APATS is believed to one of the first Canadian studies that attempted to forecast regional air passenger travel as a fraction of traffic on all modes.

The study was commissioned by the air administration of Transport Canada and contracted to ADI Ltd. The tellectual work was by A. M. Stevens and Frank Wilson. Many others worked on the study including Bart Claus and Steve Wheatley. Bart did much of the heavy data processing.

The objective of APATS was to forecast the demand for regional air passenger service to 1990 and suggest some air service supply alternatives. These would be used for developing a regional plan for airports. It was a pilot study for the national airports plan. The methodology was to consider all modes of travel for relatively long or intercity trips and deduce the air portion by comparing the competition between the modes.

The general approach was to survey the exiting long haul traffic for air, automobile, bus, and rail and to forecast the total traffic and then imply the fraction that would be carried by air in 1971, 1981, and 1991. The years were chosen because they are census times. The work was initiated by Transport Canada Air Administration.

The main features of the study were:
1. A set of zones for each of the four provinces.
2. Aggregation of the zones into airport catchment areas.
3. A screen line survey in NB and at other selected locations to develop a highway OD table.
4. A bus ticket lift for the major bus lines.
5. Rail passenger traffic between stations & some boarding & detraining data.
6. Air traffic supplied by the airlines.
7. On board aircraft and air terminal surveys.
8. A set of base year (1968) OD tables by mode.
9. Forecasts of populations and disposable incomes for the airport catchment areas.
10. Forecasts of air traffic based on a linear model of air traffic only.
11. Forecasts of air traffic as a mode choice from total traffic as estimated by a hyperbolic total traffic model.

Early in the work it was agreed that the regional economic and demographic forecasts would be supplied by others and that the main task would be to make the passenger transport forecasts. Later as the work progressed it became necessary to make population and other economic projections. The lack of good economic and demographic forecasts weakened the whole effort.

UNB had just installed an IBM 360 model 50 computer to replace an IBM 1620 which had been the UNB mainframe. All survey and other information was submitted on punched cards for batch processing. The core memory was 600K. Near the end of the study the first version of SPSS became available. This was too late to be of value and custom software in PL/1 and FORTRAN was written. The Scientific Subroutine Package (FORTRAN) provided the statistical routines.

Data was collected for all four modes and compiled into base year tables. The main data manipulating function TRIPFOR balanced the tables, produced total and air trip forecasts. Service alternatives were developed by visual interpretation of the forecasts and some simplistic rules such as the minimum annual traffic requiring one aircraft seat was 200 = (0.55 ’ 365). i.e. assume a 55% load factor. End to date; ams 950228

MYNET Links: ( to TOP of page )
• Information on the APATS models
• Node list, APATS'69
• Centroid Population and Forecasts [m31]
• Air Mode

End to date: 040720, ams