|
|
|
||||||||||||||||||||||||
|
|||||||||||||||||||||||||
By John Wiley February 12, 2004 The next time aviators gather to tell "war stories," it’s unlikely that one of the hair-raising yarns will involve taxiing. Compared with V1 engine failures, systems malfunctions and low visibility approaches, taxiing may well be the Rodney Dangerfield of aviators’ tasks; i.e., it gets little or no respect. Yet studies show that what seems to be a simple procedure can be anything but. Getting from the runway to the ramp may be easier than getting from the outer marker to the runway. But as "The Effects of Display Location and Dimensionality on Taxiway Navigation," a 1995 study by James Lasswell and Christopher Wickens of the University of Illinois at Urbana-Champaign’s Aviation Research Laboratory, points out, the flight is not over simply because you’ve managed a smooth
touchdown in Cat IIIB conditions. Noted the NASA-commissioned study, "A pilot who successfully completes a mission and lands in zero-visibility conditions will be faced with the daunting task of rolling off the runway and maneuvering to the terminal or pad with little or no visual cues."
Taxiing an aircraft can at times be likened to running a high-tech gauntlet. After all, while advancing in their awkward ground vehicles, air crews must listen for and comply with ground control directives issued over a frequently congested radio frequency. They also must correlate the outside world with the small paper chart depicting the airport layout. Add to this the fact that it can take up to two seconds for a pilot to refocus after shifting view from near to distant and back again. Moreover, at night there is the problem of having enough light in the cockpit to see the chart, while at the same time restricting internal light sources so the pilots can still see clearly objects outside the cockpit.
In the Draft for the Airport Surface Situational Awareness (RTCA Paper Number 199-03/SC186) we find another description. "Unlike inflight operations, taxiing is a 2-D task. However, the airport surface is a complex and very busy environment due to potential unfamiliarity, non-optimal weather and/or lighting conditions as well as numerous tasks required of the flight crew such as the following: gathering information including detection of hazards in the cockpit, gathering information including detection of hazards out of the window, identifying aircraft or other vehicle hazards, minimizing deviations from the centerline, achieving and maintaining best speed for given conditions, negotiating turns, communicating with crewmembers, company and ATC, completing checklists, maintaining global position awareness via some type of map and avoiding hazards (conflict with other aircraft or vehicles). All of these tasks can incur error during execution."
To help get a broader perspective on the taxiing challenge, I conducted a word search of the NASA Aviation Safety Reporting System (specifically, the NASA ASRS Incident Database, DOS version 98-4). The first search, simply for the word "taxi," produced 4,735 reports. To refine the search, I combined taxi with the words "wrong," "incorrect," "error" and "mistake," and then combined "wrong" with "taxiway" and came up with a total of 67 reports.
Reports included a crew that could not see the signs; another that received the clearance correctly but failed to make the correct turn; one that erred executing a revised clearance; and a first officer who copied the clearance correctly but gave the information to the captain incorrectly. In yet another report, the crew taxied to the wrong runway only to be told by ground control that they were the 12th airplane to misread the signs that month. CRM issues popped up in one report, which recounted the captain ignoring the F/O’s correct readback of the clearance and then taxiing to the wrong runway. Incidents happened at airports large and small with some facilities showing up more than twice. These included Detroit Wayne County (DTW), six incidents, Pittsburgh International (PIT), five incidents, Chicago O’Hare (ORD), five incidents, and St. Louis Lambert-International (STL), three incidents.
These reports were filed by pilots flying business jets, regional turboprops and jets, and airliners. Some were EFIS equipped, others not, and one aircraft had a head-up display. Meteorological conditions included day and night VMC and IMC. Flight crew experience ranged from hundreds of hours to tens of thousands of hours. The reports tell of wrong routes; wrong turns; misunderstood, incorrect and revised clearances; taxiing without reference to any chart; charts that failed to depict completed construction; and taxiway signs obscured by snow and others that were simply obfuscated. The picture that emerges from these 67 selected NASA incident reports is that getting from the gate to the runway and from the runway to the ramp can be a challenge.
What Pilots Say
In 1995, Dr. Anthony D. Andre, also of the University of Illinois at Urbana-Champaign, rode jumpseat on 35 airline flights. A few aircraft had three-man crews but the majority had two pilots. In the resulting study, "Information Requirements for Low Visibility Taxi Operations: What Pilots Say," that was done for NASA, Andre writes, "Before exploring the ways in which advanced technology can aid the pilot [and controllers] to increase efficiency of low visibility surface operations, it is important to understand the problems currently faced by pilots during taxi operations and the corresponding causes of these problems. Further, it is important to understand the pilots’ views and opinions about potential cockpit technology aids such as electronic taxi map displays as these preferences and attitudes will have a large influence on the ultimate acceptance and utility of any new technology."
