Narrative:

We were direct to arg descend via the fnchr RNAV STAR; landing north transition. First officer planned for that and set up the FMS to cross fnchr between 10;000 and 16;000 based on the north transition. At fnchr the new controller gave us runway 27. I loaded the [runway] 27 approach and inserted it glancing at the arrival seeing the transition 097 to jamla expecting that to load when setting up for the [runway] 27 approach. The FMS dumped and autopilot kicked off when I inserted. I told the first officer to fly the 097 to jamla while I entered it direct into the box. ATC queried if we were off the arrival. I advised that we lost the FMS and were proceeding outbound from jamla on the 095. They gave us a heading and descent to 6;000. Radar vectors to [runway] 27 visual were normal from that point on. First officer and I discussed the issue in debrief both believing that it was a late change to a landing west transition that caused the problem by changing the transition too close to fnchr causing the dump and ensuing confusion. We also now thought how could they do that to us and expect us to make the restrictions at jamla. I had just finished three days of west coast out and backs. I had mentioned to the first officer that I was very tired from the week and was ready for some time off. Enroute we discussed how the long legs at altitude are fatiguing and combined with the late nights and extreme heat it accumulates. I never felt; however; that I was 'not in the game' and thought I was well prepared for the arrival. I was convinced at the time that we had been switched to a landing west transition. I had not flown the fnchr before and we had briefed the landing north transition and had not reviewed the landing south procedure since it was not an option. It was not until today after sleeping and reviewing the situation that I discovered that the reason the FMS dumped is because that the fnchr has no landing west transition for [runway] 27; that the east bound depiction is the landing south transition. In the confusion of the last minute runway change and the FMS dump in trying to build what I thought was what ATC wanted I sent us on the wrong transition. I believe if we had been advised earlier prior to the perceived decision point of fnchr or later after that point there would not have been an issue. The fact that when we were given the change and the FMS dumped it was at what appeared to be the decision point for the south turn or the east turn which added to the task saturation and compression of decision processing time. In addition; for some reason I felt that the change was given at that time to modify the arrival procedure. I don't know why this was in my mind. I do know that the controller did not say 'continue via the landing north transition expect vectors for runway 27'; clearly that would have not left any room for interpretation. I interpreted the runway change call from approach as an action call combined with a quick look at the arrival to see what to expect and seeing the east bound transition I thought that was what they wanted. I feel that fatigue was a factor although that night I would have disagreed. I say this because even after the event when taking time to evaluate and debrief I was unable to process the situation effectively and understand what really happened. It was only after a night sleep that it hit me that my understanding of the event was completely wrong; including the debrief. I have subsequently contacted the first officer and accomplished a corrective debrief which identifies the reality of the mistakes and misunderstandings. The one thing that would have solved the issue from becoming a deviation event was communication. If I had immediately called approach and advised them of the FMS loss; as I should have; they would have provided an immediate vector and provided proper clearance until I could either rejoin the arrival where they wanted or break off for vectors as theygave us. This; I think; is where fatigue slows judgmental responses; and task saturation forces preoccupation and tunnel vision. If I was the pilot flying I think I may have taken the radios and focused on flying. As pilot not flying I was trying to solve the problem and did not back out to evaluate first. This idea to back out to evaluate would have prevented me from immediately going into the FMS to change the runway based on a perceived immediacy of required action due to the impending turn/decision point. I made a perceived ATC problem mine. If ATC gives me a change so late that I can't reasonably comply I should not accept the responsibility and advise ATC of the best action I can take and let them adjust to my limitations not the other way around. The next solution and one that would be preventive; would have been to ask myself questions as to what ifs? What if we get [runway] 27? How does that affect the arrival? Is there a transition? In addition the debrief included guidance for the first officer to always ensure he has a defined legal route. When his route dumped since I was distracted he should have either had me coordinate or directly coordinated with ATC. This in no way abrogates my responsibility for that duty as captain but just that with my failure if fell to him. Fly the jet; stay safe; stay legal; debrief it later.

Google
 

Original NASA ASRS Text

Title: An A300 flight crew descending via the FNCHR RNAV STAR to MEM; landing north transition; was advised to expect Runway 27 just prior to passing FNCHR. The pilot not flying; believing the 'landing south' transition would need to be programmed ASAP to provide guidance for the pilot flying; quickly programmed Runway 27--expecting the transition to program simultaneously--but; when activated; the FMS dumped all data beyond FNCHR and the autopilot kicked off. When they attempted to manually program and fly direct to JAMLA ATC noted their deviation and provided vectors for Runway 27.

Narrative: We were direct to ARG descend via the FNCHR RNAV STAR; landing north transition. First Officer planned for that and set up the FMS to cross FNCHR between 10;000 and 16;000 based on the north transition. At FNCHR the new Controller gave us Runway 27. I loaded the [Runway] 27 approach and inserted it glancing at the Arrival seeing the transition 097 to JAMLA expecting that to load when setting up for the [Runway] 27 approach. The FMS dumped and autopilot kicked off when I inserted. I told the First Officer to fly the 097 to JAMLA while I entered it direct into the box. ATC queried if we were off the arrival. I advised that we lost the FMS and were proceeding outbound from JAMLA on the 095. They gave us a heading and deScent to 6;000. Radar vectors to [Runway] 27 visual were normal from that point on. First Officer and I discussed the issue in debrief both believing that it was a late change to a landing west transition that caused the problem by changing the transition too close to FNCHR causing the dump and ensuing confusion. We also now thought how could they do that to us and expect us to make the restrictions at JAMLA. I had just finished three days of West Coast out and backs. I had mentioned to the First Officer that I was very tired from the week and was ready for some time off. Enroute we discussed how the long legs at altitude are fatiguing and combined with the late nights and extreme heat it accumulates. I never felt; however; that I was 'not in the game' and thought I was well prepared for the arrival. I was convinced at the time that we had been switched to a landing west transition. I had not flown the FNCHR before and we had briefed the landing north transition and had not reviewed the landing south procedure since it was not an option. It was not until today after sleeping and reviewing the situation that I discovered that the reason the FMS dumped is because that the FNCHR has no landing west transition for [Runway] 27; that the east bound depiction is the landing south transition. In the confusion of the last minute runway change and the FMS dump in trying to build what I thought was what ATC wanted I sent us on the wrong transition. I believe if we had been advised earlier prior to the perceived decision point of FNCHR or later after that point there would not have been an issue. The fact that when we were given the change and the FMS dumped it was at what appeared to be the decision point for the south turn or the east turn which added to the task saturation and compression of decision processing time. In addition; for some reason I felt that the change was given at that time to modify the arrival procedure. I don't know why this was in my mind. I do know that the Controller did not say 'continue via the landing north transition expect vectors for Runway 27'; clearly that would have not left any room for interpretation. I interpreted the runway change call from Approach as an action call combined with a quick look at the arrival to see what to expect and seeing the east bound transition I thought that was what they wanted. I feel that fatigue was a factor although that night I would have disagreed. I say this because even after the event when taking time to evaluate and debrief I was unable to process the situation effectively and understand what really happened. It was only after a night sleep that it hit me that my understanding of the event was completely wrong; including the debrief. I have subsequently contacted the First Officer and accomplished a corrective debrief which identifies the reality of the mistakes and misunderstandings. The one thing that would have solved the issue from becoming a deviation event was communication. If I had immediately called Approach and advised them of the FMS loss; as I should have; they would have provided an immediate vector and provided proper clearance until I could either rejoin the arrival where they wanted or break off for vectors as theygave us. This; I think; is where fatigue slows judgmental responses; and task saturation forces preoccupation and tunnel vision. If I was the pilot flying I think I may have taken the radios and focused on flying. As pilot not flying I was trying to solve the problem and did not back out to evaluate first. This idea to back out to evaluate would have prevented me from immediately going into the FMS to change the runway based on a perceived immediacy of required action due to the impending turn/decision point. I made a perceived ATC problem mine. If ATC gives me a change so late that I can't reasonably comply I should not accept the responsibility and advise ATC of the best action I can take and let them adjust to my limitations not the other way around. The next solution and one that would be preventive; would have been to ask myself questions as to what ifs? What if we get [Runway] 27? How does that affect the arrival? Is there a transition? In addition the debrief included guidance for the First Officer to always ensure he has a defined legal route. When his route dumped since I was distracted he should have either had me coordinate or directly coordinated with ATC. This in no way abrogates my responsibility for that duty as Captain but just that with my failure if fell to him. Fly the jet; stay safe; stay legal; debrief it later.

Data retrieved from NASA's ASRS site as of July 2013 and automatically converted to unabbreviated mixed upper/lowercase text. This report is for informational purposes with no guarantee of accuracy. See NASA's ASRS site for official report.