%A NIM.: 23206052009 Muhammad Fahrurrozi
%O Dr. Agus Mulyanto, S.Si., M.Kom., ASEAN Eng.
%T EVALUASI AKURASI MODEL HYBRID SIMPLE ADDITIVE WEIGHTING – WEIGHTED PRODUCT DENGAN PEMBOBOTAN ADAPTIF MENGGUNAKAN ANALYTICAL HIERARCHY PROCESS UNTUK SELEKSI CALON PENGHUNI ASRAMA
%X This study examines the application of the Simple Additive Weighting (SAW) and
Weighted Product (WP) methods, as well as their integration into a hybrid model with
adaptive weighting based on the Analytic Hierarchy Process (AHP), for the selection of
dormitory applicants. SAW has the advantage of computational simplicity, while WP is more
sensitive to data variations. Both methods have limitations, making it necessary to combine
them to obtain results that are more balanced, stable, and accurate. Adaptive weighting with
AHP is employed to ensure that the criteria weights align with the dormitory administrators’
priorities and remain logically consistent. The eight criteria used include GPA, parents’
income and occupation, number of siblings, organizational activity, non-academic
achievements, scholarship status, and semester.
This study utilizes 120 alternative data, analyzed using SAW, WP, and the Hybrid SAW–
WP method. The evaluation was carried out using Mean Absolute Error (MAE), Root Mean
Square Error (RMSE), and Spearman correlation. The results show that SAW – GT produces
MAE = 10.67 and RMSE = 13.16. These error values are relatively high because the
differences were calculated based on ranking (1–120), not the direct scores. Nevertheless,
the Spearman correlation reached ρ = 0.928, indicating a very strong relationship. WP –
GT obtained MAE = 5.96 and RMSE = 4.625, with a Spearman correlation of ρ = 0.985.
This result is better than SAW, showing that WP is closer to GT in terms of ranking
consistency. Meanwhile, Hybrid – GT achieved the best performance with MAE = 3.40 and
RMSE = 4.57. The Spearman correlation reached ρ = 0.991, the highest among the three
methods. This proves that the Hybrid method is the most consistent in replicating the GT
ranking.
In conclusion, the Hybrid SAW–WP with adaptive AHP weighting is not only consistent
in producing rankings but also more stable compared to using a single method. This
approach can be considered a more comprehensive solution for decision support systems,
particularly in the context of dormitory applicant selection.
%K Sistem Pendukung Keputusan, Hybrid, Simple Additive Weighting, Weighted Product, Analytic Hierarchy Process (AHP)
%D 2025
%I UIN SUNAN KALIJAGA YOGYAKARTA
%L digilib73819