In this paper, we propose a new approach for generating a customized forecast ensemble that considers the user's preferences across multiple criteria. The proposed algorithm takes inputs from the user and computes a set of optimal weights assignable to the n chosen criteria. Using these weights, we define a metric called the Multi-Criteria Value, which is maximized to obtain the customized ensemble. This algorithm is called Adaptive Ensemble Generator since it incorporates m distinct forecasting methods and n evaluation criteria. We demonstrate this algorithm customized

for four different users (three real-life users and one designed user) on a large database.