<![CDATA[Abstract: Resource scarcity in social assistance programs necessitates systematic prioritization mechanisms that integrate multidimensional vulnerability indicators while maintaining transparency and reproducibility. This research proposes a computational framework leveraging Fuzzy Sugeno inference systems to optimize allocation decisions for vulnerable populations, specifically persons with disabilities. The approach transforms qualitative expert assessments and socioeconomic indicators into quantifiable priority scores through four-stage fuzzy processing: fuzzification of crisp input values into membership degrees, activation of 27 IF-THEN inference rules, computation of weighted ?-predicates via minimum operator, and defuzzification via Sugeno's weighted average method producing normalized output [0,1]. Implementation utilized PHP-MySQL architecture with JavaScript-based visualizations to develop a web-accessible prototype capable of processing multifactorial household profiles. Validation across 10 representative cases demonstrates superior consistency relative to conventional manual assessment approaches. The system eliminates subjective bias through deterministic rule-based inference while maintaining semantic alignment with domain expertise. Results indicate the framework generates transparent, auditable, and scalable solutions for social policy implementation, addressing persistent challenges in allocating limited public resources to maximally deserving beneficiaries. The methodology contributes novel applications of computational intelligence to public administration and extends Sugeno fuzzy logic paradigms beyond traditional forecasting domains toward decision support systems.Keyword: fuzzy sugeno; resource allocation; vulnerability assessment; disability prioritization; social policyAbstrak: Kelangkaan sumber daya dalam program bantuan sosial menuntut mekanisme prioritisasi sistematis yang mengintegrasikan indikator kerentanan multidimensi sambil mempertahankan transparansi dan reproducibilitas. Penelitian ini mengusulkan kerangka komputasi yang memanfaatkan sistem inferensi Fuzzy Sugeno untuk mengoptimalkan keputusan alokasi bagi populasi rentan, khususnya penyandang disabilitas. Pendekatan mentransformasi penilaian ahli kualitatif dan indikator sosial-ekonomi menjadi skor prioritas terkuantifikasi melalui pemrosesan fuzzy empat tahap: fuzzifikasi nilai input crisp menjadi derajat keanggotaan, aktivasi 27 aturan inferensi IF-THEN, komputasi ?-predikat tertimbang via operasi minimum, dan defuzzifikasi via metode rata-rata tertimbang Sugeno menghasilkan output ternormalisasi [0,1]. Implementasi memanfaatkan arsitektur PHP-MySQL dengan visualisasi berbasis JavaScript untuk mengembangkan prototipe web-accessible mampu memproses profil rumah tangga multifaktorial. Validasi lintas 10 kasus representatif menunjukkan konsistensi superior relatif terhadap pendekatan penilaian manual konvensional. Sistem mengeliminasi bias subjektif melalui inferensi berbasis aturan deterministik sambil mempertahankan keselarasan semantik dengan expertise domain. Hasil mengindikasikan kerangka kerja menghasilkan solusi transparan, dapat diaudit, dan scalable untuk implementasi kebijakan sosial, mengatasi tantangan persisten alokasi sumber daya publik terbatas kepada beneficiary paling layak. Metodologi berkontribusi aplikasi novel soft computing terhadap administrasi publik dan memperluas paradigma logika fuzzy Sugeno melampaui domain forecasting tradisional menuju sistem pendukung keputusan.Kata kunci: fuzzy sugeno; alokasi sumber daya; penilaian kerentanan; prioritisasi disabilitas; kebijakan sosial]]>
Copyrights © 2025
