Posts Tagged ‘human error’
Wednesday, August 20, 2014 @ 02:08 PM gHale
Human error was the cause of a 5,000 gallon diesel fuel spill in the Ohio River right outside Cincinnati, OH, Monday.
There is, however, “excellent improvement” in water conditions Wednesday after cleanup crews recovered about 1,000 gallons of the 5,000 gallons of fuel that spilled into the river late Monday during a routine transfer at Duke Energy’s W.C. Beckjord Station in New Richmond, about 20 miles upstream of Cincinnati.
Recovery of the fuel might be complete by the end of the day, and shoreline cleanup will begin after that. Wildlife damage has been minimal, said officials at the Environmental Protection Agency (EPA).
Drinking water is safe, Cincinnati Mayor John Cranley said Tuesday after the diesel spill.
“Our drinking water is safe and will be safe,” Cranley said at a press conference, before taking a large gulp from a glass of water at his podium.
EPA and Coast Guard officials estimated about 5,000 gallons of the fuel spilled during
The spill occurred at 11:15 p.m. Monday, according to Duke Energy officials. Crews were able to stop the release by 11:30 p.m.
“The diesel fuel, which is a reddish color, appears to be along the Ohio side of the river in small pockets,” EPA On-Scene Coordinator Steven Renninger said. “The biggest pocket is several miles downstream.”
Tony Parrott, head of Greater Cincinnati Water Works, said the department learned of the spill just after midnight.
Parrott said crews shut down the Ohio River intakes quickly so the spill did not make it into the Water Works. He said the fuel reached Water Works at about 7 a.m.
Water Works leaders said reserves were near capacity when intakes were closed, meaning Water Works can operate this way for an extended period of time. Crews said they expect to keep intakes closed for as long as it takes for the spill to completely pass.
“Even if we hadn’t shut down the intakes, our treatment methods would clean the water and would be safe for drinking,” Cranley said. “However, out of an abundance of caution… we decided to close our intake valves until we are confident.”
Parrott said the agency does have the capacity to feed active carbon into the water supply if they must open the intakes before the spill passes.
The Clermont County Water Resources Department also implemented a contingency plan in response to the spill. Officials there said they shut down wells in the Pierce-Union-Batavia area with the most potential to have an issue.
Lyle Bloom, director of Utilities at the department, said the agency is operating only wells that have no potential for the spill to have an impact on them.
“This is a precautionary measure,” Bloom said. “We do not expect the well field to be impacted by the diesel spill. The diesel fuel will remain along the water surface and should not adversely impact the aquifer.”
Duke Energy spokesperson Sally Thelen said Duke is looking at this spill as a result of “human error” and not a mechanical failure.
“It wasn’t as if something failed,” Thelen said. “We had alarms that did sound that did notify us that we did have an issue with an overflow situation. The alarms worked.”
According to Hamilton County Emergency Management Agency (HCEMA) officials, the spill happened when an open valve caused a secondary containment unit to lose fuel.
The process of moving fuel from two large tanks to smaller tanks is an everyday process, Thelen said. The tanks and valves are land-based and above ground. She said the diesel spilled down a hill and into the riverbank for about 15 minutes before workers stopped it.
The plant, which is set to close down Jan. 1, 2015, has the capacity to put out 1,124 megawatts of power and has six coal/steam units. It began operating in June 1952.
Monday, December 16, 2013 @ 03:12 PM gHale
By Ellen Fussell Policastro
In dangerous operations, especially the offshore oil industry, safety has to remain top priority.
Yet safety incidents continue in spite of highly experienced operators’ efforts. More often than a lack of training, a lack of judgment places these operators in the line of fire. One explanation for operators’ cloudy judgment is they may have reached a saturation point with automation systems and technology overload. That coupled with environmental stress leads to inevitable failure and increases the potential for disaster.
Luis Duran, product marketing manager for ABB control technologies in Houston, and Gregory Hale, founder of ISSSource.com, discussed how corporate culture, combined with technology, can help reduce the impact of safety incidents offshore in an ABB-sponsored OE Expert Access, “Empower Offshore Operations with Human-Centered Technology.”
