Reliable pre-launch forecasting of a new product plays an important part in helping companies to make investment decisions, allocate resources, and minimize risks. However, the creation of such forecasts is difficult and challenging because historical sales data are lacking. A popular approach to solve this problem is to use data from analogous products. However, the accuracy of this method depends on the set of similar products used for the analogies and their weights, and a more objective weighting method would be an improvement. Therefore, we developed a new integrated demand forecasting method that combines similarity analysis and the Bass model, which we used in an analytic hierarchy process (AHP) model to appropriately weight the forecasting results from the diffusion model of analogies. We illustrate the applicability of this forecasting method by simulating future Long-Term Evolution (LTE) 4G mobile communication subscribers in Taiwan.
展开▼
