New Product Short-Term Demands Forecasting with Boxplot-Based Fractional Grey Prediction Model

The cost of investing in new product development (NPD) is high, and it a feasible way to use demand forecasts for customer or end-users as decisive reference. However, this short-term time-series data has difficulties learning because there no past performance on which base the estimates. In past, been proven that cumulative method fractional grey prediction model (FGM) better than traditional integer (GM) model. There are many studies using different optimal algorithms determine moderate score order. How set coefficient ? FGM also worth exploring. Therefore, research reveals uses box-and-whisker plots estimate trends data, known boxplot-based scale (boxplot-based FGM, BP-FGM) improve accuracy predictors by setting sets ?. experiment, examined dataset was collected from well-known equipment manufacturer object. For modeling, mean absolute percentage error (MAPE) established objective function optimization model, results three datasets verified effect through commodity attributes public test its production, experimental show BP-FGM FGM.