The numbers clearly reveal where the offshore environment fits into the statistics of incidents throughout the industry. The U.S. Bureau of Labor Statistics reports an occurrence of 2.3 incidents in 2011 due to injury and illness per 100 oil and gas workers. The U.S. offshore industry experienced an even lower rate of 0.8 incidents per 100 full-time workers. That compares with 3.5 incidents per 100 workers for the entire private sector. While the last few decades have seen a noticeable drop in process safety incidents, the last five years have seen a plateau. So while the numbers haven’t gone up, they haven’t gone down either. And those that do occur are more severe and costly. The industry has experienced an annual loss of $20 billion due to safety incidents, and 80 percent of these incidents were preventable.
Basic automated actions are reliable in one in several hundred thousand occurrences. And certified safety systems bring up the reliability levels to nearly one failure in one million occurrences. But when it comes to manual actions – the human element – reliability drops dramatically to 1 in 100 occurrences or less, depending on environmental conditions, such as mental stress, during an abnormal event. Root-cause analysis indicates these human errors persist in spite of training and behavior enforcement.
Poor Communication Means Human Failures
Lack of communication is one of the top reasons for human failures offshore. A glaring example of communication failures is that of the 1988 Piper Alpha rig disaster in the North Sea. Alarm handling played a role as well. With so many alarms going off, operations had a hard time deciding which were important and which ones could wait. Setting up these alarms to react in real time will help operators make a decision. But a culture of safety needs to exist first – an environment in which people understand and work toward a common goal to ensure a safe operating system.
Corporate culture is vital to ensuring a strong safety environment. “You can’t just say, ‘Let’s be safe,’ and that’s the end of it,” Hale said. “There needs to be a framework that comes from the top and says, ‘Safety is Job 1.’ Second you have to understand the risk and be clear in your assessment of where that risk lies. Also, organizations with vigorous safety cultures are in a more secure position to avoid accidents and are better prepared when an incident happens.”
“Training is not necessarily a technology topic, but it is a key issue in the industry today,” Duran said. “We have a group of experts in industry getting ready to retire. So we have to capture their expertise. Technology is constantly changing as well. But we can use technology to assist the workforce. We can help train operators to work more effectively in difficult situations,” he said.
Hale likened safety training to the regimen airline pilots go through on a frequent basis.
“With pilot training, they have a chance to deal with every kind of potential disaster out there,” Hale said. “In the normal world they will probably not encounter these disasters, but if they do, they’re ready to go.”
Technology Can Help
While a strong corporate culture of safety is critical, technology can also come into play to assist operations in ensuring a safe environment. Stressors, such as organizational change and management transition, fatigue from shiftwork and overtime, and ineffective communication can contribute to poor decisions, Duran said. Technology can step in with efforts in improved ergonomics, display design, and training simulators to assist operations, Duran said.
In the process sector offshore, the design of the system as it relates to the operator is an important part of the safety design. It’s important to keep the operator’s attention, as he’s supervising the system and not controlling it, Duran said. “We can improve comfort and provide information that’s easy to understand. Simple moves, such as changing the operator’s physical environment and changing the lighting of the control room, can provide an environment in which the operator can be more alert and able to handle stress during abnormal conditions in the plant,” he said.
Graphic design is another area Duran’s team can improve. “The majority of today’s process automation assemblies rely on HMI,” he said. “Graphic design is a critical element to respond to a system design. In the late 1980s, it was common to see graphics that looked like a piping integration diagram. Inconsistent use of colors and graphic design practices were key contributors to confusion in the control room.” High performance HMI can improve operator situational awareness with consistent use of color, regardless of the information source. Within the same graphic, the operator can visualize the context with a multitude of subsystems.
In a similar situation, “when so many alarms are going off – what we call crying babies – that can lead to an urgent situation,” Duran said. “You soon have a whole bunch of crying babies that need attention all at once. That’s where an effective graphic design for alarm panels can help.”
“With traditional HMI you can get good results. But why settle for good when you can get better results with high performance HMI? You can see an increase with a five-fold hike in abnormal situations before alarms occur. You’ll see a 96 percent rate of handling abnormal situations with high performance HMI,” Hale said.
Using high performance HMI doesn’t have to be complicated. But to keep it simple, it’s essential to adhere to maintenance practices. An asset integrity management system can help deliver proactive maintenance strategies. One proactive strategy is to figure out possible causes for abnormal events. This type of strategy allows for easier communication between maintenance and operators, Hale said. “As we know, greater collaboration allows for stronger communication, which begs the question about integrated safety and control.”
Integrated Safety Systems
With the right design, integrated control and safety systems (ICSS) can provide a consistent technology environment for engineers and operators. “The whole premise of an integrated system is to give the operator access to information and decision-making, regardless of where information is coming from,” Duran said. “When information comes from process displays, especially in cases of abnormal conditions, ICSS can make it easier to exploit an area in the system where the operator needs to take timely action,” Duran said.
The key is to design all this technology around the human, Duran said. “The operators are running the system, but the technology can help with lower engineering and lifecycle costs as well as providing access and event management.” The technology works in tandem with humans because it requires engineers to think through the process. “So the designer has to ensure a process control system does not affect the safety of the procedure. He needs to introduce management of change consistently across the board in the system. This relies on the competence of the people using the system,” he said. And that’s where integrated control and safety systems come in – “to provide technology to help eliminate the consequences of human error.”
The bottom line is safety is not a job for one person, Hale said. Companies that employ best-in-case safety practices achieve 90 percent overall equipment effectiveness. But in order to make these changes in culture, the impetus has to come from executive leadership. And that leadership needs to establish a proactive strategy. Everyone has to understand the strategy and practice it on a daily basis.
“When there’s a small incident or near miss, there should be a root-cause analysis that’s a teaching point. Make sure your hazop is up to date; things change over time. You have to update your layers of protection analysis and your alarm strategy,” he said.
Today’s displays can give the operator enough information so he can take action to prevent problems from escalating. The key is consistency in the working environment, which can add to collaboration, so operators can respond to the unexpected when automation cannot, Hale said.
“We can design everything we know into an automated system,” Hale said. “The issues are how to deal with those things we don’t expect to happen. That’s where the operator is king. His ability to respond to those conditions will drive the difference between the incident and the recovery of the abnormal condition. That’s where these technologies can support the operator, so he can understand issues and take appropriate action to avoid a potential incident.”
Ellen Fussell Policastro is a freelance writer in Raleigh, NC. Her email is firstname.lastname@example.org.
Wednesday, October 9, 2013 @ 04:10 PM gHale
Human error mishaps are just multiplying as six workers at Japan’s crippled nuclear power plant in Fukushima ended up accidentally splashed with highly radioactive water, the plant operator said Wednesday.
The workers removed the wrong pipe from equipment at the plant, sending toxic water spilling onto them and the entire floor of the facility housing a set of three units designed for primary, partial water treatment, said Yoshimi Hitosugi, the spokesman for Fukushima Dai-ichi nuclear plant operator Tokyo Electric Power Co. (Tepco).
The workers, who were wearing face masks with filters, protective hazmat suits and raingear, suffered minor exposure, but the incident is still under investigation, Hitosugi said. The six were part of an 11-member team; the remaining five did not get hit with the water, he said. The workers also reattached the pipe.
The accident is the latest in a spate of leaks and other problems caused by human error that have added to public criticism of Tepco’s handling of the crisis at Dai-ichi, which is still in precarious condition since its triple meltdowns following the March 2011 earthquake and tsunami.
In order to keep the melted reactors cool, they must be continuously doused with water, which then becomes contaminated with radiation and then needs to pumped out and stored in tanks at the site.
Last week, workers overfilled a storage tank without fully checking water levels, causing a leak, possibly to the sea.
In August, the utility reported a 300-ton leak from another storage tank. That came after the utility and the government acknowledged that contaminated groundwater was seeping into ocean at a rate of 300 tons a day for some time.
Japan’s Nuclear Regulation Authority Chairman Shunichi Tanaka said repeated mishaps could be a sign of the harsh work environment.
“Careless mistakes are often linked to (declining) morale,” Tanaka said at a news conference. “People usually don’t make silly, careless mistakes when working in positive environment and motivated. The lack of it, I think, may be related to the recent problems.”
Tuesday, August 6, 2013 @ 07:08 PM gHale
Human error and equipment malfunction are the possible causes of a series of explosions that spawned a 20-by-20 foot fireball at a central Florida propane plant that left eight injured.
Tavares, FL, Fire Chief Richard Keith identified the possible causes Tuesday, but was leaning toward ruling out sabotage.
After two days officials in Tavares said they capped all of the propane leaks after a series of explosions last Monday night at the gas plant. As of Wednesday morning, authorities were burning off any propane left in the area.
The federal Occupational Safety and Health Administration launched an investigation into the Monday night incident at the Blue Rhino propane plant in the town northwest of Orlando, as did the Florida State Fire Marshal’s office.
No one died, but eight workers ended up injured, four critically.
Those injured include a worker hit by a car on a nearby road while fleeing the explosions.
A Blue Rhino employee who was at the plant when the explosions started told The Associated Press based on what a forklift operator told him, the explosion was likely caused by a “combination of human error and bad practices.”
Hoses designed to spray water on the large tanks didn’t go off because they had to manually activate, which required someone braving dangerous conditions.
Wednesday, June 12, 2013 @ 03:06 PM gHale
A propane tank explosion that sent two Ferrell gas employees to the hospital suffering from burns was the result of human error, officials said.
The two men were unloading propane tanks at a facility on Highway 32 in Coffee County, GA, remain hospitalized at an Augusta burn center following the accidental explosion Monday, officials said.
The employees were transferring propane from one truck to another when a leak occurred followed by the blast, State Fire Marshals said.
Residents from more than a mile away reported hearing the blast which officials say sent flames upwards of 40 feet. When firefighters arrived they found two trucks burning at the tank farm on Highway 32.
It took 45 minutes to put the truck fires out but a leak continued, which forced officials to shut down the highway for several hours. Tuesday morning crews removed the trucks from the scene.
While the two men remain hospitalized, a third employee injured in the blast went to the hospital and then released.
Fire investigators said propane blasts like this are rare and have only seen two such explosions in the past decade.
“Training really paid off,” said Douglas Fire Department Capt. Casey Wright. “They did an exceptional job firefighting. They were able to extinguish the fire, contain the fire, you know not within two or three minutes, but it was relatively quick.”
Because a propane leak continued, firefighters blocked off the highway and let the tank empty out on its own.
Coffee County Fire Chief Steve Carter said, “At this point, we have a small leak still on the delivery truck, and we’re diluting it down and going to allow it to bleed off until it’s empty.”
Friday, June 7, 2013 @ 03:06 PM gHale
The costs of data breaches is on the rise across the world, but in the U.S. companies have managed to continue bringing breach costs down, according to the eighth annual Cost of Data Breach Study.
The research found mistakes and human errors accounted for the bulk of all breaches, but malicious or criminal attacks cost businesses more when they are at the root of breaches, according to the report by the Ponemon Institute on behalf of Symantec.
Examining breach experiences from 277 organizations in nine countries, the study found the average global cost of data breaches reached $136 per compromised record, while in the U.S. the cost was $188 per compromised record.
“It’s still not chump change, but it definitely seems to be trending down in the U.S. with two years of downward movement,” said Dr. Larry Ponemon, chairman and founder of the Ponemon Institute.
The downward pressure could likely be attributable to more mature breach prevention and response practices. According to the study, the factors most likely to push breach costs down were instituting incident response plans, establishing a strong security posture, and appointing a CISO. Meanwhile, factors most likely to raise costs included third-party error, rushed breach notification, and lost or stolen devices.
Delving into the root causes of breaches studied in this report, Ponemon found 64 percent of breaches were the result of negligence or system failures, while 37 percent were the result of malicious insiders or criminal hackers. Ponemon said this should be a wake-up call.
“Everyone wants to hear about cyber attacks, and everyone wants to hear about cyber attacks and exfiltration of data by the Chinese or the proverbial bad guy, and those things are happening, but in our data, since the beginning of time the majority of cases are involving people problems or system failures,” he said. “Both are the result of negligence in a way.”
The catch is, malicious insider or criminal attacks cost more compared to human error and system woes.
An attack costs $157 per breached record compared to the $122 per record for breaches caused by system glitches and $117 for those caused by human error. Ponemon and Symantec feel the study results generally point to a greater need to address malicious and negligent insider threats within the enterprise.
“Our conversations with customers and our research does point to the insider threat continuing to be the bigger cause behind data loss and data breach events,” said Linda Park, product marketing manager for Symantec. “While there is an uptick in the malicious attacks, companies really are still focused on insider threats overall and making sure that employees are trained, aware, and that they have the right enforcement in place to make sure people are doing the right things.”
Click here to download the entire eighth annual “Cost of Data Breach Study.”
Wednesday, April 24, 2013 @ 07:04 AM gHale
By Gregory Hale
It is time to focus on human reliability.
“We have done a fantastic job in improving productivity, let’s try to prevent human error,” said Eddie Habibi, chief executive and founder of PAS, during his keynote address Tuesday at the PAS Technical Conference in Houston. “Human error is to human reliability as pump failure is to asset reliability.”
In these days of faster, better, more, the reliance on technology is becoming greater. The catch is technology is solid and it keeps getting better, but what about the people running the technology? Are they getting better?
There is a tremendous value automation systems bring as asset reliability helps bring in a solid return on investment, but now manufacturers need to also focus on human reliability and the human automation relationship. “We can’t take the human out,” Habibi said. “The operator is the most critical element in production.”
“Technology is available to improve human reliability,” said Larry Evans, founder and former chief executive at AspenTech and current advisor to PAS, during his keynote address Tuesday.
“These solutions allow engineers and operators to make good decisions and fewer bad ones,” Evans said.
We all know about what happens when there is a big disaster, but what about the small upsets that occur. They end up costing companies quite a bit and they don’t have to happen, Habibi said. In addition, the small upsets can lead to bigger problems. “The little things can add up to make huge things.”
To Habibi, the operator is the vital link in the automaton chain. That person needs to always have a good day because “when he has a bad day, everyone has a bad day.”
“The human is the weakest link. The role of the operator is to supervise the automation at the plant, making sure the automation system is doing the right things,” he said.
There are two types of human error, Habibi said: Unintentional and intentional.
Unintentional errors are the things we don’t mean to do. He gave the example of when he was in a plant one time and an operator changed a setting from 4.7 percent to 4.9 percent, but instead changed it to 49 percent. He said he never saw anyone run so fast as soon as he realized what he did. He ran out of the room to fix something and that averted a potential catastrophe.
Intentional errors, he said, come when the operator thinks he or she is smarter than the procedure. “They happen because we think we are doing the right thing,” he said. “The goal is to prevent human error, similar to physical asset reliability.”
To help alleviate human error, operators need to have situation awareness, which is having a collective understanding and cognizance of the environment around you. It is ongoing, constantly looking to do the right thing.
Habibi said there are three types of situation awareness:
• Physical environment
• Organizational culture
• Human automation relationship.
Physical factors include control room ergonomics, lighting, temperature comfort, along with traffic and noise issues.
Organizational culture deals with policies and procedures, shift schedule and reporting, work ethic and motivation, and training, knowledge and skills. “You have got to take care of people; give people the right tools,” he said.
The human automation relationship has different tiers. Tier one is direct operator interaction. Tier two is decision support systems. Tier three is automation asset management.
When it all comes down to it, as Evans said technology is going to continue to grow, but humans need to be able to keep up with it and make sure everything is running safely.
“Faster, better, more is great,” Habibi said. “But faster, better, more and safe is a hell of a lot better.”
Friday, November 2, 2012 @ 01:11 PM gHale
The shutdown of the No. 1 nuclear reactor at Korea’s Wolseong power station Monday was the result of human error, Korea Hydro and Nuclear Power (KHNP) officials said.
The plant operator made a mistake while manipulating a circuit breaker, causing a power shortage to some equipment which led to a malfunction in the cooling system, officials said.
“An internal probe revealed that a senior employee handed over his responsibilities to a junior with only two years of experience, who misunderstood his orders and operated the wrong switch,” a KHNP staffer said.
This is similar to the mistake that caused a malfunction at reactor No. 1 at the Gori nuclear power station that led to a blackout in February. At that time, a subcontracted worker with little experience failed to follow procedure and pushed the circuit breaker, causing the entire nuclear plant to lose power for 12 minutes.
“Human error is a more serious problem than faulty components and is the result of a lack of safety awareness,” said Prof. Hwang Il-soon at Seoul National University.
The latest incident appears to spell the death warrant for the Wolseong No. 1 reactor, which reaches the end of its 30-year life cycle Nov. 20.
A KHNP official said an inspection revealed serious damage to the inside of the reactor and a detailed investigation will take place after they take apart the reactor. The inspection will take more than a month.