Majority of incidents caused or made worse by human error
LONDON (Reuters) — Following every major disaster there is an intense focus on the safety of the technology concerned, but accident reports show the majority of incidents are caused or made much worse by human error.
This month's fiery train derailment in Quebec and dramatic plane crash in San Francisco have inevitably triggered an intense focus on the safety of the technology concerned.
In many instances it is not the technology that is unsafe but the operational procedures and practices used by employees and supervisors.
Time and again, reports published by the United States National Transportation Safety Board (NTSB) and other accident investigators have pinned blame on poorly trained personnel, who have failed to work effectively in teams, and have allowed a "culture of deviance" to develop in which required procedures were ignored for the sake of convenience.
"No system can operate safely when a culture of deviance from procedural adherence has become the norm," the NTSB observed in a recent report.
The instinctive reaction to disasters is to label the technology as hazardous and seek to restrict it or ban it altogether. But identifying operational lapses and fostering a more effective safety culture is usually a more relevant response.
Ignoring safety restrictions
Research into the 1986 space shuttle Challenger accident found that technical personnel allowed a culture of deviance to develop, in which it became acceptable to ignore or work around safety systems and requirements.
"An early decision to continue shuttle operations in violation of requirements cultivated an operating culture in which not adhering to requirements became the norm," the NTSB said. "Decisions made thereafter made it easier for shuttle personnel to avoid adhering to other requirements, thus 'normalising' the deviation from technical requirements.”
Investigating a 2010 oil spill at Marshall in Michigan, when almost 850,000 gallons of crude was discharged into a wetland from a broken pipeline, the NTSB found similar failures made the incident far worse. The spill was not detected or addressed for over 17 hours, during which time control room staff repeatedly ignored alarms and tried to restart the pipeline twice.
"Ineffective performance of control centre staff led them to misinterpret the rupture... which led them to attempt two subsequent start-ups of the line," the NTSB concluded.
Enbridge, the pipeline operator, "failed to train control centre staff in team performance," it determined, as the supervisor in charge of operations delegated authority to a subordinate.
"Although Enbridge had procedures that required a pipeline shutdown after 10 minutes of uncertain operational status, Enbridge control centre staff had developed a culture that accepted not adhering to procedures," the NTSB found.
Procedural deviance is not limited to pipelines. Investigating the collision of a freight train with 19 stationary railcars at Bettendorf, Iowa, in 2009, the board found that the crew of one train "failed to follow a critical operating rule." If they had complied "the accident would not have occurred."
Investigating another rail crash in 2011, the NTSB determined the probable cause was "the failure of the crew of the striking train to comply with the signal indication requiring them to operate in accordance with restricted speed requirements and stop short of the standing train because they had fallen asleep due to fatigue."
Irregular work schedules and sleep disorders contributed to the fatigue, and had not been handled in line with best operational practice, in part because staff training was not mandatory.
Crew resource management
In many incidents, the way individuals work together as a team has been at fault. In the Enbridge disaster, the NTSB concluded "the ineffective performance of the control centre team... is consistent with human factors research on team performance, which has shown that the quality of team performance is influenced by team structure and team leadership."
"Although Enbridge control centre staff worked in teams, they were not trained to do so," the NTSB observed.
By contrast, the U.S. Federal Aviation Administration (FAA) requires all airline pilots to have specific training in team performance. The FAA's Crew Resource Management (CRM) program provides guidance to airlines on all aspects of developing, implementing, reinforcing and assessing team performance.
"Team training prepares people to work efficiently and effectively as members of a group," the NTSB explains. "CRM in commercial aviation seeks to reduce human errors in the cockpit by improving interpersonal communications, leadership skills and human decision-making."
As part of CRM, pilots, flight attendants and dispatchers learn to function as members of teams, not just as a collection of technically competent individuals, and behave in ways that foster crew effectiveness.
The U.S. Coast Guard requires licensed mariners to participate in similar Bridge Resource Management (BRM) training, and the Federal Railroad Administration is funding research with the aim of developing equivalent programmes for rail crews.
Creating a safety culture
The NTSB has pushed operators in the industries that it regulates to develop a safety culture, which the board defines as "doing the right things, even when no one is watching."
The NTSB's Roadmap to a Safety Culture begins with demonstrating management commitment, ensuring the standardisation and compliance with operating procedures, and promoting an open culture in which deviations can be reported without fear of retaliation.
In the airline industry, failure to develop adequate standard operating procedures, failure to adhere to them, and instances where the flight crew deliberately disregard them are a consistent theme in the cause of accidents. Crew members who intentionally deviate from procedures made three times more errors, the NTSB found.
Failure to adhere to procedures is the single most common cause of plane crashes, according to aircraft-maker Boeing, far ahead of equipment failure or maintenance and inspection problems, and is a contributory factor in 30 per cent to 50 per cent of all hull losses.
Regulators encourage airlines to have hotlines for staff to report non-compliance in confidence, with an assurance they will be acted upon, information will be kept confidential and that they will not be punished or ridiculed for reporting deviations from operating procedures.
Deviant cultures can exist in the military too. Earlier this year, the U.S. Air Force stripped 17 officers of their authority to control and if necessary launch intercontinental ballistic missiles from the Minot Air Force Base in North Dakota, after an inspection faulted poor compliance with operating procedures and "rot" within the unit.
In a recent presentation, the NTSB noted that a safety culture is triggered by management at the top, but measured by the day-to-day performance of employees at the bottom of the corporate pyramid.
Safety culture is a process of striving for continuous improvement, not an outcome, the NTSB warns. "If your organization is convinced it has a good safety culture, you are almost certainly mistaken."
Procedures not technologies
All industrial processes and modes of transportation involve some level of risk, but the evidence suggests that if proper procedures are adhered to, these risks can be substantially reduced to levels most people find acceptable.
Every day, the transport industry moves millions of passengers and hundreds of millions of tonnes of cargo most of them without accident. Refineries, chemical plants, power stations and manufacturing establish handle huge quantities of flammable, explosive and highly corrosive products, often at high temperatures and under intense pressure.
All of these processes are in some sense inherently unsafe. But the evidence strongly suggests that if proper protocols are followed, the chances of an incident are vanishingly small. The risk of death or injury to workers or local residents is very low, certainly compared with the risk of being injured or killed in a motor vehicle accident or falling from a height.
In the wake of a major disaster, the first instinct is to blame the technology, but the cause is more often poor handling practices. If we want to improve safety, the priority is normally better training and operational controls, rather than switching technology.
