SYSTEMATIC LITERATURE REVIEW: PERAN KONSEP MATEMATIKA DISKRIT DALAM PENGEMBANGAN MODEL CREDIT SCORING PEMBIAYAAN PADA LEMBAGA KEUANGAN SYARIAH NON-BANK

Abstract

This study examines the integration of discrete mathematics into financing scoring systems for non-bank Islamic financial institutions (NBIFIs). A Systematic Literature Review was selected to consolidate evidence on mathematical structures, decision-support methods, predictive algorithms, and sharia governance, with reporting guided by PRISMA 2020. Searches in Google Scholar, Scopus, Garuda, Semantic Scholar, and PMC for publications from 2020 to 2026 identified 487 records; 89 remained after deduplication and title-abstract screening, 34 full texts were assessed, and 20 articles were included. Machine-learning and credit-scoring studies dominated the evidence base (35%), followed by studies on NBIFIs and Islamic non-bank finance (25%), PRISMA-based reviews (15%), discrete mathematics and decision modelling (15%), and DSS/MCDM applications (10%). The findings show that finite sets, logical rules, decision trees, graph structures, and ranking matrices are relevant to financing assessment, but explicit discrete-mathematics applications remain rare. Major gaps concern limited NBIFI-specific datasets, weak operationalisation of maqashid al-shariah, insufficient fairness testing, and scarce external validation. The review proposes a four-layer model linking input data, discrete-mathematical structures, hybrid scoring algorithms, and explainable decisions under sharia governance. Practically, NBIFIs should develop auditable scoring rules, involve Sharia Supervisory Boards in model validation, and provide assistance pathways for applicants below the eligibility threshold.