Past research proposed that it is possible to forecast cargo demand using time series models and that there exists potential cost savings in the way that Civilian Reserve Air Fleet (CRAF) is used for cargo airlift. United States Transportation Command (USTRANSCOM) performs annual 'fixed-buys' of CRAF to support airlift needs. Forecasted cargo demand would allow for reasonably accurate cargo projections vs. the current expected value estimation. Accurate forecasting allows for greater 'fixed-buys,' further incentivizing CRAF airlines as well as reducing the number of additional aircraft purchases during the quarterly and monthly buys. Multiple forecasting models are constructed and the results compared. A Monte Carlo simulation using a discrete pallet destinations distribution and a discrete pallet port arrival date distribution (based on historical data) outputs a month of projected pallet weights (with date and destination) that are equivalent to the forecasted cargo amount. The simulated pallets are then used in a heuristic cargo loading algorithm. The loading algorithm places cargo onto available aircraft (based on real schedules) given the date and the destination and outputs statistics based on the aircraft ton and pallet utilization as well as number of aircraft types used and the total cost of the projected airlift schedule. A technical approach to the operational planning of cargo airlift could provide significant cost savings or could provide an alternative planning approach changing the future of USTRANSCOM operations.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite