Narrative:

We were cleared by lax center direct blh as filed. We were filed for the blh GEELA6 RNAV arrival. I had programmed the FMC for the runway 7L transition; since runway 7R was closed by NOTAM kphx A0865/13 201308/0500 201308/1400 and our arrival time was during that entire period. This NOTAM was the first contributing issue to this incident; as I interpreted it as meaning that the runway was going to be closed continuously for 3 days; rather than the usual night time closings of various phx runways. If the runway was going to be closed each night but open in the morning; I expected the NOTAM would say (as it usually does) runway 07R/25L clsd daily 0500-1400 wef 201308/0500 201308/1400. We had completed the arrival verification procedure before top of descent; and had been cleared direct to scole. After starting our descent to cross scole at FL250 we received the latest phx ATIS which showed arrivals on runway 7R. I attempted to load the 7R transition by selecting GEELA6/blh/ILS 7R and had to do it several times before the route looked satisfactory and discontinuities were removed. During each unsuccessful attempt the autopilot was transitioning back and forth from VNAV path to VNAV speed; with changing vtk error information and autothrottle activity. My difficulty was probably caused by our direct clearance to scole (bypassing blh) combined with my fatigue level; having been up all night. The net result was that my task loading went up very rapidly and my ability to figure out what I was doing wrong deteriorated. We did not have the chance to do a complete 2-pilot arrival verification procedure again after changing to the 7R transition because in very short order we got a frequency change to abq center; crossed scole properly at 280 knots and FL250; received and read back a clearance to descend via the geela 6 arrival 7R transition; set 4000 in the altitude window; and noted that we were high on the profile and the autopilot was correcting with a significant pitch down. My first officer (pilot monitoring) then asked; 'where did spink go? We need to cross at or above 240.' I didn't see spink on the HSI; although I thought we had not yet crossed it. We were rapidly approaching FL240. I selected altitude hold and saw that spink had apparently dropped out of the active route. I didn't think I had time to reload the route so I selected direct to spink; saw that it was about two miles ahead of us; and executed it. The aircraft was in a steady pull up and had leveled at FL236 as we passed spink. The execution occurred too close to spink; because as soon as we passed it the FMC drew a right circle back to spink; the aircraft rolled into a turn to the right; and the active route ahead disappeared up to geela. I went into heading sel then direct to snrra and on the way was able to successfully reload and verify the entire route with no further issues. Apparently I had not properly loaded the blh transition; so the route I had executed was from scole direct to geela. Distraction and sudden workload increase occurred just as we needed to complete another arrival verification. My first officer clearly saved the day with good monitoring; I'm sure we would have crossed spink significantly lower if he hadn't caught the mistake. ATC did not comment on our altitude or minor course divergence.fatigue; distraction; and a change to a different RNAV arrival transition caused me to fail to resolve FMC route change issues fast enough to prevent crossing an altitude restriction 400 ft too low. I could have avoided the issue altogether by not attempting to change the runway transition until much later (for example; approaching cagor); when our ground speed was much lower; we had more time; and ATC had made a final decision on our runway assignment. In retrospect; as soon as my task loading spiked I should have said; 'I have the aircraft and the radio; would you verify the route right now?' I think that would have divided the workload and managed the distractions and task loading more effectively. It might be worthwhile to check NOTAM kphx A0865/13 to see if it is coded correctly; since that's what set me up to begin with. Regarding the fatigue issue; it is my opinion that most if not all of us flying schedules that require us to stay up all night are always in a degraded condition. Operating this way is the norm; and the effects of the fatigue do not become obvious until things go wrong. This isn't going to change with the new flight/duty time far; solving the problem will require the company to stop building pairings that clearly violate safe scheduling practices that have been well researched and publicized for years. The schedule I fly is legal under the new fars. It is; however; as my doctor puts it; 'just crazy.'

Related reports:

Google
 

Original NASA ASRS Text

Title: A fatigued Captain described the complexity and confusion around a PHX RNAV runway transition change after the initial descent began which resulted an altitude constraint overshoot after the constraining waypoint dropped from the FMS during programming.

Narrative: We were cleared by LAX Center direct BLH as filed. We were filed for the BLH GEELA6 RNAV arrival. I had programmed the FMC for the Runway 7L transition; since Runway 7R was closed by NOTAM KPHX A0865/13 201308/0500 201308/1400 and our arrival time was during that entire period. This NOTAM was the first contributing issue to this incident; as I interpreted it as meaning that the Runway was going to be closed continuously for 3 days; rather than the usual night time closings of various PHX runways. If the runway was going to be closed each night but open in the morning; I expected the NOTAM would say (as it usually does) RWY 07R/25L CLSD DAILY 0500-1400 WEF 201308/0500 201308/1400. We had completed the arrival verification procedure before top of descent; and had been cleared direct to SCOLE. After starting our descent to cross SCOLE at FL250 we received the latest PHX ATIS which showed arrivals on Runway 7R. I attempted to load the 7R transition by selecting GEELA6/BLH/ILS 7R and had to do it several times before the route looked satisfactory and discontinuities were removed. During each unsuccessful attempt the autopilot was transitioning back and forth from VNAV PATH to VNAV SPD; with changing VTK ERROR information and autothrottle activity. My difficulty was probably caused by our direct clearance to SCOLE (bypassing BLH) combined with my fatigue level; having been up all night. The net result was that my task loading went up very rapidly and my ability to figure out what I was doing wrong deteriorated. We did not have the chance to do a complete 2-pilot arrival verification procedure again after changing to the 7R transition because in very short order we got a frequency change to ABQ Center; crossed SCOLE properly at 280 knots and FL250; received and read back a clearance to descend via the GEELA 6 arrival 7R transition; set 4000 in the altitude window; and noted that we were high on the profile and the autopilot was correcting with a significant pitch down. My First Officer (Pilot Monitoring) then asked; 'Where did SPINK go? We need to cross at or above 240.' I didn't see SPINK on the HSI; although I thought we had not yet crossed it. We were rapidly approaching FL240. I selected ALT HOLD and saw that SPINK had apparently dropped out of the active route. I didn't think I had time to reload the route so I selected direct to SPINK; saw that it was about two miles ahead of us; and executed it. The aircraft was in a steady pull up and had leveled at FL236 as we passed SPINK. The execution occurred too close to SPINK; because as soon as we passed it the FMC drew a right circle back to SPINK; the aircraft rolled into a turn to the right; and the active route ahead disappeared up to GEELA. I went into HDG SEL then direct to SNRRA and on the way was able to successfully reload and verify the entire route with no further issues. Apparently I had not properly loaded the BLH transition; so the route I had executed was from SCOLE direct to GEELA. Distraction and sudden workload increase occurred just as we needed to complete another arrival verification. My First Officer clearly saved the day with good monitoring; I'm sure we would have crossed SPINK significantly lower if he hadn't caught the mistake. ATC did not comment on our altitude or minor course divergence.Fatigue; distraction; and a change to a different RNAV arrival transition caused me to fail to resolve FMC route change issues fast enough to prevent crossing an altitude restriction 400 FT too low. I could have avoided the issue altogether by not attempting to change the runway transition until much later (for example; approaching CAGOR); when our ground speed was much lower; we had more time; and ATC had made a final decision on our runway assignment. In retrospect; as soon as my task loading spiked I should have said; 'I have the aircraft and the radio; would you verify the route right now?' I think that would have divided the workload and managed the distractions and task loading more effectively. It might be worthwhile to check NOTAM KPHX A0865/13 to see if it is coded correctly; since that's what set me up to begin with. Regarding the fatigue issue; it is my opinion that most if not all of us flying schedules that require us to stay up all night are always in a degraded condition. Operating this way is the norm; and the effects of the fatigue do not become obvious until things go wrong. This isn't going to change with the new flight/duty time FAR; solving the problem will require the company to stop building pairings that clearly violate safe scheduling practices that have been well researched and publicized for years. The schedule I fly is legal under the new FARs. It is; however; as my doctor puts it; 'just crazy.'

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.