Narrative:

We had preprogrammed the FMS with the civet 5 arrival to runway 25L; as was our normal expectation. Center had cleared us to cross gramm at FL190. The aircraft was on autoplt and descending to make the crossing. Approximately 7 mi from gramm; center handed us off to approach. I checked in with approach and he cleared us to descend via civet 5 to runway 24R. I acknowledged the clearance and the captain reprogrammed the FMS. When LNAV was selected; the aircraft began a right turn away from the desired course. I noted this immediately and announced it to the PF. He immediately switched to heading mode and turned the aircraft back to re-intercept the course. By this time we were past gramm and almost on top of rustt. When LNAV was selected the aircraft was apparently too close to rustt and it overshot the course. The PF then disconnected the autoplt and manually re-intercepted the course. After rustt the course changes by 22 degrees to the right. The PF was attempting to regain the course prior to civet. At this time I requested direct to skoll or decor to give a little more time/distance to regain the course. Approach cleared us to palac and descend to 8000 ft. He then asked if we had the field. We said we had the field in sight and he cleared us for the visual to runway 24R. I selected direct to palac in the FMS and the PF engaged the autoplt. The remainder of the approach was normal and uneventful. Basically; the PF did 2 small s-turns across the arrival course. I think that the factors that led to the course deviations were: 1) preprogramming an expected runway and not getting it. 2) center did not give us any indication of which runway to expect even though the arrival is runway dependent. 3) approach cleared us for a runway dependent arrival when we were almost over the first fix. 4) time spent reprogramming the FMS put the aircraft in a situation that caused the initial deviation. I have heard of this happening before and after thinking about possible solutions have come up with these: 1) program both runways; one in route #1; the other in route #2 for immediate use. 2) ask center for an anticipated runway assignment prior to the fix where we get handed off to approach. Doing these 2 things should help alleviate this problem in the future. Another contributing factor is that the fixes gramm; rustt; and civet are very close together. This makes it more difficult for the autoplt to re-intercept if it gets even a little off course. The solution to this part is to do what we did; get a vector further down the arrival path. Another possible contributing factor was the decreasing speed assignments during the arrival. Every time ATC changes something; the FMS gets reprogrammed and has to re-compute the paths.

Related reports:

Google
 

Original NASA ASRS Text

Title: B757 CREW IS TRIPPED UP BY LATE RWY CHANGE ON THE CIVET 5 INTO LAX.

Narrative: WE HAD PREPROGRAMMED THE FMS WITH THE CIVET 5 ARR TO RWY 25L; AS WAS OUR NORMAL EXPECTATION. CTR HAD CLRED US TO CROSS GRAMM AT FL190. THE ACFT WAS ON AUTOPLT AND DSNDING TO MAKE THE XING. APPROX 7 MI FROM GRAMM; CTR HANDED US OFF TO APCH. I CHKED IN WITH APCH AND HE CLRED US TO DSND VIA CIVET 5 TO RWY 24R. I ACKNOWLEDGED THE CLRNC AND THE CAPT REPROGRAMMED THE FMS. WHEN LNAV WAS SELECTED; THE ACFT BEGAN A R TURN AWAY FROM THE DESIRED COURSE. I NOTED THIS IMMEDIATELY AND ANNOUNCED IT TO THE PF. HE IMMEDIATELY SWITCHED TO HDG MODE AND TURNED THE ACFT BACK TO RE-INTERCEPT THE COURSE. BY THIS TIME WE WERE PAST GRAMM AND ALMOST ON TOP OF RUSTT. WHEN LNAV WAS SELECTED THE ACFT WAS APPARENTLY TOO CLOSE TO RUSTT AND IT OVERSHOT THE COURSE. THE PF THEN DISCONNECTED THE AUTOPLT AND MANUALLY RE-INTERCEPTED THE COURSE. AFTER RUSTT THE COURSE CHANGES BY 22 DEGS TO THE R. THE PF WAS ATTEMPTING TO REGAIN THE COURSE PRIOR TO CIVET. AT THIS TIME I REQUESTED DIRECT TO SKOLL OR DECOR TO GIVE A LITTLE MORE TIME/DISTANCE TO REGAIN THE COURSE. APCH CLRED US TO PALAC AND DSND TO 8000 FT. HE THEN ASKED IF WE HAD THE FIELD. WE SAID WE HAD THE FIELD IN SIGHT AND HE CLRED US FOR THE VISUAL TO RWY 24R. I SELECTED DIRECT TO PALAC IN THE FMS AND THE PF ENGAGED THE AUTOPLT. THE REMAINDER OF THE APCH WAS NORMAL AND UNEVENTFUL. BASICALLY; THE PF DID 2 SMALL S-TURNS ACROSS THE ARR COURSE. I THINK THAT THE FACTORS THAT LED TO THE COURSE DEVS WERE: 1) PREPROGRAMMING AN EXPECTED RWY AND NOT GETTING IT. 2) CTR DID NOT GIVE US ANY INDICATION OF WHICH RWY TO EXPECT EVEN THOUGH THE ARR IS RWY DEPENDENT. 3) APCH CLRED US FOR A RWY DEPENDENT ARR WHEN WE WERE ALMOST OVER THE FIRST FIX. 4) TIME SPENT REPROGRAMMING THE FMS PUT THE ACFT IN A SIT THAT CAUSED THE INITIAL DEV. I HAVE HEARD OF THIS HAPPENING BEFORE AND AFTER THINKING ABOUT POSSIBLE SOLUTIONS HAVE COME UP WITH THESE: 1) PROGRAM BOTH RWYS; ONE IN RTE #1; THE OTHER IN RTE #2 FOR IMMEDIATE USE. 2) ASK CTR FOR AN ANTICIPATED RWY ASSIGNMENT PRIOR TO THE FIX WHERE WE GET HANDED OFF TO APCH. DOING THESE 2 THINGS SHOULD HELP ALLEVIATE THIS PROB IN THE FUTURE. ANOTHER CONTRIBUTING FACTOR IS THAT THE FIXES GRAMM; RUSTT; AND CIVET ARE VERY CLOSE TOGETHER. THIS MAKES IT MORE DIFFICULT FOR THE AUTOPLT TO RE-INTERCEPT IF IT GETS EVEN A LITTLE OFF COURSE. THE SOLUTION TO THIS PART IS TO DO WHAT WE DID; GET A VECTOR FURTHER DOWN THE ARR PATH. ANOTHER POSSIBLE CONTRIBUTING FACTOR WAS THE DECREASING SPD ASSIGNMENTS DURING THE ARR. EVERY TIME ATC CHANGES SOMETHING; THE FMS GETS REPROGRAMMED AND HAS TO RE-COMPUTE THE PATHS.

Data retrieved from NASA's ASRS site as of January 2009 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.