Perbandingan Metode SAW dan AHP pada Sistem Pengambilan Keputusan Pemilihan Supplier Berbasis Web dengan Algoritma Regresi ANN dan SVM

UD. Berkah Jaya is a cold storage company facing challenges in selecting the appropriate suppliers to meet its stock requirements. Decision Support Systems (DSS) methods such as Simple Additive Weighting (SAW) and Analytical Hierarchy Process (AHP) are utilized to address this issue. To enhance future decision-making, these methods are combined with predictive regression systems, namely Neural Network (NN) and Support Vector Machine (SVM). By integrating these systems, a predictive decision-making method is developed.Based on this research, it is found that the regression algorithm NN is more suitable for use, with the Mean Absolute Error (MAE) values for Neural Network and SVM being 920.81 and 2496.62, respectively, and Mean Absolute Percentage Error (MAPE) values of 8.64% and 32.99%, respectively. For DSS, SAW is more appropriate because AHP uses a comparison matrix, where the weight consistency ratio in AHP results in a negative value.