Narrative:

Cleared for approach, visual conditions. Traffic was commuter aircraft being vectored from south to runway 28L. Picked traffic up visually early in approach. Initially was well ahead of traffic. Approach control issued 3 speed reductions which resulted in rendezvous with traffic abeam and closing (due to angling nature of approach). There was no traffic ahead or behind to cause spacing concerns. Full automation was in use. LNAV and VNAV controling autoplt with ILS for runway 28R armed. Tracking was accurate, GS captured first while LNAV tracked FMS visual track. This approach angles in to intercept localizer at 4.4 DME. Closure rate continued. Just above 1000 ft got TCASII RA with 'descend, descend' command. Traffic was in sight and lateral separation rapidly decreasing. Figures of the pilots in the commuter cockpit were clearly visible. I cannot say if the commuter was overshooting. We had not yet intercepted the localizer. With lateral separation rapidly decreasing, I disconnected the autoplt at an estimated 6-8 seconds from impact and banked the aircraft right, away from the localizer and the traffic. Once acceptable separation was obtained, I continued an offset angling approach, intercepting centerline on short final for landing. After landing, I phoned approach control and asked why the controller had engineered our 'formation arrival.' in short, their goal was to have both aircraft touch down simultaneously. This minimizes the amount of time that the departure runways (1/19L&right) is fouled due to arrs. This maximizes their traffic flow rate. The norcal controller I spoke with said they had received numerous complaints from pilots about this procedure. In my view this is a classic case of minimizing safety margins to maximize traffic flows at a congested airport. The box note on the approach chart says 'caution aircraft operating within 500 ft to parallel runway possible wake turbulence.' it is noteworthy our manual defines an near midair collision as when another aircraft penetrates a 500 ft bubble around your aircraft. Runway spacing is 750 ft. This places wingtip clearance at less than 650 ft. This leaves total flight path tolerance for both aircraft at 150 ft before near miss criteria. Consider effects of crosswind. Also, if captain is flying runway 28L approach or first officer is flying runway 28R, the other aircraft may not be visible if it is slightly aft. Consider how distracting it is to be flying this approach procedure, monitoring localizer and GS capture, airspeed, altitude, conducting landing checklist, and contacting tower while watching your side window fill up with airplane and have the TCASII yelling 'descend, descend' at just over 1000 ft in the air. In summary, with a lateral separation of only 650 ft with perfect runway alignment and controllers striving to zero out nose-to-tail separation, this procedure in my opinion (with over 2 decades of airline experience) poses an extreme hazard for a midair collision or landing accident/incident due to distrs caused by close proximity traffic.

Google
 

Original NASA ASRS Text

Title: B757 CREW ON VISUAL APCH INTO SFO RWY 28R IS REQUIRED TO TAKE EVASIVE ACTION ON FINAL FROM A PROP COMMUTER INBOUND TO RWY 28L.

Narrative: CLRED FOR APCH, VISUAL CONDITIONS. TFC WAS COMMUTER ACFT BEING VECTORED FROM S TO RWY 28L. PICKED TFC UP VISUALLY EARLY IN APCH. INITIALLY WAS WELL AHEAD OF TFC. APCH CTL ISSUED 3 SPD REDUCTIONS WHICH RESULTED IN RENDEZVOUS WITH TFC ABEAM AND CLOSING (DUE TO ANGLING NATURE OF APCH). THERE WAS NO TFC AHEAD OR BEHIND TO CAUSE SPACING CONCERNS. FULL AUTOMATION WAS IN USE. LNAV AND VNAV CTLING AUTOPLT WITH ILS FOR RWY 28R ARMED. TRACKING WAS ACCURATE, GS CAPTURED FIRST WHILE LNAV TRACKED FMS VISUAL TRACK. THIS APCH ANGLES IN TO INTERCEPT LOC AT 4.4 DME. CLOSURE RATE CONTINUED. JUST ABOVE 1000 FT GOT TCASII RA WITH 'DSND, DSND' COMMAND. TFC WAS IN SIGHT AND LATERAL SEPARATION RAPIDLY DECREASING. FIGURES OF THE PLTS IN THE COMMUTER COCKPIT WERE CLRLY VISIBLE. I CANNOT SAY IF THE COMMUTER WAS OVERSHOOTING. WE HAD NOT YET INTERCEPTED THE LOC. WITH LATERAL SEPARATION RAPIDLY DECREASING, I DISCONNECTED THE AUTOPLT AT AN ESTIMATED 6-8 SECONDS FROM IMPACT AND BANKED THE ACFT R, AWAY FROM THE LOC AND THE TFC. ONCE ACCEPTABLE SEPARATION WAS OBTAINED, I CONTINUED AN OFFSET ANGLING APCH, INTERCEPTING CTRLINE ON SHORT FINAL FOR LNDG. AFTER LNDG, I PHONED APCH CTL AND ASKED WHY THE CTLR HAD ENGINEERED OUR 'FORMATION ARR.' IN SHORT, THEIR GOAL WAS TO HAVE BOTH ACFT TOUCH DOWN SIMULTANEOUSLY. THIS MINIMIZES THE AMOUNT OF TIME THAT THE DEP RWYS (1/19L&R) IS FOULED DUE TO ARRS. THIS MAXIMIZES THEIR TFC FLOW RATE. THE NORCAL CTLR I SPOKE WITH SAID THEY HAD RECEIVED NUMEROUS COMPLAINTS FROM PLTS ABOUT THIS PROC. IN MY VIEW THIS IS A CLASSIC CASE OF MINIMIZING SAFETY MARGINS TO MAXIMIZE TFC FLOWS AT A CONGESTED ARPT. THE BOX NOTE ON THE APCH CHART SAYS 'CAUTION ACFT OPERATING WITHIN 500 FT TO PARALLEL RWY POSSIBLE WAKE TURB.' IT IS NOTEWORTHY OUR MANUAL DEFINES AN NMAC AS WHEN ANOTHER ACFT PENETRATES A 500 FT BUBBLE AROUND YOUR ACFT. RWY SPACING IS 750 FT. THIS PLACES WINGTIP CLRNC AT LESS THAN 650 FT. THIS LEAVES TOTAL FLT PATH TOLERANCE FOR BOTH ACFT AT 150 FT BEFORE NEAR MISS CRITERIA. CONSIDER EFFECTS OF XWIND. ALSO, IF CAPT IS FLYING RWY 28L APCH OR FO IS FLYING RWY 28R, THE OTHER ACFT MAY NOT BE VISIBLE IF IT IS SLIGHTLY AFT. CONSIDER HOW DISTRACTING IT IS TO BE FLYING THIS APCH PROC, MONITORING LOC AND GS CAPTURE, AIRSPD, ALT, CONDUCTING LNDG CHKLIST, AND CONTACTING TWR WHILE WATCHING YOUR SIDE WINDOW FILL UP WITH AIRPLANE AND HAVE THE TCASII YELLING 'DSND, DSND' AT JUST OVER 1000 FT IN THE AIR. IN SUMMARY, WITH A LATERAL SEPARATION OF ONLY 650 FT WITH PERFECT RWY ALIGNMENT AND CTLRS STRIVING TO ZERO OUT NOSE-TO-TAIL SEPARATION, THIS PROC IN MY OPINION (WITH OVER 2 DECADES OF AIRLINE EXPERIENCE) POSES AN EXTREME HAZARD FOR A MIDAIR COLLISION OR LNDG ACCIDENT/INCIDENT DUE TO DISTRS CAUSED BY CLOSE PROX TFC.

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.