1974_美国_HAR7405.pdf

8 $$ BG^{m-X} $$

NATIONAL TRANSPORTATION SAFETY BOARD

FOR RELEASE: 6:30 AM FEB · 6 1975

HIGHWAY ACCIDENT REPORT

GREYHOUND BUS COLLISION WITH CONCRETE OVERPASS SUPPORT COLUMN ON I-880, SAN JUAN OVERPASS, SACRAMENTO, CALIFORNIA NOVEMBER 3, 1973

〔图〕NATIONALSAFETYBOARD\n NATIONAL TRANSPORTATION SAFETY BOARD Washington, D.C. 20591 REPORT NUMBER: NTSB-HAR-74-5

SS-H-34

HIGHWAY ACCIDENT REPORT

GREYHOUND BUS COLLISION WITH CONCRETE OVERPASS SUPPORT COLUMN ON I-880, SAN JUAN OVERPASS, SACRAMENTO, CALIFORNIA NOVEMBER 3,1973

ADOPTED: DECEMBER 13, 1974

NATIONAL TRANSPORTATION SAFETY BOARD Wechiosato, P. O., 20501

Washington, D.C. 20591

REPORT NUMBER; NTSB-HAR-74-5

TECHNICAL REPORT DOCUMENTATION PAGE

FOREWORD

The accident described in this report has been determined to be a major accident by the National Transportation Safety Board under the criteria established in the Safety Board's regulations.

The Bureau of Motor Carrier Safety of the Federal Highway Administration, the California Highway Patrol (CHP), the Coroner of Sacramento County, and the Civil Aeromedical Institute of the Federal Aviation Administration cooperated in this investigation.

The conclusions, the determination of probable cause, and the recommendations herein are those of the Safety Board.

TABLE OF CONTENTS

NATIONAL TRANSPORTATION SAFETY BOARD WASHINGTON, D. C. 20591

HIGHWAY ACC IDENT REPORT

Adopted: December 13, 1974

Greyhound Bus Collision with Concrete Overpass Support Column on I-880, San Juan Overpass Sacramento, California November 3, 1973

SYNOPSIS

At 9 p.m., P.s.t., on November 3, 1973, a charter Greyhound bus, eastbound on I-880 near Sacramento, California, ran off the road, overrode a guardrail, and collided with a column which supported the San Juan Bridge Overpass. The speed of the bus was approximately 67 mph. The driver was seen to be incapacitated, but an effort by a passenger to take control was too late.

The bridge support column penetrated the center front of the bus to a depth of 21 feet. The driver and 12 passengers were killed; 22 pas-sengers were hospitalized for moderate to critical injuries; 11 passengers were treated for minor injuries.

The National Transportation Safety Board determines that the probable cause of this accident was driver incapacitation. The reason for the in-capacitation could not be determined. Contributing to the fatalities and injuries were: (a) The failure of the guardrail to direct the bus away from the bridge column; (b) the velocity with which the bus struck the bridge column; and (c) the structural failure of the seats in areas of the bus where crash forces were moderate.

FACTS

The Accident

Just before 9 p.m. on November 3, 1973, a charter Greyhound bus, carrying 45 passengers and a driver, was approaching the San Juan Bridge Overpass eastbound on I-880 near Sacramento, California.

As the bus approached the overpass in the No. 1 (left) lane, it passed a passenger car and stabilized in the center lane. The bus then turned into the No. 3 (outside) lane. During this maneuver, it veered in front of a small car and caused its driver to take strong braking and steering action to avoid colliding with the bus. The bus did not recover from the turn and continued to its right. It crossed the outside lane, passed over the shoulder, mounted a 6-inch curb, override the guardrail,

• 2 -

and impacted a 48-inch-diameter support column of the bridge overpass. The bus' speed was approximately 67 mph.

Witnesses stated that they did not see turn signals nor brake lights during any maneuvers by the bus after it left the No. 1 lane.

Accident Site

Interstate 880 is a six-lane divided highway with three lanes for eastbound traffic and three lanes for westbound traffic. The lanes are 12 feet wide, with a 10-foot asphalt concrete shoulder on the left and a 10-foot asphalt concrete shoulder on the right. I-880 is generally straight and flat in the area of the accident site and had a speed limit of 65 mph. (See Figure 1.) At the time of the accident, the weather was clear, the road was dry, and the moon was full. There were no lights under the bridge, which, by contrast with the road, was dark and shadowy. Before November 3, no accidents had occurred at this site.

Except for the height of the guardrail, highway features met California Highway Standards. The guardrail was installed 21 to 23 inches above the surface of the 6-inch long curb, rather than at 27 inches as required by the standard. (See Figure 2.)

The Bus

The bus was a two-level, Greyhound Sceni-Cruiser, GMC Model 4501-SL, and was manufactured in 1953. The lower level accommodated 10 passengers, and the main passenger compartment, which was 28 inches higher, seated 37 passengers. (See Figure 3.)

The bus was designed according to the lightweight, fuel-saving concept. The basic structure was of relatively light gage steel and aluminum. The engine and transmission were located at the rear. The front space was occupied by the driver and passengers. The bus engine was governed so that the bus could not exceed a speed of 67 mph on a flat surface.

In addition to the outward-opening door on the right side, the bus had five large windows on each side which opened. They were hinged at the top; the windowsills were 8 feet above the ground. Windows could be released by lifting a release bar and pushing the windows out. The bus had no rear or operable roof exit.

After the accident, the Bureau of Motor Carrier Safety (BMCS) made a mechanical inspection of the bus. Their inspection revealed that the tires were in compliance. Because of the severe vehicle damage, complete system tests could not be performed. However, brakes, brake lights, taillights, and turn signals operated normally. No abnormalities could be found with the steering system which could have resulted in unwanted

• 3 -

〔图〕I-880 & SAN JUANOVERPASS -POINT OF IMPACTAERIAL VIEWON RAMPEAST BOUND LANEZ\n

Figure 1. Aerial view - I-880 & San Juan overpass - point of impact.

• 4 -

〔图〕21" HIGH W-SECTION GUARD RAILBRIDGE SUPPORT COLUMNS6"HIGH ISLAND CURBYELLOW EDGE STRIPEIMPACT AREA.(SEE DETAIL 'A')NBUSC

Figure 2. Accident site, showing path of bus within 900 feet of impact.

• 5 -

〔图〕NON STRUCTURAL21STRUCTURALHEIGHT GUARDRAIL\n

Figure 3. Sceni-Cruiser, GMC Model 4501SL, 1953.

• 6 -

steering action or inability to steer if the power steering pump failed. Wheels, brakes, and suspension were without defect. The engine governor was removed and tested in an independent laboratory under Safety Board supervision. Later, it was installed on a similar bus. In both the laboratory tests and actual performance tests, it performed according to specifications. The BMCS review of the maintenance record revealed no violation of regulations and indicated that the bus, rebuilt three times during its 20 years, had been serviced and inspected regularly.

The Driver

The driver of the bus was a 26-year-old male, 5 feet 10 inches tall, and weighed 210 pounds. He had been employed by Greyhound since April 1972. In addition to a truckdriver training course in 1971, he had received 3 weeks of training as a busdriver from the carrier. His driver training record was average, and his record with Greyhound was good. His California motor vehicle record shows one accident and no violations or arrests. The driver had the required State of California Driver's License and met the medical qualifications of the Federal Motor Carrier Safety Regulations. His first and only medical examination pursuant to CFR 391.43, was on April 14, 1972. His driving log was current and in compliance with regulations regarding hours of duty. The driver's family indicated that he had 13 hours of sleep within the 24 hours before the accident. They also reported that he normally spent a good deal of his off-duty time at home and resting. He went out only one night a week to bowl; he was considered a social drinker.

The Coroner of Sacramento County attributed the driver's death to multiple traumatic injuries with contusions of the brain, positive hemorrhages, and multiple fractures of facial bones. Blood alcohol and drug levels were negative. There was no reported indication of physiological failure. The Coroner's Record also indicated that the driver suffered from fatty alteration of the liver and chronic gastritis.

Bus Occupants

The 45 passengers had chartered the bus for a trip to Reno, Nevada. Their ages ranged from 19 to 78 years. Most passengers were asleep at the time of the accident.

Damage

Vehicle. -- The overpass column penetrated the center front of the bus to a depth of 21 feet. The collision tore the front of the bus apart, drove the front axle and wheels back to the cargo compartment, and ripped up the floor in the center and on the right side of the bus. (See Figures 4 and 5.) The driver's station and controls separated from the bus structure and were thrown against another support column, 52 feet beyond the point of impact. (See Figure 6.)

• 7 -

〔图〕83A00910\n

Figure 4. Right side showing depth of penetration and general damage.

• 8 -

〔图〕插图

Figure 5. Left side showing extent of damage. NOTE: Front wheels in cargo area.

〔图〕er's station, controls, and seat at the support columm, 50 feet from impabus.\n

Figure 6. The driver's station, controls, and seat at the support column, 50 feet from impact point of bus.

• 10 -

Horizontal scrape marks were located on the right rearward side of the bus along the lower side. Rub marks were found on the right front wheel. Wood fibers were also lodged in the wheel lugs and on a bracket attached to the underside of the front axle. Timber splinters were lodged between the tire and wheel on the right rear outside wheel.

Scuff marks were found on the right front tire and on the two outside tires of the right tandems. The left front tire was cut completely through the wall and tread.

The front bumper was bent into a "V" and fractured in two locations. A brown substance resembling creosote preservative was present on the bumper surface and extended upward 3½ inches from the bumper's lower edge. The lower edge of the bumper as shown on a standard bus drawing is 18 inches above the ground.

The interior of the bus was severely damaged. (See Figure 7.) The only seats that did not fail were the five seats across the back. Seatback hinges, legs, and floor attachments failed on all seats available for examination following the accident. Although some seats had been moved to facilitate rescue, those seats which remained in the bus were not completely separated from the outboard wall attachment points. The inboard legs, however, were broken from the floor attachment plates.

Guardrails. -- The left line of timber posts between the guardrail nose and the first bridge support column was fragmented. The right line of posts was partially sheared and split vertically. Although the right line of posts remained in position after the accident, they were damaged too severely to be reused.

The W-beam rail was damaged on the left side of the installation. The return section (nose) was deformed on its left side and scratched horizontally. The W-beam was sheared about 6 feet from the nose. The W-beam was torn vertically from its top to a point below its horizontal centerline and sheared longitudinally for 6 feet. The shearing was such that the continuity of the W-beam was not broken. (See Figures 8 and 9.)

After the crash, the beam was lying on the island, face up. The top edge of the beam was displaced slightly from its original horizontal position.

Past the first column, only small chips of wood were dislodged from tops of several of the timber posts in the left rail.

Only one tire mark less than 3 feet long was located on the roadway surface. It was located on the left curb top and face 16 feet from the guardrail nose and 50 feet west from the face of the first column. The tire mark was not the result of braking.

• 11 -

〔图〕插图

Figure 7. Interior view of the bus showing extent of damage and failed seats.

• 12 -

〔图〕插图

Figure 8. Damage to guardrail, view of sheared posts and paint marks from the bus on bridge support column.

· 13 ·

〔图〕插图

Figure 9. The guardrail after repair. The tape represents the path of the bus.

• 14 -

Injuries and Fatalities. -- Thirteen of the 46 occupants of the bus died of multiple traumatic injuries, including crushing injuries of the chest, rupture of the heart and aorta, skull fractures, multiple fractures, and, in one case, transection.

Eleven passengers suffered minor injuries and bruises and were released following brief hospitalization. The remaining 22 persons were moderately to critically injured and sustained head, chest, abdominal, and upper and lower limb injuries.

Information regarding the most likely seating of 38 of the 45 passengers before the accident was obtained through interviews with 23 of the passengers. (See Figure 10.) There was some shifting of passengers during the trip; consequently, there is the possibility that some of the locations are incorrect and do not reflect injury relatable to that location. Three or four of the passengers were ejected from the bus.

Incapacitation of Driver

The Safety Board was able to interview 19 of the passengers. The general consensus was that the passengers were orderly, and that, at the time of the accident, many of them were asleep. None of those interviewed suggested that the driving was other than smooth and normal before the accident.

The host on the bus was in the window seat next to the cohost and just behind the driver at the time of the accident. The host reported that he saw the driver's hand go up and out on the left and "...then he clutched his chest. He took his hands off the wheel, and they were both off then. He did that twice..then I hit the cohost..,and I said, 'Hey, something's going wrong here!' We'd just come into the lane going toward the big abutment. I could see that we were going to hit, so I said to the cohost, 'See what you can do!' He stood up (the cohost) and by the time he stood up, it was too late." The host gave a similar account of the accident to the California Highway Patrol (CHP).

Postaccident Activities

Immediately after the accident, some bus windows were opened, and a few slightly injured passengers were helped out of the bus by passersby and the Sacramento Police, who arrived 5 to 6 minutes after the accident. Passengers reported that getting out of the bus was difficult, because it was dark, they were confused and jammed between seats, and it was 8 feet from the windowsill to the ground. They commented that the windows were heavy and that there was no way to hold them open.

The first fire and rescue equipment was dispatched at 9:09 p.m. and arrived 10 to 15 minutes later. Fire and rescue personnel used ladders to reach the windows which were held open with rope and supports. (See

• 15 -

〔图〕PENETRATION PATHFRONTF534FWash21FTRoomFFU52-U-3U312321U115511513U2U1U22U22\n

Figure 10. Passenger seating and injury disposition.

• 16 -

Figure 11.) The last victim was removed from the bus at 11:49 p.m. Although the rescue effort took almost 3 hours, it was timely and efficient, considering the extent of damage to the bus. The Safety Board believes that the organized rescue and medical response minimized the losses resulting from this accident.

ANALYSIS

The driver changed from lane No. 1 and continued briefly in lane No. 2. He was probably in control at that time. Shortly after the maneuver, however, the bus made a shallow turn to the right and continued on that course until it impacted the bridge support column.

There was absolutely no indication that the road or any other highway feature could have induced the unwanted change in the bus' direction. Examination of the highway provided no evidence of design features, including signing and marking, which could have miscued the driver. Examination of the mechanical, hydraulic, and electrical systems of the bus, as well as the bus maintenance record, indicated that the course change was not caused by a failure of any of the vehicle's systems. In addition, evidence indicates that the brakes were not used.

There was no reason to believe that the maneuver to the right was intentional. There were no highway conditions or requirements, such as signs or location of exits, that would have necessitated an immediate transition into lane No. 3. Therefore, although the Coroner's Record does not provide evidence of physiological failure, the testimony of the host, who sat behind the driver and observed his unusual behavior, the reports by outside observers that turn signals were not displayed, and the fact that there was no evidence of evasive action or braking, justify the conclusion that the driver was incapacitated and was, therefore, not in control of the vehicle.

Since there was no reason for the driver to consciously direct the vehicle to the right, the maneuver was probably initiated by involuntary inputs to the steering wheel when the driver was incapacitated. Once such turning started, it could be aggravated by protuberances on the road surface, low tire pressure, faulty wheel alignment, unbalanced distribution of weight, and the slope of the road.

As there is no provision for a backup operator or for automatic stopping of buses, the BMCS attempts to exclude drivers who might fail for "physical, mental, or organic defects of such a character and extent as to affect the applicants' ability to operate a motor vehicle safely." 1/ The BMCS recognizes the importance of the physical and mental integrity of drivers as a means of protecting the public in its general instructions.

1/ Federal Motor Carrier Safety Regulations, Subpart E, Physical Qualifications and Examinations, 49 CFR 391, 41-49.

• 17 -

〔图〕ROPESUPPORTLNATOMAS16\n

Figure 11. Rescue operations showing use of ropes, ladders, and window supports.

• 18 -

to the examining physician contained in 49 CFR 391.43 and 391.49, in which waiver of certain physical defects does not apply to the drivers of vehicles transporting passengers or placarded for hazardous cargo.

The good intention of the BMCS is not, however, reflected in the physical examination which is specified. The examination covers in general all of the major physiological systems but, with few exceptions, it places the burden of indepth examination on the examining physician. Since any physician or osteopath may conduct the examination, it is not to be expected, except in those instances where the physician or osteopath is fully employed by a carrier, that he is conversant with the requirements of intercity bus and truck operations and the medical considerations which would "affect the applicants' ability to operate a motor vehicle safely...." In addition, he may not necessarily be knowledgeable of the accident literature that would alert him to clinical observations critical to bus and trucking operations.

Aviation, from the beginning, recognized the need for medical specialists to evaluate the fitness of pilots. The Federal Aviation Administration requires that all medical examinations of airman certificate holders be accomplished by certified Aviation Medical Examiners. Through this means, only physicians who are conversant with the medical aspects of the flight environment are involved in examinations. Through this approach, the examination is better standardized; that is, there is better quality control, there is more assurance that clinical observations will be evalu- ated in relation to flight requirements and that tests of special import- ance in evaluating flight personnel are performed by experienced personnel.

Physicians who specialize in the medical aspects of motor carrier operation have increased ability to recognize clinically disabilities that could lead to sudden incapacitation and could employ techniques not routine in general practice.

The driver in this accident was about 45 pounds overweight for his age and height; he seemed to require more than the usual amount of sleep; and based on the autopsy report, he had fatty alteration of the liver and chronic gastritis. Except for the record of overweight, the certifying physician had no information to suggest any medical disability.

Obesity, the need for more than the usual amount of sleep, degeneration of the liver, and chronic gastritis suggest a host of recognizable disabilities that could have resulted in the incapacitation experienced by the driver in this accident. Narcolepsy (pathologic sleepiness) 2/ and

2/ The Sleepy Driver: A Test to Measure Ability to Maintain Alertness, R. E. Yoss, MD. Mayo Clinic Proceedings, Vol. 44, Rochester, Minnesota, November 1969.

• 19 -

reactive hypoglycemia are among the more likely possibilities. A more demanding medical examination might have disclosed any disabilities of the driver.

The Board realizes the limitations of the suggested certified Highway Medical Examiner program; however, the need to detect the driver who is subject to sudden incapacitation is so important that a concerted effort must be made to accomplish the objective of better detection.

Factors Contributing to the Magnitude of the Loss

1. Bus Design Features. -- Because of the severity of the accident and the urgent requirement to remove the injured from the bus, much of the structure of the bus was pulled apart and interior equipment was removed to facilitate the rescue. Enough of the bus was untouched by the rescue operation, however, to establish the primary contributors to passenger injury. The head, leg, and chest injuries observed in this accident are not unlike those suffered by unbelted passengers in aircraft accidents 3/ who impact the seats in front of them. The failed legs at the attachment plates, the rip out of the arm rest attachment nut plates from their metal interwall support, and the interviews with passengers support this view.

The seatback frames were constructed of tubular metal with little padding. Only light plastic deformation (bending) was found on the units. Clean breaks at weld joints and fittings were the most frequently observed failures and attest to the unyielding nature of the seat structure. This type of construction is conducive to injury.

Some of the passengers impacted the penetrating bridge column and some were thrown out of the bus. (These passengers were exposed conditions beyond human tolerance.) On the other hand, 11 passengers sustained minor injuries and were quickly released from the hospital; 22 others, though more severely injured, survived. This accident may, therefore, with the exception of the front of the bus and the path of the support column, be considered a survivable accident. The acceleration forces were calculated, based on the bus' speed and stopping distance, to average about 7.5g, with an approximate duration of 0.42 seconds. Such forces are well within the bounds of human tolerance, if the seats had stayed in place and if seatbelts had been available and used.

There are currently no Federal Standards for bus seats. Nonetheless, there have been improvements in seat design, and the crashworthiness of the newer buses is greater than that of older buses such as the one involved in this accident.

• 20 -

If the seats had met the standards in the National Highway Traffic Safety Administration's (NHTSA) Notice of Proposed Rulemaking, Docket 73-3, Notice 1, proposed on February 1, 1973, (and withdrawn in Docket No. 73-4, Notice 2, on July 30, 1974) the injuries suffered by passengers to the rear of row 3 would possibly have been less severe. However, whether more passengers would have survived in the new seats without seatbelts cannot be determined. The higher seatback proposed in the withdrawn rule would have benefited the seatbelt user, because with seatbelts, the proposed seats would probably have reduced lower and upper leg fractures and, in conjunction with the padded seatbacks, reduced head injuries and lacerations.

1. Window Design Features. -- Unquestionably, rescue and removal of the seriously injured were made more difficult and more time-consuming because the windows were 8 feet above the ground, difficult to open from the outside, and difficult to hold open. Had a fire occurred, the delay in evacuating passengers through the windows would undoubtedly have resulted in a greater number of fatalities.

1. Structural Crash Resistance. -- Because of the design of the bus, the first two rows of passengers were seated above the front axle and front wheel wells. As much as 2/3 of the weight of the bus was located in the rear half of the bus.

In addition, the bumper at the right front door, capable of contacting a guardrail, is not structural, but is merely attached to the door. (See Figure 3.) The frontal member of the bumper is structural. Where contact of the front corner of the bus with guardrails is possible, the structural members are the end of the bumper and the reinforced edge of the wheel-housing which supports the bottom step. The latter was bent rearward during the collision, apparently producing some upward force on the bus. These elements are not well adapted to guardrail contact.

The weight distribution of this bus was favorable from the stand-point of guardrail action in directing the front of the bus away from the bridge column. There was no other point on the bus so suitable for producing a large turning effect by applying a small force as the right front corner. However, there was no apparent consideration of a deflecting function in the structural design of the bus, and the guardrail was not designed to be impacted by a bus. Thus, the advantage for barrier deflection of a light front end was not explored.

When the bumper struck the column, the bus decelerated. Although the relatively light gage bus structure was adequate to sustain road forces, it lacked continuous heavy members which could resist deformation by the concentrated force applied by the bridge column. Only the front axle resisted severe bending by the bridge column. Thus, the lighter gage distributed structure of the front portion of the bus was overcome in sequence, and the rear half of the bus was decelerated to

• 21 -

a stop somewhat evenly from about 65 mph in 21 feet. In effect, the invasion of the front of the bus cushioned those seated in the rear. The cushioning, at 7.5g, is less than currently required by aviation standards for commercial aircraft seat design. Although this accident occurred at a high speed, it is possible that light-weight, rear-engined buses which have more weight in the rear than in the front may be unnecessarily penetrated in a forward collision, and that passengers may be seated in zones subject to collapse. There are no Federal standards for structural crash resistance of buses, and relatively few tests of bus crash resistance have been made. The Board has no knowledge of any efforts to improve crash performance of modern buses. Seating and luggage space location as a method of protecting passengers and driver deserves investigation.

4. Highway Factors. --

a) The traffic control devices at and near the scene were reviewed under both daylight and dark conditions. The scene was studied to determine whether a driver would have likely mistaken the area in front and to the right of the bridge columns for an offramp. In addition, the approach signing was reviewed to determine whether messages to a driver could be misunderstood. No condition was found that would definitely indicate that any of the traffic control devices, singularly or in combination, could have indicated to the busdriver that an exit ramp existed at that location.

b) The bus struck the guardrail to the left of the forward rail terminal and slightly deformed the terminal. The vehicle's bumper then contacted the left side of the guardrail installation that flared to the left. The guardrail, at an approximate 15° angle with the centerline of the bus, yielded to the loading by moving to the right. At a point 6 feet from the forward guardrail end, the bumper override the W-beam. Both the W-beam and the underside of the bumper sheared at that point. Shearing of the W-beam continued below its longitudinal centerline for about 7 feet until a second support post, about 7 feet from the initial point of shear, was sheared and the beam cross-section was rotated clockwise. The W-beam most likely sheared longitudinally when it was contacted by the front axle, after it had been sheared vertically by the bumper and overridden by the vehicle.

