Montreal - SAUDI ARABIAN AIRLINES announced the purchase of two WOMBAT-CS systems from ÆRO INNOVATION INC. The U$50,000 order includes two WOMBAT-CS consoles and the DuoWOMBAT-CS CRM Evaluation System. The Saudia Flight Training Academy will be in charge of translating the WOMBAT-CS instructions into Arabic. The Jeddah-based Academy will incorporate WOMBAT-CS into the actual prehiring pilot selection process. Mr. Jean LaRoche, President of Montreal-based ÆRO INNOVATION INC. reported, "It is the first time that WOMBAT has been installed in the Middle-East, but we expect to announce more sales from this area very shortly." Since the test itself is near culture-free, only the instructions will have to be translated with on-screen arabic characters. Printed material, normally distributed to the testees ahead of time, will also be translated into Arabic.
The Academy Director, Captain Al-Sowayigh first experienced WOMBAT-CS at the 1993 Ohio State University's Symposium on Aviation Psychology. After sending a delegation to Aero Innovation's Headquarters in October 1993, the Academy proceeded with a two-month validation study in Jeddah. Positive results further led to purchase negotiations. "We have been looking for this kind of testing throughout the world. I think we have finally found it. Now the goal is to learn how to use WOMBAT to maximize the results," stated Mr. Paul Koller, Flight Training Specialist in Jeddah.
Fairfax, Virginia - WOMBAT-CS v4.0 underwent experimental evaluation at GEORGE MASON UNIVERSITY, just outside Washington D.C., to determine if there were gender differences associated with WOMBAT's situational awareness assessment, and if so, if the differences are tied to past computer gaming experience. George Mason University has an active human factors engineering program that supports an aviation research group. This research, conducted by Ms. Deborah Bruce, a doctoral student working under Dr. Boehm-Davis, studied 32 subjects, 16 males and 16 females with an average age of 25, ranging from 18 to 48.
Results showed a smooth, consistent learning curve that reached a plateau about midway through the 90-minute testing period. When individuals were retested, the results showed no significant increase in their scoring rates relative to the final 30 minutes of the first administration. Most importantly, the results showed no evidence of a gender difference as is typically found in batteries of psychomotor tests, and no confounding from past computer experience. "These results are good news for WOMBAT-CS users worldwide as it indicates that WOMBAT-CS appears not to be susceptible to these common forms of bias," reported Ms. Bruce, a Human Factors Specialist for Systems, Requirements & Services Associates, Inc. (SRSA) working in Washington, DC.
How window design and pilot scanning habits work against you. by Dr. Stanley N. Roscoe V.-P. Research & Development
Of the four accidents, which resulted in 335 deaths, the one over Zagreb is the most notable because the Trident was leaving a seven-mile-long contrail against the blue sky. For almost three minutes, the leading end of its white contrail was on a stationary collision bearing of 30.7° to the left of the DC-9's forward flight path. In the earlier midairs (Urbana, Ohio, 1967; Fairland, Indiana, 1970; and Duarte, California, 1971), the visibility had been good, but there had been no contrail to attract a pilot's attention. Is it possible that the crew of the DC-9 could not see the other aircraft? Unlikely as that may seem, our research shows that the Trident coming from the left, with its seven-mile contrail, could have been completely obscured from the DC-9 pilot's view until the instant before the collision.
At 10:12:03 UTC, the copilot of the DC-9 told the middle sector controller that the plane had just passed through 31,000 feet on the way to Flight Level 350. Seven seconds later, the middle sector controller, who was controlling traffic from FL 240 to FL 310, instructed the DC-9 to "squawk Standby" on its transponder and to switch to the upper sector controller's radio frequency. When the transponder is in the standby mode, the controller's radar set does not receive and display the code altitude of the aircraft. But at the time the middle sector controller instructed the DC-9 to switch to the upper sector, the upper sector controller was busy talking to three other aircraft. He didn't make contact with the DC-9 until two minutes later. After asking the DC-9 for its present level, the controller said to maintain that level "because you have an aircraft in front of you (unintelligible word) 335 from left to right." The DC-9 responded, "Okay, we remain on exactly 330." One second after this last transmission, the planes collided. Neither pilot made any attempt to avoid the other aircraft. The handoff of the DC-9 from the middle to the upper sector was unquestionably not properly coordinated. The DC-9 had been cleared onto a precise collision course with the Trident. And the Zagreb controller was sentenced to seven years in prison for his part in the tragedy. However, a pilot is legally responsible for seeing and avoiding other traffic when conditions permit, even under instrument flight rules.
As the DC-9 approached the Zagreb VOR, the situation called for extreme care. The crew was aware of the local traffic congestion because the upper sector controller was so busy with other aircraft that they couldn't contact him for two minutes after entering his airspace. Furthermore, they knew they were climbing through flight levels that had been denied them because of other traffic. And they knew that the upper sector controller had no way of knowing their altitude because they'd set their transponder to standby. The Trident crew had been flying across Europe at a constant altitude for more than an hour and reasonably expected controllers to continue to protect that altitude from other aircraft. The DC-9 was more difficult to see because, according to the crew of a Luftansa B-737 about 19 km behind the Trident at FL 290, it wasn't leaving a contrail. Finally, the Trident was flying into the sun, although at a relatively high angle. The DC-9 crew had the sun at their backs.
In view of the combined circumstances-the Trident's seven-mile contrail, the clear visibility, and the traffic situation-the crew of the DC-9 should have seen the collision coming in time to avoid it. Why didn't they? As the accompanying diagram shows, the DC-9 has two identical windshield posts separating the pilot's and copilot's windshields from their side windows. From the pilot's and copilot's reference eye positions, these two posts measure 112 mm in projected width. That's not the actual measurement of the posts, but the width as it appears to the pilot and copilot from their normal position. The DC-9 also has two identical posts separating the 43 X 53 cm center windshield from the pilot's and copilot's windshields. From the copilot's reference eye position, the left post measures 82 mm, projected width.
With my associate J.C. Hull, a behavioral researcher who's worked at New Mexico State University and ILLIANA Aviation Sciences Limited, I performed a series of rooftop and laboratory experiments to see what effect such obstructions actually have on the pilot's ability to see what's going on outside. In one experiment, Hull and I projected simulated contrails on a textureless background to the right or left of a window post 30 cm from the subjects' eye. Their ability to detect the contrails was seriously reduced when we increased the post width from 63 mm (the average distance between the eyes and the width recommended in the U.S. Code of Federal Regulations) to 117 mm (approximate width of DC-9 posts). As the contrails extended farther and farther into monocular view from either side of the central sector of binocular obstruction, subjects were more likely to see them. Here's how the posts affected subjects' vision:
In our experiment, observers looked straight ahead and attempted to detect contrails briefly visible on either side of a post. But although pilots tend to focus on the post when they're flying in an empty sky, they almost never look directly at a window post when they're scanning the sky for other traffic. When the eyes look to one side of a window post, objects near the same side appear in the periphery of one eye only, the eye on the same side. Objects on the other side of the post appear only in the periphery of the other eye at an even greater angular distance -in the case of a 117 mm post, more than 20 degrees. So by looking to one side of an oversized post, a pilot is more likely to see a contrail on the same side, but less likely to see one on the other side. But the losses on the opposite side do not offset the gains on the same side. So the net improvement is substantial. Evidently, a pilot's normal scanning habit of fixating no closer than about 15 degrees from a window post is most effective. Not looking at a post should also make it easier to focus at a greater distance and thereby improve the clarity of outside vision.
Our experiments seem to warrant the following conclusions:
Few pilots are aware of the danger caused by wide window posts a few centimeters from the eyes. And neither the aircraft manufacturers nor the FAA has brought this danger to our attention. No airline, to my knowledge, specifically trains pilots to look around posts.
Pilots using the swivel & bob technique in a Boeing 747
Pilots generally pride themselves on keeping the head on a swivel, never realizing that this is not sufficient. As the word gets around, we can expect to see more and more heads bobbing fore and aft and from side to side, particularly as the airplane approaches VORs.
A binocular camera duplicates what you see with your eyes. The FAA has binocular photos of many general aviation aircraft cockpits, with all reference points clearly marked so you can readily see how much vision you have in any direction. You can get a print of the cockpit of your aircraft from the FAA Technical Center, Atlantic City, NJ 08405 USA. Once you know where you can't see, you can learn to compensate-to swivel and bob and do whatever you must to see the whole sky.
|| Home page || AeroNews Index ||