Expert judgement in resource forecasting: insights from the application of the Delphi method

Through application in a world-leading automotive business, this paper explores the practicalities of applying a new method for forecasting resource requirements in the absence of data. The method involves a one off effort to capture expert knowledge in a very structured fashion leading to the formation of regression equations for prediction. Creating such models creates a new conundrum: how can quantitative forecasting models, constructed through structured expert estimations, be validated and accepted in the absence of data? We employ Delphi and find that, with adaptation, it can lead to acceptance of the models generated using the new data-less method.