The contact with the left guardrail beam and shearing of its post apparently had little effect on the trajectory of the bus.

With increasing penetration of the bridge support column into the bus, the rear of the bus rotated to the left about 5 feet at point of rest. The bus' longitudinal centerline struck an estimated 6 inches to the left of the column center. The path of the bus, from a tire mark on the curb to a post on the right guardrail about 6 feet from the column, transcribes a path on a slight left curve, after the initial contact

• 22 -

with the forward end of the guardrail. The effect of the guardrail on the direction of the bus trajectory was, therefore, inconsequential.

The energy dissipated in shearing the timber post of the left guard-rail was estimated to be less than 1 percent of that generated by the bus. Therefore, little deceleration from the estimated speed of 67 mph could be expected before impact with the column.

The compatibility of the bus and a 27-inch-high, W-section guardrail was reviewed to determine if any change in the bus trajectory and damage would have resulted if the guardrail had been at its required height. The Safety Board believes that the results of the accident would not have been changed significantly if the guardrail had been 6 inches higher. The weight, speed, angle of impact, and height of the bus' center of gravity would have produced forces too great to be controlled by the standard W-section guardrail used in California.

Traffic barrier systems in the United States have not been designed to withstand impacts by heavier vehicles or by vehicles with high centers of gravity. In barrier system tests, the criterion accepted by the FHWA prior to this accident was based on the use of a vehicle weighing 4,000 pounds to impact the barrier at speeds up to 60 mph and at angles up to 25°. California dynamically tests its barrier systems with a 4,500-pound vehicle, at angles up to 25°, and at 65 mph. The weight of California's standard test vehicle is only 1/9 that of the bus in this accident. Test data on impacts of heavier vehicles with guardrail systems are limited. However, some median barriers in European countries have been designed to withstand vehicles up to 18,000 pounds. 4/

Substantial research in the area of vehicle and highway comptability is needed. Until recently, research to design a barrier system that would safely restrain both heavy and light vehicles has not been undertaken because of the expected high cost to install such systems and also the belief that a system that would restrain heavier vehicles would be a hazard to occupants of lighter vehicles. Current research conducted by FHWA on new concepts of bridge rails indicates that barrier systems can be designed to manage the energy of both large and small vehicles and inter-city buses without adverse effects to occupants of either vehicle.

There is little evidence to indicate that substantial unified efforts have been made between those organizations responsible for barriers and those responsible for vehicles to make the design elements of the two structural systems compatible. For instance, the protruding lugs on the front wheels of a bus or truck can become a metal-cutting buzz saw when contacting a barrier rail. The result of permitting damaging design elements on the vehicle could be a destructed railing

23

system that permits penetration by the colliding vehicle. The exposed lugs could also contribute to lifting the vehicle vertically and cause override of a barrier system.

In a test conducted by FHWA on a new bridge rail concept, a "school-bus-type" vehicle was used in an impact test. The protruding lugs on the front wheel of the bus that contacted the rail were sheared and permitted the wheel to come loose. Other damage to the vehicle was minor. Wheel loss can, of course, cause tragic consequences.

In addition to the bridge rail test, under the FHWA program for "advanced vehicle protective systems," a concrete safety-shape median barrier was dynamically tested with a 48,000-pound tractor-semitrailer with good results. One truck test at 45 mph and at a $15^{\circ}$ angle resulted in minor damage to the vehicle.

The program by FHWA to develop barrier systems for various applica-tions that are compatible with a wide range of vehicles is a necessary and important step forward.

CONCLUSIONS

1. The busdriver was incapacitated at the time of the accident for reasons unknown.

1. Some medical causes of sudden incapacitation of drivers are detectable by more extensive physical examination of candidates than those now routinely required by BMCS regulations.

1. There is no evidence that the highway delineation, marking, or lighting contributed to the accident.

1. There is no evidence that the bus was in any way mechanically defective.

1. The guardrail was inadequate to redirect the bus.

1. The W-section guardrail was lower than required by California standards. However, a standard height guardrail of the same design would not have redirected the bus nor changed the severity of the accident.

1. Guardrails installed on interstate highways are, in general, inadequate for the full range of vehicles that use the highway.

1. There are no mandatory Federal requirements for the design and location of highway barriers or for their restraint performance.

1. The depth of penetration of the bus by the concrete column to over half the length of the bus was not a result of any abnormality of

• 24 -

the bus. There are no Federal standards which govern structural crash-worthiness of interstate buses, nor has the crash-restraint design of buses been influenced by private testing.

1. Except in the zone of penetration, the accident was survivable. Because of the low average deceleration which resulted from the collapse of the penetrated parts of the bus, fewer serious injuries could have resulted if the seats had remained intact.

1. If the seats in the bus involved in this accident had met the withdrawn NHTSA standard, the injuries to those who survived in the rear of the bus would have been less severe. Seatbelts undoubtedly would have reduced leg injuries had seats able to sustain seatbelt loads been installed.

1. The penetration of the front of the bus was not unexpected in view of the bus design and the forces produced by the high-speed impact with the bridge support column. The occurrence, however, raises the question whether present designs are the most desirable.

1. Evacuation of the bus was difficult because of

$^{0}$ the height of the windowsills above the ground,

• the absence of emergency illumination in the bus,

• the difficulty of opening the windows from the inside,

• the lack of a device to hold the windows open, and

• the absence of any planned means of opening windows by rescuers from the outside.

1. The occurrence of a fire involving the fuel from the ruptured fuel tank, in the circumstances of the difficult evacuation, would probably have produced many additional fatalities. The fact that diesel fuel, as opposed to gasoline, was used reduced the likelihood of fire.

1. Knowledge regarding the crash interaction between the exiting variety of vehicles and roadway protective structures is limited.

PROBABLE CAUSE

The National Transportation Safety Board determines that the probable cause of this accident was driver incapacitation. The reason for the incapacitation could not be determined. Contributing to the fatalities and injuries were: (a) The failure of the guardrail to direct the bus away from the bridge column; (b) the velocity with which the bus struck the bridge column; and (c) the structural failure of the seats in areas of the bus where crash forces were moderate.

25

RECOMMENDATIONS

The National Transportation Safety Board recommends that the Federal Highway Administration:

1. Investigate and develop criteria and issue regulations for medical examinations which will serve objectively to detect individuals susceptible to sudden incapacitation, such as heart conditions, epilepsy, narcolepsy, and hypoglycemia, and which will be required at least for the initial examination of those drivers who operate buses in interstate commerce. (Recommendation H-74-36)

1. Establish regulations to facilitate evacuation of buses in an emergency. The incorporation of emergency lighting systems actuated through impact and entry for rescuers should be included in the regulation. Attention is called to the Board's investigation of the interstate bus accident in Baker, California, in 1968, in which it was recommended that "no new type buses go into service which have not been tested to insure that all occupants can escape rapidly." (Recommendation H-74-37)

1. Promulgate mandatory national performance standards for traffic barrier systems. Those standards should contain criteria for dynamic testing or analytical procedures substantiated by such tests for each design to increase the compatibility of barriers with both light and heavy vehicles. The standards should also contain requirements regarding the placement of the barriers in the field to assure that compatibility of the vehicle/barrier is not compromised by adjacent environment. (Recommendation H-74-38)

1. Require the California Department of Transportation to review existing guardrail installations to assure that, where the guardrail has been offset behind a curb or a dike, the top of the guardrail is at a height in conformance with the California standard. (Recommendation H-74-39)

BY THE NATIONAL TRANSPORTATION SAFETY BOARD

/s/ JOHN H. REED Chairman /s/ LOUIS M. THAYER Member /s/ ISABEL A. BURGESS Member /s/ WILLIAM R. HALEY Member participate in the adoption of this report.

Francis H. McAdams, Member, did not participate in the adoption of this report. December 13, 1974