Narrative:

Our night freight flight had just lifted off when the tower called with a message that our company wanted us to return to the loading ramp. After about 20 mins of radio communications with the operations agent who was relaying messages from the load supervisor and our chief pilot; we learned that the freight had not been loaded according to the load plan. This meant that the center of gravity that we had computed was not correct because the data we used did not reflect where the pallets of freight had actually been loaded. We turned around and made an overweight landing 36 mins after taking off. This is my understanding of the error process: 1) our company's loading system idents each pallet of freight by serial number. 2) these pallets are large 'cookie sheets' of aluminum with the serial number stamped on the aluminum rim of the pallet. 3) when the pallet is weighed; that weight is assigned to the pallet's serial number in the weight and balance computer program. 4) when the pallets are marshalled at the aircraft; the ramp worker in charge of loading the pallets on the aircraft uses a computer generated load plan to assign each pallet to its planned position in the aircraft. 5) the error occurred when the ramp worker in charge of loading the pallets confused 2 similar serial numbers. Dyslexia or the dim lighting conditions on the ramp could have been the source of the error. I do not know how the load supervisor discovered the loading error; but that is why he called us back to the loading ramp after we had taxied out. 6) 2 or more pallets were placed in the wrong position; and that caused the dry center of gravity to be out of limits. 7) fortunately for us; because we were on a 4 hour flight; the center fuel tank was full. This put the wet center of gravity within limits and the aircraft flew normally. The overweight landing turned out to be a better choice since dumping 15000 pounds of fuel out of the center tank could have moved the center of gravity out of limits. This error might have been prevented by identing each pallet inside the freight handling building out of the WX and in better lighting conditions. Further; a large clip-on tag should be used to identify the loading position of the pallet (ie; P-2). That tag would make it possible for every ramp worker to verify that the pallet is loaded in the proper position when it is pushed into position on the aircraft. Time pressure is another source of errors. Calculating the load plan is the last step in the loading process; and everyone is waiting on the load plan. So the worker who calculates the load plan is usually trying to work as fast as possible; and that leads to errors. Our company's load planning is done with a computer program that requires the worker to manually juggle the loading to get the center of gravity within limits while considering hazmat locations; zone weight limits; and what has to be loaded or unloaded at the next stop. It seems to me that a computer program would work faster and do a more reliable job; but no one seems to have written anything that would take the inputs of the pallet weights and other limitations and calculate the best load plan. A broader part of the loading problem in my view is that the FAA does not seem to want to control the handling of freight. The FAA's solution to all loading problems like this is to make the crew responsible for supervising the loading. This is not a practical solution because the freight handling system extends a long way 'up stream' from the main cargo door of the aircraft. Most of the freight is accepted; loaded on pallets; weighed; and calculated long before the crew lands and parks the aircraft. The situation may be complicated by the company organization. In our case the freight handling division weighs and loads the freight; and the airline division operates the aircraft. It may seem that regulating the loading would involve inspecting another company division. My view is that a simple organizational change could resolve the problem. The responsibility for calculating the load plan could be shifted to the airline so that aircraft weight and balance is an airline responsibility. If the airline weight and balance personnel calculated the load plan and placed the position tags on each pallet the freight handling division ramp workers could still load the pallets on the aircraft. My thinking is that the FAA regulatory authority/authorized applies to everything that affects the airworthiness of an aircraft; including the weight and balance. It was finally recognized that properly handling hazmat requires a system that starts when the freight is accepted; not when it is loaded. The system used to load an aircraft also requires regulatory oversight and possibly an FAA licensed loadmaster.

Related reports:

Google
 

Original NASA ASRS Text

Title: B727 FREIGHTER CREW IS INFORMED SHORTLY AFTER TKOF THAT THE CG IS OUT OF LIMITS; AND RETURNS SAFELY FOR AN OVERWEIGHT LANDING.

Narrative: OUR NIGHT FREIGHT FLT HAD JUST LIFTED OFF WHEN THE TWR CALLED WITH A MESSAGE THAT OUR COMPANY WANTED US TO RETURN TO THE LOADING RAMP. AFTER ABOUT 20 MINS OF RADIO COMS WITH THE OPS AGENT WHO WAS RELAYING MESSAGES FROM THE LOAD SUPVR AND OUR CHIEF PLT; WE LEARNED THAT THE FREIGHT HAD NOT BEEN LOADED ACCORDING TO THE LOAD PLAN. THIS MEANT THAT THE CTR OF GRAVITY THAT WE HAD COMPUTED WAS NOT CORRECT BECAUSE THE DATA WE USED DID NOT REFLECT WHERE THE PALLETS OF FREIGHT HAD ACTUALLY BEEN LOADED. WE TURNED AROUND AND MADE AN OVERWT LNDG 36 MINS AFTER TAKING OFF. THIS IS MY UNDERSTANDING OF THE ERROR PROCESS: 1) OUR COMPANY'S LOADING SYS IDENTS EACH PALLET OF FREIGHT BY SERIAL NUMBER. 2) THESE PALLETS ARE LARGE 'COOKIE SHEETS' OF ALUMINUM WITH THE SERIAL NUMBER STAMPED ON THE ALUMINUM RIM OF THE PALLET. 3) WHEN THE PALLET IS WEIGHED; THAT WT IS ASSIGNED TO THE PALLET'S SERIAL NUMBER IN THE WT AND BAL COMPUTER PROGRAM. 4) WHEN THE PALLETS ARE MARSHALLED AT THE ACFT; THE RAMP WORKER IN CHARGE OF LOADING THE PALLETS ON THE ACFT USES A COMPUTER GENERATED LOAD PLAN TO ASSIGN EACH PALLET TO ITS PLANNED POS IN THE ACFT. 5) THE ERROR OCCURRED WHEN THE RAMP WORKER IN CHARGE OF LOADING THE PALLETS CONFUSED 2 SIMILAR SERIAL NUMBERS. DYSLEXIA OR THE DIM LIGHTING CONDITIONS ON THE RAMP COULD HAVE BEEN THE SOURCE OF THE ERROR. I DO NOT KNOW HOW THE LOAD SUPVR DISCOVERED THE LOADING ERROR; BUT THAT IS WHY HE CALLED US BACK TO THE LOADING RAMP AFTER WE HAD TAXIED OUT. 6) 2 OR MORE PALLETS WERE PLACED IN THE WRONG POS; AND THAT CAUSED THE DRY CTR OF GRAVITY TO BE OUT OF LIMITS. 7) FORTUNATELY FOR US; BECAUSE WE WERE ON A 4 HR FLT; THE CTR FUEL TANK WAS FULL. THIS PUT THE WET CTR OF GRAVITY WITHIN LIMITS AND THE ACFT FLEW NORMALLY. THE OVERWT LNDG TURNED OUT TO BE A BETTER CHOICE SINCE DUMPING 15000 LBS OF FUEL OUT OF THE CTR TANK COULD HAVE MOVED THE CTR OF GRAVITY OUT OF LIMITS. THIS ERROR MIGHT HAVE BEEN PREVENTED BY IDENTING EACH PALLET INSIDE THE FREIGHT HANDLING BUILDING OUT OF THE WX AND IN BETTER LIGHTING CONDITIONS. FURTHER; A LARGE CLIP-ON TAG SHOULD BE USED TO IDENT THE LOADING POS OF THE PALLET (IE; P-2). THAT TAG WOULD MAKE IT POSSIBLE FOR EVERY RAMP WORKER TO VERIFY THAT THE PALLET IS LOADED IN THE PROPER POS WHEN IT IS PUSHED INTO POS ON THE ACFT. TIME PRESSURE IS ANOTHER SOURCE OF ERRORS. CALCULATING THE LOAD PLAN IS THE LAST STEP IN THE LOADING PROCESS; AND EVERYONE IS WAITING ON THE LOAD PLAN. SO THE WORKER WHO CALCULATES THE LOAD PLAN IS USUALLY TRYING TO WORK AS FAST AS POSSIBLE; AND THAT LEADS TO ERRORS. OUR COMPANY'S LOAD PLANNING IS DONE WITH A COMPUTER PROGRAM THAT REQUIRES THE WORKER TO MANUALLY JUGGLE THE LOADING TO GET THE CTR OF GRAVITY WITHIN LIMITS WHILE CONSIDERING HAZMAT LOCATIONS; ZONE WT LIMITS; AND WHAT HAS TO BE LOADED OR UNLOADED AT THE NEXT STOP. IT SEEMS TO ME THAT A COMPUTER PROGRAM WOULD WORK FASTER AND DO A MORE RELIABLE JOB; BUT NO ONE SEEMS TO HAVE WRITTEN ANYTHING THAT WOULD TAKE THE INPUTS OF THE PALLET WTS AND OTHER LIMITATIONS AND CALCULATE THE BEST LOAD PLAN. A BROADER PART OF THE LOADING PROB IN MY VIEW IS THAT THE FAA DOES NOT SEEM TO WANT TO CTL THE HANDLING OF FREIGHT. THE FAA'S SOLUTION TO ALL LOADING PROBS LIKE THIS IS TO MAKE THE CREW RESPONSIBLE FOR SUPERVISING THE LOADING. THIS IS NOT A PRACTICAL SOLUTION BECAUSE THE FREIGHT HANDLING SYS EXTENDS A LONG WAY 'UP STREAM' FROM THE MAIN CARGO DOOR OF THE ACFT. MOST OF THE FREIGHT IS ACCEPTED; LOADED ON PALLETS; WEIGHED; AND CALCULATED LONG BEFORE THE CREW LANDS AND PARKS THE ACFT. THE SITUATION MAY BE COMPLICATED BY THE COMPANY ORGANIZATION. IN OUR CASE THE FREIGHT HANDLING DIVISION WEIGHS AND LOADS THE FREIGHT; AND THE AIRLINE DIVISION OPERATES THE ACFT. IT MAY SEEM THAT REGULATING THE LOADING WOULD INVOLVE INSPECTING ANOTHER COMPANY DIVISION. MY VIEW IS THAT A SIMPLE ORGANIZATIONAL CHANGE COULD RESOLVE THE PROB. THE RESPONSIBILITY FOR CALCULATING THE LOAD PLAN COULD BE SHIFTED TO THE AIRLINE SO THAT ACFT WT AND BAL IS AN AIRLINE RESPONSIBILITY. IF THE AIRLINE WT AND BAL PERSONNEL CALCULATED THE LOAD PLAN AND PLACED THE POS TAGS ON EACH PALLET THE FREIGHT HANDLING DIVISION RAMP WORKERS COULD STILL LOAD THE PALLETS ON THE ACFT. MY THINKING IS THAT THE FAA REGULATORY AUTH APPLIES TO EVERYTHING THAT AFFECTS THE AIRWORTHINESS OF AN ACFT; INCLUDING THE WT AND BAL. IT WAS FINALLY RECOGNIZED THAT PROPERLY HANDLING HAZMAT REQUIRES A SYS THAT STARTS WHEN THE FREIGHT IS ACCEPTED; NOT WHEN IT IS LOADED. THE SYS USED TO LOAD AN ACFT ALSO REQUIRES REGULATORY OVERSIGHT AND POSSIBLY AN FAA LICENSED LOADMASTER.

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.