to the Committee.

Question. The General Accounting Office raised a similar liability concern related to the States. The GAO suggested that part of the reluctance of the States to develop proper assessment of the dangers of their grade crossings and the need for signals at these crossings result from the fact that, "the State could be held liable for an accident at a crossing that met the criteria for warning device installation but did not have the device in place." Again, do you believe that this is a problem? The Department does not feel there has been a reluctance on the States' part to properly assess the hazards at crossings because of a liability concern. Under existing regulations governing he categorical railroad-highway crossing safety program, the States have the responsibility of assessing the relative hazards of crossings and establishing priorites for improvements. The States have been fulfilling

this task.

The need for signals or establishment of criteria or warrants for signals, either on a State-wide basis or national basis may raise liability concerns. We would note several States have adopted specific criteria for active warning devices and we are not aware of any specific liability issues that have resulted from these State criteria. However, the Department is undertaking an effort to establish uniform nationwide criteria for the installation of warning systems at railroad-highway crossings and the liability question has been raised.

Question. If so, what recommendations, if any, do you have to combat such situation?

Answer. Although the liability issue has been raised concerning establishment of warning device criteria, the Department does not feel it is a significant problem at this time. The Department believes the best approach for combating the potential of liability is to have a well documented aggressive program for addressing railroadhighway crossing problems and the States are encouraged to adopt this approach. Question. The General Accouting Office study cited earlier suggested that some of the signaling and other safety devices might not be as effective as had previously been expected, noting, "In fact, at least 30 percent of the collisions between trains and motor vehicles in 1975 occurred at crossings with functioning train-activated flashing lights and an additional 8 percent at crossings with functioning gates." Has any further analysis been carried out on the life-saving and injury-preventing potential of the various grade crossing safety devices?

Answer. During 1978, over 10 percent of railroad-highway crossing accidents involving motor vehicles occurred where gates were in place and over 36 percent occurred where other types of active warning devices were installed. It must be realized, however, that active devices are located at the most hazardous crossings and that the accident rates at these crossing would have been even higher had the active waring device not been there.

The Department has undertaken additional research efforts to determine and verify the accident-preventing potential of the various crossing warning devices. These efforts have included flashing lights, flashing lights with gates, and various advance warning sign designs to include activated advance warning signs.

Question. What types of predictions can we make about the effectiveness of these various devices?

Answer. The Transportation Systems Center has conducted before and after studies of nearly 3,000 crossings at which active warning devices have been installed and expects to complete a report titled: "The Effectiveness of Flashing Lights and Flashing Lights with Gates in Reducing Accident Frequency at Public Rail-highway Crossings, 1975-1978," during May 1980. This report will show that the installation of flashing lights with gates where only passive warning devices (signs) existed previously reduced the number of accidents by more than 84 percent. The installation of flashing lights (without gates) reduced accidents by nearly 65 percent. The addition of gates to already installed flashing lights reduced accidents by 64 percent. Question. Do you believe we are making progress in the rail crossing area? On what basis do you make this assessment?

Answer. Definite progress has been made in this area. The States have increased the obligation of funds authorized for crossing safety improvements each year since the Rail-Highway Crossings Programs was created in the Highway Safety Act of 1973. More and more States are reporting that standard signs and markings are being placed at all public crossings. The States and the railroads are cooperating to a much greater extent than ever before to work out agreements leading to the installation or upgrading of active warning devices and imporve crossing surfaces. An increasing number of States are initiating Operating Lifesaver programs to educate motorists of their responsibilities at crossings. A considerable amount of research is nearing completion or is underway or planned which deal with accident prediction capabilities, innovative and improved warning devices, and accident causation. All these efforts by the many participants in the railroad-highway crossing safety area certainly have contributed to the drop in fatalities since the beginning of the program. Preliminary 1979 figures indicate that fatalites should be the lowest since 1920. Further, the number of crossing accidents is holding approximately level in spite of the increased exposure due primarily to the number of vehicle miles driven.

Question. Have you carried out any estimates as to how much lifesaving potential and injury avoidance is likely at various levels of expenditures for specific safety protection? If so, would you provide this information and the basis for these estimates to the Committee?

Answer. The work which we have done has been accomplished at the Transportation Systems Center. To date, it has been done only to demonstrate the capability