Narrative:

I missed a crossing altitude restr by less than 2000 ft. This was due to entering the descent phase of flight in the FMGC and the aircraft system 'dropped' a previously entered crossing restr and I did not catch the omission until I was too close to the restr to comply. Flight continued normally to landing and block in. Human factors included placing a higher degree of trust than deserved in the FMGC prediction of leveloff system and not mentally calculating a 3:1 glide radio to back up the autoflt. I also had a 50 KT tailwind in the descent further complicating the issue. Supplemental information from acn 583737: the flight management computers on the A320 are programmed to selectively ignore pilot inputs. As a result the system can be extremely unforgiving of any loss of crew vigilance. A crossing restr of 17000 ft was pre-loaded into the flight management computers at ramms intersection. When 17000 ft was later set into the flight control unit altitude window, the flight management computers disregarded and deleted the crossing restr. We did not catch the deletion and erroneous descent guidance until it was too late to make the crossing restr. The flight management computers functioned precisely as designed. The computers intentionally delete a crossing restr if certain common conditions are met. Among others, this includes beginning descent more than 200 nautical air mi from destination. It is not clear why the computers were intentionally engineered to ignore pilot input in such cases.

Google
 

Original NASA ASRS Text

Title: A319 FLT CREW MISSES XING RESTR DURING STAR INTO DEN.

Narrative: I MISSED A XING ALT RESTR BY LESS THAN 2000 FT. THIS WAS DUE TO ENTERING THE DSCNT PHASE OF FLT IN THE FMGC AND THE ACFT SYS 'DROPPED' A PREVIOUSLY ENTERED XING RESTR AND I DID NOT CATCH THE OMISSION UNTIL I WAS TOO CLOSE TO THE RESTR TO COMPLY. FLT CONTINUED NORMALLY TO LNDG AND BLOCK IN. HUMAN FACTORS INCLUDED PLACING A HIGHER DEGREE OF TRUST THAN DESERVED IN THE FMGC PREDICTION OF LEVELOFF SYS AND NOT MENTALLY CALCULATING A 3:1 GLIDE RADIO TO BACK UP THE AUTOFLT. I ALSO HAD A 50 KT TAILWIND IN THE DSCNT FURTHER COMPLICATING THE ISSUE. SUPPLEMENTAL INFO FROM ACN 583737: THE FLT MGMNT COMPUTERS ON THE A320 ARE PROGRAMMED TO SELECTIVELY IGNORE PLT INPUTS. AS A RESULT THE SYS CAN BE EXTREMELY UNFORGIVING OF ANY LOSS OF CREW VIGILANCE. A XING RESTR OF 17000 FT WAS PRE-LOADED INTO THE FLT MGMNT COMPUTERS AT RAMMS INTXN. WHEN 17000 FT WAS LATER SET INTO THE FLT CTL UNIT ALT WINDOW, THE FLT MGMNT COMPUTERS DISREGARDED AND DELETED THE XING RESTR. WE DID NOT CATCH THE DELETION AND ERRONEOUS DSCNT GUIDANCE UNTIL IT WAS TOO LATE TO MAKE THE XING RESTR. THE FLT MGMNT COMPUTERS FUNCTIONED PRECISELY AS DESIGNED. THE COMPUTERS INTENTIONALLY DELETE A XING RESTR IF CERTAIN COMMON CONDITIONS ARE MET. AMONG OTHERS, THIS INCLUDES BEGINNING DSCNT MORE THAN 200 NAUTICAL AIR MI FROM DEST. IT IS NOT CLR WHY THE COMPUTERS WERE INTENTIONALLY ENGINEERED TO IGNORE PLT INPUT IN SUCH CASES.

Data retrieved from NASA's ASRS site as of July 2007 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.