Puspita, Pratiwi Ekaİnkaya, TülinAkansel, Mehmet2025-01-212025-01-2120191308-5514https://dergipark.org.tr/tr/download/article-file/650608https://dergipark.org.tr/tr/pub/umagd/issue/39915/473977https://doi.org/10.29137/umagd.473977https://hdl.handle.net/20.500.12587/19789Sales forecasting refers to the prediction offuture demand based on past data. A vast literature on sales forecasting hasaccumulated due to its vital role in balancing demand and supply. Among these,data mining has emerged as a powerful tool to facilitate sales forecasting. Inthis study, we use data mining methods for accurate and reliable salesforecasts in a forklift distributor company. Monthly sales data for 100different types of forklifts between 1998 and 2016 are used. The proposedforecasting methodology includes three steps. First, products with similarsales patterns are determined using hierarchical clustering. Dynamic timewarping is applied to calculate the similarities among product sales data.Second, features are extracted and selected for each cluster. In addition tothe features adopted from the literature, four new features are proposed tocharacterize intermittency. Multivariate adaptive regression splines model isused for feature selection. Third, support vector regression is used to predictfuture sales of each product cluster. Finally, the performance of the proposedapproach is evaluated according to forecasting error and complexity. Thenumerical analysis shows that the proposed approach gives reasonable accuracywith less complexity.eninfo:eu-repo/semantics/openAccessData miningclusteringforecastingmultivariate adaptive regression splines (MARS)dynamic time warping (DTW)support vector regression (SVR)Article11-254010.29137/umagd.473977473977