Andre posed a simple question for the flight crews: "What problems do you currently face when taxiing your aircraft in low visibility conditions?" Crews responded, stating, "Low visibility taxiing is the hardest," that "Taxiing at ORD the first time in low visibility conditions is impossible," and that "Under low visibility conditions, I taxi at one-half to one-third the speed of high-visibility conditions." One pilot said conditions were once so bad he had to have a follow-me truck guide him, and another said conditions were so bad at SFO, it took one hour to taxi from the gate to the runway. Andre accurately noted that as weather goes down and at complex airports, pilots taxi slower, notably, ". . . sometimes with less-than-accurate awareness of where they are on the airport surface."
Crews at the time also complained about the apparent lack of standardization in taxiway designations. "All airports should be consistent within themselves and other airports. . . . Why is A sometimes next to K?" Another pilot pointed out that at one airport, weeds mostly cover the signs. Another asked why the taxiway designation cannot be painted on the taxiway "like they do on the runways?"
Communication, or more accurately, miscommunication, was cited as a major problem. Among pilot complaints were frequency congestion, cockpit distractions, inconsistent terminology, misinterpreted clearances, incomplete clearances, and confusing phraseology each adding to possible errors. Crews told Andre they disliked having the taxi route changed once they were under way since they felt that modifying a plan while taxiing increased the chance of error. Another crew said they wanted more information about where to taxi than when to taxi. Also, crews commented they did not like clearances such as, "Follow that guy."
Andre focused next on proposed electronic moving map (EMM) displays. As one might expect, comments ranged from enthusiastic anticipation to bah-humbug. Supporters wanted to see the diagrams north-oriented but also "track up." The EMM should be "useful in all conditions, not just low visibility," one commented, while another volunteered that "The most difficult thing is not knowing where you are and what you’re coming up to next; that would be tremendous information to have on the map." And, said another, "Of course I want this. It took me 30 years to learn O’Hare!"
However, a common concern among many of those interviewed about an EMM focused on the amount of time pilots would give such a device rather than looking out the window. Andre summarized his findings noting that low-tech options such as improved signage, better surface markings and lighting, as well as communication protocols, can substantially improve the taxiing process.
Three Required Components
In their 1995 study, Lasswell and Wickens identified three requirements for successful taxi navigation: local guidance, global awareness and route awareness. Local guidance is basically taxiing the airplane, that is, turning, advancing, gathering information from signs, and such. Global awareness is knowing the general layout of the airport, the whereabouts of the ramp or runways, and traffic. Route awareness is knowing the route to the objective as well as your current position. The study notes that omitting any one of these three components results in increased workload with potential for catastrophe.
One potential solution to the problem is represented by NASA’s highly sophisticated avionics suite, the Taxiway Navigation and Situation Awareness, or T-NASA, system. The agency developed the T-NASA through a process that included more than 300 commercial pilots participating in part task trainer simulations, high fidelity simulations and flight tests. The T-NASA suite consisted of an HUD and an EMM and a high-speed data link that put the taxi clearance in text form on the EMM.
Problems cropped up when just using the HUD or EMM independent of the other since the HUD alone gave local guidance but no global awareness and the EMM gave global awareness but no local guidance. Additionally, each device presented its own problems.
Pilots using HUDs can suffer from "attention tunneling," which occurs when the viewer looks at the HUD’s plate and information/symbols displayed on it rather than through the HUD. "Attention tunneling" can cause crews to miss important information such as another airplane blocking the taxiway. In one study, crews using an HUD in a simulation took longer to detect an airplane in their path than those without the HUD. Worse, some crews never even saw the aircraft and "collided."
The EMM studies showed that map orientation preference depended upon the task being undertaken. Track-up was preferred when used with reference to outside landmarks, while north-up was preferred for route planning. The T-NASA EMM provided both depictions as well as the ability to zoom in/out to four levels. The study found that crews tended to spend more time studying the EMM when weather conditions were good; that decreased as weather and conditions deteriorated. Also captains used the EMM differently from F/Os, with the former zooming in while the F/Os zoomed out.
The T-NASA suite helped eliminate taxiing errors almost completely. With the EMM crews taxied faster, caught and corrected errors faster and without assistance from ATC, their tracking improved, deviations decreased and situational awareness was better. Unfortunately for most of us, we will not have a T-NASA suite installed in our airplane anytime soon. So was all the effort for naught? Absolutely not. Read on.
T-NASA Simulations
In 1997, researchers D.R. Kelly and G.L. Adams examined questionnaire data gathered from 2,000 U.S. airline pilots about factors that could contribute to taxiing errors and identified five areas specific to navigation: (1) pilot’s unfamiliarity with airports; (2) inadequate airport navigation aids; (3) ATC-pilot miscommunications; (4) lack of standardized cockpit procedures; and (5) pilot fatigue and poor eating habits. These findings were presented at the 9th International Symposium of Aviation Psychology in Columbus, Ohio.
In 2001, Bechy L. Hooey, David C. Foyle and Anthony D. Andre used the Kelly and Adams study as their baseline for another NASA-sponsored study, "Design of Aircraft Displays for Low Visibility Taxi." The new study found that taxiing errors were neither random nor the result of pilot inattention. Rather, the problems cited included frequency congestion at large airports, confusing signage, extensive clearances, and distractions in the cockpit while taxiing.
The study showed again that taxiing, especially in low visibility conditions or at night, turned out to be a complex and challenging exercise. It also noted that while many cockpits have state-of-the-art glass, getting to and from the runway is still done most often using a compass and a paper chart.
The report noted that "taxi charts can be confusing, cluttered and difficult to read, promoting excessive head-down time. Further, pilots must translate information on the chart to an out-of-window view, which often requires mental rotation from the north-up chart to their actual heading. The difficulty of the taxi task is compounded further by the complexity of the navigational environment. Airport surfaces consist of a tangled network of taxiways and runways identified by signs and painted markers. As signs cannot be placed overhead, they are placed on grass and cement islands to the side. Navigation errors are often attributed to the necessarily awkward placement of the taxiway signs and complex taxiway geometry."
Between 1998 and 2000, Hooey and Foyle’s low-visibility study noted, there were 3,420 runway incursions with 48 percent of these incursions caused by pilots deviating from their clearance. The FAA subsequently held workshops to solicit suggestions on how to identify problems. Suggestions included "procedural and operational changes, and improvements to pavement markings and signage and in-cockpit technology."
As part of their research project, Hooey and Foyle conducted two full mission simulations using a full-motion simulator with an 180-degree field. The views depicted Chicago ORD, including accurate replications of the airport, taxiways, signs, structures, lights, markings and concourses. The simulator was equipped with the T-NASA avionics suite. Errors were broken into three groups: planning, decision and execution. Each error group had distinct contributing factors and as such required different solutions.
Sixteen two-man crews completed 18 taxi trials, six trials using only Jeppesen charts, six trials using only the EMM and six trials using the EMM and HUD. Half of the 16 crews completed their trials with conditions set for RVR 700 while the other half completed their runs in simulated night VFR conditions.
The second simulation series had 18 two-man crews complete nine scenarios with an RVR 1000. Three scenarios were done without the advanced technology, three were done using data-linked text clearances, and three were done using the EMM and HUD with data-linked text messages. The scenarios without the advanced technology included just a verbal clearance and the paper chart. The taxi clearance was issued after clearing the runway.
Taxiing anywhere not cleared or a deviation from centerline of more than 50 feet constituted a navigation error. Of the 150 trials using paper charts only, 17.3 percent (26) contained navigation errors. The most interesting point of this simulation was more errors occurred at night than in low visibility conditions, suggesting navigation errors were not limited to bad weather but ". . . are pervasive." It is these 150 current operation trials we want to focus on.
Errors occurred when crews improperly planned their taxi route but carried out the flawed plan without deviation, accounting for 23 percent (six) of the 26 navigation errors. Two main factors found in this group were miscommunications and a bias in expectations and confirmation.
Frequency congestion, lengthy clearances, multiple clearances and read-back errors all can combine to create potential planning errors. In one instance the F/O incorrectly read back the clearance to ATC and then gave the mistaken clearance to the captain who then executed a plan based on the flawed information. A second mistake occurred when the F/O read the clearance back correctly but gave the captain incorrect information. The captain planned his course on this errant information.
Expectation and bias caused the four other errors. Crews correctly copied and read back the clearance and initially enacted the correct plan. However, as they neared their gate, the clearance turned them away because an airplane was blocking the direct route. Nevertheless, the crew chose to ignore a clearance that seemed incorrect and erred when they deviated from their original plan.
Decision Errors
In 11 of the 26 errors, crews got the information correct and planned correctly but failed to execute their plan properly. Contributing factors in these errors were inadequate navigational awareness and "excessive operational demands." Of this set of errors, 55 percent occurred at the first decision point after exiting the runway. Examples of excessive operational demands include taxiing while the F/O was involved in other tasks such as changing radio frequencies, receiving clearances, writing down information, checking charts for routing and reading clearance to the captain.
Inadequate navigational awareness accounted for four of the 11 decision errors. Crews erred when they believed they were at a point other than their actual location. Again, a contributing factor was the F/O’s involvement in other tasks rather than monitoring and concentrating on taxiing. In seven of the 11 decision errors, the captain turned in the wrong direction, indicating a lack of knowledge or a misunderstanding of where they were and where they were going. In two of these seven incidents, the captain made the decision without involving the F/O. The F/O provided incorrect information in another two incidents, and in two more incidents, the captain and F/O contributed to the mistake. In the final error, the F/O tried to help with the correct information but the captain ignored the input.
Execution Errors
In another group of errors, crews got the clearance correct, identified their position correctly, but then erred in execution. This happened in 35 percent of the 26 errors. Factors contributing to these errors included complex taxiway geometry, confusing signs and "a sea of blue lights."
Crews committing these "execution errors" noted that taxiways changed names but not direction, or they encountered multiple intersecting taxiways and intersections where taxiways had turns in the same directions but at different angles. This complex geometry accounted for seven of nine execution errors (78 percent). Crews complained about the difficulty of deciding which centerline to follow and problems deciphering the Jeppesen charts.
While T-NASA or a derivative may help solve some of these problems, they will likely come at a steep price. Meanwhile, there are low-tech and low-cost solutions available.
Back to Basics
Following recent crew resource management evolutions, a major step for flight crews could be recognizing the potential for error or difficulty in taxiing, especially at night or in low visibility conditions. Underestimating this "threat," especially at airports such as LAX, ORD, DTW or CLE, is just asking for errors to occur.
The combination of aging eyes trying to decipher airport diagrams crammed with information in tiny fonts is another problem. As the eyes age, they not only require help focusing the image but also need more light just to see the chart at all. A problem, yes, but fixable. First, some study before going to the airplane can help. It is safe to say that more often than not, crews spend more time planning the en route portion of the flight than how to taxi. Study the airport diagram and have an idea of how to get to the runway. Second, take the chart and make a copy that is much larger and easier to read. Third, go online to www1.faa.gov/runwaysafety/naco.cfm and download a large copy of the airport diagram.
To preclude communications problems, both pilots should listen to the taxi clearance. This is not what happens in many cockpits whose captain is busy navigating out of the parking spot and watching the ground crew while the F/O is configuring the airplane and calling for the clearance. Slow down. Get the clearance, and make sure both crewmembers understand it and agree on it before releasing the brakes. If there is any question, the solution has to come from ATC. No in-cockpit answers permitted.
To prevent deviation and runway incursions, one airline insists on "No solo taxi." When taxiing, if duties such as weight and balance, company radio calls, problems in the cabin, checklist items or any other task take precedence over taxiing, the F/O should let the captain know. A call such as "I’m out of the loop" or "Heads down," lets the captain know the second set of eyes is unavailable. Also, if there is clearance that requires crossing a runway, cockpit workload is suspended until the airplane clears the runway. Again, any question of being cleared to cross should be posed to and answered by ATC, and not resolved among the cockpit’s inhabitants.
Prior to descent is a good time to complete your study of the arrival airport’s layout. You should know approximately how much runway you will need for landing and rollout. Once armed with that knowledge, you can estimate where you will be able to exit the runway. Knowing which taxiway you’ll likely use helps you plan the route from the runway to the ramp.
Some airports use standard taxi routes, but again, this is often printed in the small font and deciphering the route is better done before receiving the clearance. Of course, if you do not know where you are or where you are going, shelve the pride and ask for progressive. Again, it is easier and takes less time to ask for progressive than to fill out all the paperwork explaining the error. If necessary, stop the airplane until you can verify where you are, where you are going and how you are going to get there.
Absent a flight deck full of money, most of us are going to have to continue using a paper chart and compass to get to and from the runway. While the solutions offered seem quite basic, part of the problems in taxiing may stem from lack of respect for the hazards, the workload or the skills required for a task that can be anything but simple. B/CA
Reprinted from the February 2004 issue of Business & Commercial Aviation magazine
