The weight and balance calculations performed by the crew underestimated the actual take-off weight of the aircraft at Gander by about 14,000 pounds. The underestimation of the take-off weight was primarily due to the use of a standard average weight that did not take into account the nature of the passengers being carried. Contributing to the underestimation was the lower cargo weight used by the Cologne/Cairo crew and the company's use of a basic operating weight that did not take into account the weight of removable galley and cabin equipment and potable water.

The standard weight used was applicable to an average civilian adult with five pounds of carry- on baggage. The Board determined that the average weight of the passengers carried on MF1285R was approximately 220 pounds, 30 per cent higher than the 170-pound average used for flight planning purposes.

The original incorrect figures continued to be used for the flights to Gander, and the planned flight from Gander to Fort Campbell. As a result of the underestimation of the weight of the aircraft load, the Board believes that the maximum authorized take-off weight was exceeded by 8,000 pounds on take-off from Cologne.

Although the use of actual passenger weights was required by the Arrow Air Operations Manual, the system employed by the company for determining weight and centre of gravity did not provide specific direction on how to use actual weights. It was evident that weights on previous flights had been used in actual passenger weight and balance calculations; thus, it is apparent that crews were familiar with a method to adjust passenger weights to reflect a more accurate weight. The actual weight of individual passengers was not determined in Cairo by either MFO personnel or Arrow Air. It should have been apparent to the crew who completed the initial weight and balance calculations that an average weight of 170 pounds was considerably less than the actual weight, and the load sheet should have reflected this higher weight.

There was further evidence to indicate that Arrow Air flight crews were not determining the weight and centre of gravity for every flight. A review of weight and centre of gravity documentation for the series of MFO rotation flights which commenced on 03 December 1985 and the series of flights which commenced on 10 December 1985 determined that the passenger and cargo weights used on the flights from Cairo to Fort Campbell were identical to the weights used on the inbound flights from McChord AFB to Cairo. Despite the fact that a different load was being boarded at Cairo, it is apparent that the flight crew was copying the load figures for the inbound flight.

The Board also noted significant inconsistencies in documentation regarding loads being carried on the two series of rotation flights. The Board obtained considerable evidence that suggested the loads carried from McChord AFB to Cairo on 03 December 1985 and 10 December 1985 were substantially the same. Despite this similarity, the passenger weight as indicated on load sheets prepared by the same flight crew differed by 8,000 pounds. The cargo loads carried on these flights were reportedly also similar in weight nonetheless, on the load sheets, the indicated weights were again 8,000 pounds different Because new weight and balance calculations were not performed for the return flight to the United States, these same inconsistencies were present in the load documentation for the flights originating in Cairo. In addition, the Board notes that the number of passengers indicated on the load sheets prepared on departure from Cairo, Cologne, and Gander was incorrect.

These inconsistencies are further evidence that the weight of loads being carried on Arrow Air aircraft was not being determined accurately.

Contributing to this situation were inadequate load documentation and record keeping. Throughout its investigation, the Board experienced difficulties in obtaining accurate documentation regarding the weight of passengers and cargo carried on the MFO chartered flights both to and from Cairo.

Although the cargo was being weighed prior to departure from both Cairo and McChord AFB, no manifests or records of the scaled weights were being kept. Nor were such records kept of the scaled weight of passengers departing McChord AFB. The only U.S. military load records recovered that pertained to the series of flights were the McChord AFB Records/Audit manifests which did not agree with either the scaled weights or the figures used on the Arrow Air load sheets. Weight information prepared by U.S. military and MFO personnel was passed to Arrow Air personnel on slips of paper. It could not be determined what, if any, use was made by the Arrow Air personnel of this weight information. None of the load sheets prepared prior to flight reflected the weights calculated by U.S. Army or MFO personnel.

The Board also noted numerous inconsistencies regarding load weights in the load planning guidance material available to personnel from Arrow Air, the U.S. Army, and the MFO. These inconsistencies added to what the Board believes was considerable uncertainty regarding the actual weight of the loads carried on the MFO flights.

In calculating the actual weights of loads carried on the two series of rotation flights, the Board determined that, on each flight the maximum authorized ZFW was exceeded. Furthermore, it is the conclusion of the Board that Arrow Air flight crews and management were aware that the maximum ZFW was being exceeded on a regular basis.

The flight crew members who were responsible for the calculation of the weight and centre of gravity in Cairo acknowledged that they believed the load to be about 10,000 pounds heavier than that indicated on the load sheet. The ZFW indicated on the load sheet was 229,621 pounds, less than 400 pounds under the maximum authorized ZFW. Therefore, the crew operated the aircraft almost 10,000 pounds over the maximum authorized ZFW. On those occasions where the passenger weights on the load sheets were higher than the standard average weight, the Board noted that the cargo weight was always less than the cargo weight shown on the load sheets where a lower passenger weight was used. The reduction in cargo weight corresponded closely to the increase in passenger weight. In every case, the ZFW was just under the maximum allowable. It is the opinion of the Board that the load sheet calculations performed by the flight crew were planned to demonstrate adherence to the maximum allowable ZFW. It further believes that the standard average passenger weight, although it did not accurately reflect the weight of passengers being carried, was being used in an effort to keep the ZFW indicated on the load sheet below the maximum authorized.

Arrow Air management was concerned about the ability of the aircraft to carry the MFO contracted loads within its ZFW limits. In 1985, they had contemplated action to raise the ZFW limit of the aircraft, although this action was not actively pursued. In discussions with Arrow Air management personnel following the accident, it was evident they were aware that, in order to conduct MFO flights, the maximum design ZFW of the aircraft was a problem. The contract between Arrow Air and the MFO specified a baggage allowance of 154 pounds per passenger. Assuming an average passenger weight of 170 pounds, Arrow Air had contracted to carry payloads of up to 81,000 pounds on the MFO flights. This value was approximately 13,500 pounds in excess of the payload capability of the aircraft used for the MFO flights. This discrepancy between contractual obligations and the payload capacity of the aircraft was known to management; however. action to increase the maximum design ZFW was not being pursued.