COLUMN ONE : FAA’s Safety Response Record Hits Turbulence : Over the past decade, the agency has been slow to heed safety warnings--sometimes acting only after fatal crashes, according to a Times study.
As USAir Flight 1493 prepared to touch down at Los Angeles International Airport on Feb. 1, 1991, passenger David Richman, a Harvard-educated college professor and proud new father, knew nothing about the potential danger of runway collisions. Nor did any of the other 33 people who were about to die that Friday evening.
But the Federal Aviation Administration did know.
As far back as 1983, records show, air safety experts had urged FAA action to do more to prevent runway accidents. Even within the agency, officials had prodded their superiors to expedite ground-based radar systems and relatively inexpensive traffic lighting systems.
The FAA did take those actions and more--but only after Flight 1493 suddenly landed atop a Skywest commuter plane, killing 34 people. This was the third fatal runway collision in the United States in 13 months.
“How many crashes do you need?” asked Barry M. Sweedler, director of safety recommendations for the National Transportation Safety Board. “If you have one, there should never be another one with the same problem. Why do we need two, three, four or five?”
In dozens of instances over the past decade, the FAA has taken two years or more to respond to repeated warnings of air safety problems, and the agency often has acted only after loss of life, according to a four-month study by The Times.
The FAA, the federal agency responsible by law for ensuring air safety, has been named by the NTSB as a cause or factor in 103 airplane accidents and incidents between 1983 and last July that together killed 574 and injured 421.
And records also show that hundreds more people have died in crashes caused by problems to which the FAA had been alerted but failed to act.
In addition to runway collisions, these problems included turbulence caused by the wings of Boeing 757 jetliners, procedures for de-icing wings before takeoff, the refurbishment of aging aircraft, passenger access to emergency exits and the installation of devices that warn if a plane is flying too low.
Although he said safety is the agency’s “No. 1 job,” David Hinson, the administrator of the FAA, said of the list of lapses: “I think you’ve put your finger on something that we’re really trying to deal with.”
While Hinson was unfamiliar with the specific areas The Times was focusing on--most of which preceded his arrival in 1993--he said: “Hopefully, we’ll be more responsible in quicker fashion.”
FAA officials say the agency’s day-to-day operation addresses most issues “rather effectively,” often detecting and remedying safety problems before accidents occur. Each morning, they say, the agency deploys about 3,000 inspectors and runs the busiest airspace in the world--without major incident on most days.
“This is a formidable task,” Hinson said.
The FAA “acknowledges that it can be reactive in some cases, but that can be a proper response in many cases,” according to a statement by the agency in response to The Times.
“ Reaction often means we insist on learning from past accidents. At other times, reaction may mean that we are properly responding to public demand or concern, which is proper in our democratic form of government.”
Said Hinson: “If you look at the way the FAA deals with safety, about 95% of what we do . . . of what you or I would call preemptive safety efforts, those are efforts you probably never read about or write about.
“About 5% of what we do is after there is an incident or an accident.”
Airplane manufacturers and airlines said safety comes first with both the FAA and them, and that they would never compromise the lives of the flying public. “We always want to try to make a good system better,” said Joe Hopkins, spokesman for United Airlines.
A scheduled commercial airplane trip in the United States is regarded as among the safest forms of travel in the world. The odds of an air traveler dying in a crash are roughly the same as the odds of being killed by lightning, according to the National Safety Council.
The U.S airline industry had gone more than two years without a major accident before the string of fatal accidents this year. As of October, an estimated 843 people have died in plane crashes in 1994--one of the worst records since 1988, according to preliminary statistics by the NTSB, the agency that investigates accidents.
Now, NTSB sources say, investigators are exploring whether deficiencies in FAA oversight and operations may have been contributing factors in at least two of the three major accidents this year in Charlotte, N.C., Pittsburgh, Pa., and near Chicago.
After the FAA admitted in July that it may have mishandled reports on turbulence problems associated with Boeing 757 jetliners, The Times reviewed hundreds of airline crashes and thousands of government documents obtained under the Freedom of Information Act. Scores of past and present FAA officials, members of Congress, airline industry sources and safety experts were interviewed.
What emerged was a portrait of an agency that many times has been slow to address safety problems, particularly when they were controversial or costly to correct. Among the findings:
* The FAA’s performance sometimes has been compromised by poor communication between those charged with identifying potential safety problems and those with the power to act on them, and pressure from an influential industry.
“Any time you want to change the rules, the manufacturers scream,” said one FAA aircraft-certification official who spoke on condition of anonymity. “Anything that cost money to the manufacturers, we have to fight them on. They say: ‘Well, there hasn’t been an accident.’ We say: ‘Well, there could be.’ ”
* Deadly delays have occurred in part because a law requires the FAA to justify the cost of implementing proposed safety measures by showing that enough lives will be saved.
“It’s strangulation of safety by regulatory process,” observed Rep. James L. Oberstar (D-Minn.), chairman of the House Public Works and Transportation Committee’s panel on aviation.
* The FAA failed to heed repeated admonitions from oversight agencies and from within its own ranks to make safety a higher priority. And there exists, both inside and outside the FAA, a tacit acceptance that sometimes only accidents can spur the agency to take meaningful action.
Charles O. Miller, former head of the NTSB’s Bureau of Air Safety, said he has kept a file called “known precedents.”
“I have been keeping it for awhile because I was getting disturbed about seeing accidents happen from causes I had seen before,” he said.
Consider the case of runway safety.
In 1978, two researchers, using statistics compiled by the National Aeronautics and Space Administration’s Aviation Safety Reporting System, issued a report that said “incursions by aircraft on the runways of controlled airports represent a significant safety hazard.”
Between 1978 and 1983, at least three near-collisions involving major jetliners occurred on U.S. runways.
In 1985, after two Northwest Airlines DC-10s were involved in what the NTSB termed a “potentially disastrous” incident in Minneapolis, the safety board urged the FAA to expedite projects designed to prevent collisions.
For years the FAA had such projects in the works, including the development of ground-based radar systems to help air traffic controllers track airplanes on the tarmac and simple lighting systems to ensure that departing planes do not stray onto active runways, especially during inclement weather.
Even within the FAA, officials recognized that progress had been too slow.
An internal FAA memorandum sent to air traffic managers in 1986 acknowledged that the runway incursion threat required further action. It noted that many FAA projects designed to deal with the problem were incomplete and, even if finished, would have questionable efficacy.
That year the NTSB released the results of a special investigation on runway incursions, proffering to the FAA 33 recommendations designed to prevent them. One recommendation originally had been issued 13 years earlier, but was never acted upon.
During the next four years, according to internal FAA documents, the agency received dispatches from many corners of the industry, urging it to take preventive measures.
Oct. 27, 1987: John O’Brien, director of engineering and air safety for the Air Line Pilots Assn., said in a letter to FAA Associate Administrator Anthony Broderick: “Several accidents and incidents over the last few years have documented the seriousness of runway incursions.”
Aug. 11, 1989: An internal FAA memorandum declared: “The Air Traffic Operations Service has stated an immediate need for a runway incursion alert system. . . . We agree that the potential for a runway accident is a national concern.”
But that concern turned to catastrophe.
The first of three fatal runway incursions occurred in Atlanta on Jan. 18, 1990, followed by one in Detroit on Dec. 3, 1990, and then the Los Angeles crash in February, 1991. The death toll for the three crashes: 47 people, including David Richman, the college professor.
“The world changed at that moment,” said Richman’s father, Alex, a psychiatrist and professor in Nova Scotia. “On that . . . morning, we were ignorant. We thought we could trust the airlines and the government. We thought safety came first.”
After the Atlanta crash, which killed one man, the FAA mustered action teams to survey airports and identify potential problem areas. It also released a report in April, 1990, emphasizing that pilots could help cut runway incursions by being more aware of surrounding aircraft.
At the time, the agency also acknowledged: “FAA has not always coordinated its efforts to reduce incursions.”
Still, it was not until after the Los Angeles crash that the agency placed more emphasis and money into its long-extant runway-incursion project.
By then, according to the FAA’s own estimate, 11 runway accidents had occurred in the United States since 1970, resulting in 644 deaths and injuries, not including the casualties from the Atlanta, Los Angeles and Detroit crashes.
“I’m not a great fan of how fast we get things done in this organization,” said the FAA’s Michael J. Harrison, who spearheaded the runway collision prevention program after the crash of USAir Flight 1493. “Should the runway-incursion stuff have been done sooner? The answer to that is: of course.”
Since that time, the agency’s plan to develop a sophisticated ground-based radar system and place it at the nation’s busiest airports has fallen behind schedule and may not be completed until the turn of the century, according to the General Accounting Office.
“Something simmers on the back burner, and it doesn’t get done,” said Hugh E. Waterman, a former FAA manager who worked for the agency for 27 years before retiring in 1986. “But then, one governor or 60 regular human beings dies, then it goes onto the front burner.”
*
Created in 1958, the FAA has a double-barreled task that critics believe is a conflicting one: to promote the aviation industry and to ensure safety.
The NTSB is charged with investigating the causes of transportation accidents and making safety recommendations. Created in 1974, it has no regulatory powers and cannot force the FAA to act. Historically, the NTSB has wielded its influence most successfully by publicly drawing attention to the FAA’s lapses and leveraging legislative outrage.
The FAA points out that through the years it has identified numerous safety hazards and acted before accidents ever occurred. In a 20-page statement to The Times, FAA officials said the agency routinely detects safety problems but receives no credit because there is no way to tell how many accidents have been prevented.
“Airline safety is among the least recognized success stories in public or corporate policy in the United States,” the FAA statement said. “Improvements in engine reliability, aircraft design, avionics, cockpit technology, navigational aids and air traffic control have made serious accidents rare events.”
Although the FAA has adopted eight of every 10 safety recommendations made by the NTSB, the average amount of time it takes to implement the recommendations is slightly more than two years. In some cases, it’s been much longer.
While he believes that the NTSB functions as a “great auditor” for the FAA, Hinson said the FAA sometimes respectfully disagrees with the safety board, and with good reason.
“I think it’s probably OK to have disagreement on the 15% or so,” Hinson said, referring to the number of safety recommendations the agency doesn’t accept. “It’s not just . . . ‘We don’t like them’ and put them away. We have to defend our position.
“There are often cases in aviation where knowledgeable people can disagree.”
The NTSB’s Sweedler said the safety board was pleased with the “82.5% acceptance rate,” but, he said, “there are quite a few important issues in the other 18%.”
*
Records show a pattern of delay in correcting a number of problems:
* It took the FAA eight years to act on a problem with Cessna carburetors after John R. James, manager of the FAA’s aircraft certification office in Atlanta, had written a memo warning: “CAUTION: there have been reports of power loss, severe in some cases.”
The problem was causing planes to stall and crash. During that eight-year period, at least 10 accidents had occurred and half a dozen people had died.
* A year before the top half of a Boeing 737 ripped off over Hawaii in 1988 because of metal fatigue, FAA researcher Thomas Swift wrote a 77-page report saying fatigue in America’s aging air fleet was a safety risk: “It is possible for a number of cracks, each not easily inspected, to suddenly join together and form a long critical crack. . . . A number of fleets are currently operating at double their initially anticipated design life goals.”
* It took a decade of admonitions and at least four major crashes before the FAA acknowledged that certain planes, such as early model DC-9s and Fokker F-28s, were especially susceptible to control problems with minute amounts of ice on their wings.
Ten years after the NTSB had made safety recommendations following the deaths of 78 people in a 1982 Air Florida accident in Washington, a similar tragedy happened at La Guardia Airport in New York. That accident, involving USAir Flight 405, killed 27.
The Air Line Pilots Assn. said in a letter to the FAA at the time: “The lack of an adequate response to (the NTSB’s earlier) recommendation (on de-icing wings), which was made approximately 9 years ago, contributed to the accident involving USAir Flight 405.”
“The FAA,” said veteran air safety investigator Rudolf Kapustin, “makes it sound like: ‘Hey, we’ve got a brand-new problem here.’ Well, a lot of times it’s not a new problem. It’s an old problem.”
Kapustin, who has worked for the FAA and the NTSB, was the lead investigator in the 1982 Air Florida crash, which was caused, in part, by ice on the wings. “After the Fokker accident (at La Guardia Airport in 1992), the FAA convened this worldwide conference on de-icing. They said: ‘Now you’ve got to put a ladder up and touch (the wing to see if it has ice on it).’ Well, Jesus, we knew that years ago.”
The FAA itself concluded last July that its ability to act upon safety concerns in a timely manner was lacking. The agency had launched an internal investigation after The Times reported that it had mishandled the Boeing 757 turbulence issue.
But the acknowledgment came as no surprise to air safety experts.
In 1988, then-FAA Administrator Allan McArtor, in an internal memorandum, acknowledged that the agency lacked an internal clearinghouse for safety data, and that management of safety data within the agency was disorganized.
A July, 1988, Office of Technology Assessment report concluded: “More stringent safety standards usually follow a widely publicized airline accident and vocal public and congressional concern than from FAA initiatives.”
And last year, Robert E. Machol, the FAA’s chief scientist before retiring last summer, pointed out the agency’s shortcomings. In a March, 1993, memo, he detailed how the Boeing 757 wake-turbulence problem was handled. Before two crashes that claimed 13 lives, Machol had predicted that turbulence created by the jetliners would cause a “major crash” if the FAA failed to take preventive measures.
“This is symptomatic of a bigger problem, which is that we react very slowly to things where maybe we ought to act more rapidly,” Machol wrote. “We need to be more alert to safety questions.”
Next: Why delays in response occur.
Times researcher Sheila A. Kern and correspondent Shelby Grad contributed to this report.
The FAA Factor in Crashes
The Federal Aviation Administration is listed as a cause or factor in 93 plane crashes from1983 to July, 1994. Injury and fatality records for the 93 crashes show:
* 576 dead
* 151 seriously injured
* 270 moderately injured
An additional 10 crashes were attributed to errors by specific FAA employees, resulting in:
* Seven dead
* Eight moderately injured
Source: National Transportation Safety Board
Researched by JEFF BRAZIL and SHELBY GRAD / Los Angeles Times
Runway Incursions
From 1990 to 1993, the number of aircraft runway collisions has declined by one-third despite an increase in air traffic: ‘93: 188 Source: Federal Aviation Administration
Researched by JEFF BRAZIL and SHELBY GRAD / Los Angeles Times
How the Study Was Conducted
To complete this report, The Times reviewed roughly 20,000 pages of internal documents obtained from the Federal Aviation Administration through the Freedom of Information Act. The newspaper also reviewed a computer analysis of the causes of airplane crashes between 1983 and last July by the National Transportation Safety Board. The Times also studied dozens of reports by government oversight agencies, and interviewed scores of present and past FAA officials, airline industry sources, aviation safety experts, members of Congress, crash survivors and the families of crash victims.
