Model Sistem Pendukung Keputusan Kelayakan Pemberian Kredit Pada BUMDes

Ni Kadek Harum Aripuspa (STMIK Primakara)
Ni Made Estiyanti (STMIK Primakara)
Eddy Muntina Dharma (STMIK Primakara)

Progresif: Jurnal Ilmiah Komputer
Vol 18, No 2: Agustus 2022

Decision-making for granting credit to BUMDes XYZ customers, which is still done manually, allows the results of decisions taken to be subjective, resulting in inaccurate decision results. This inaccurate credit decision has the potential to increase the number of bad loans that occur. This article presents a Decision Support System (DSS) model for assessing creditworthiness using the Simple Additive Weighting (SAW) method, with reference to the 5C criteria (Character, Capacity, Capital, Collateral and Condition). The performance accuracy of the SAW method was tested with 30 samples of bad credit data and current loans of customers of a BUMDes in Bali, using the Confusion Matrix technique. The test results show the Precision value of 95.4% and recall of 91.3%. The accuracy value is classified as Excellent Classification based on the ROC (Receiver Operating Characteristic) curve. The final result of the SAW-based SPK model is then implemented in the form of an SPK application for providing credit to BUMDes XYZ customers.Keywords: Simple Additive Weighting; Accuracy; Confusion Matrix; Precision and Recall; Excellent Classification Abstrak. Pengambilan keputusan pemberian kredit kepada nasabah BUMDes XYZ yang masih dilakukan secara manual, memungkinkan hasil keputusan yang diambil bersifat subjektif, berakibat pada hasil keputusan yang tidak akurat. Keputusan pemberian kredit yang tidak akurat tersebut berpotensi meningkatkan jumlah kredit macet yang terjadi. Artikel ini menyajikan model Sistem Pendukung Keputusan (SPK) penilaian kelayakan pemberian kredit dengan menggunakan metode Simple Additive Weighting (SAW), dengan mengacu pada kriteria 5C (Character, Capacity, Capital, Collateral dan Condition). Akurasi kinerja motode SAW diuji dengan 30 sampel data kredit macet dan kredit lancar nasabah sebuah BUMDes di Bali, menggunakan teknik Confusion Matrix. Hasil uji menunjukkan nilai Precision sebesar 95,4% dan Recall sebesar 91,3%. Nilai akurasi tersebut tergolong dalam Excellent Classification berdasarkan kurva ROC (Receiver Operating Characteristic). Hasil akhir model SPK berbasis SAW kemudian diimplementasikan dalam bentuk aplikasi SPK pemberian kredit kepada nasabah BUMDes XYZ.Kata kunci: Simple Additive Weighting; Akurasi; Confusion Matrix; Precision and Recall; Excellent Classification