A “demand Pull” Approach to Short Term Forecast Development and Testing

Product development can be motivated by “demand pull” from user needs or by “technology push” from research. Historically, new aviation weather forecasts have been motivated by “technology push” considerations, in which better scientific validation results were the principal criterion for the forecast “goodness”. For example, one might demonstrate that the forecast had a higher probability of detection (Pd) and/or lower false alarm probability (Pfa) than an alternative forecast. There were no specific numerical criteria for forecast performance that would warrant operational use of the new forecast in place of the preexisting ones, except for a generic requirement for “improved” forecasting of a given phenomenon. However, in an era of significant government and airline budget austerity for civil aviation investments, it is becoming increasingly important to quantitatively demonstrate the benefits to the operational user community of improved aviation weather forecasts. If demonstrating benefits is very important, then we propose that there should be a greater element of “demand pull” involved in the product development wherein there is: