Formulation of a Model to Predict Second Destination Transportation Tonnage Estimates for Future Budget Requirements.

摘要

The purpose of the research was to develop a better method for computing SDT tonnage estimates which are used to derive future budget requirements. The objectives were to determine whether flying hours, manpower, or both were significantly reliable predictors of tonnage estimates and to develop a computerized model for computation of SDT tonnage and budget estimates. Obtaining both actual and programmed data with MAC as a data source, discontinuous linear regression was used to derive twelve equations for the five overseas geographic areas serviced by MAC. The equations were synthesized into four models which were tested by various statistical methods to determine their overall validity. The optimal model chosen verified that programmed flying hours and manpower were significantly reliable predictors of tonnage estimates. A comparison between the optimal model and the method currently used to estimate tonnage revealed that the model provided a significantly more accurate estimation of tonnage to be moved. Computerization of the model was developed utilizing FORTRAN and the Statistical Package for the Social Sciences. After reviewing the results, the authors concluded that tonnage can be estimated using programmed flying hours and manpower as variables to derive future SDT budget requirements. (Author)