SYSTEMATIC LITERATURE REVIEW SISTEM PENDUKUNG KEPUTUSAN PEMELIHARAAN PERKERASAN JALAN BERBASIS MULTI-KRITERIA BEBAN LALU LINTAS DAN IKLIM
Kata Kunci:
Kecerdasan Buatan, Bibliometrik, Data Mining, Sistem Pendukung Keputusan, Pembelajaran Mesin, Multi-Beban, Perkerasan Jalan, Tinjauan Pustaka SistematisAbstrak
Perkerasan jalan menghadapi tantangan multi-beban yang kompleks berupa beban lalu lintas, iklim ekstrem, dan fenomena Urban Heat Island (UHI) secara simultan. Penelitian ini menerapkan pendekatan Systematic Literature Review (SLR) berbasis kerangka PICO yang diintegrasikan dengan teknik Data Mining bibliometrik menggunakan VOSviewer dan R-Studio, guna memetakan tren dan pola penelitian dari 50 artikel terindeks Scopus (2020–2026). Metode berbasis komputasi mencakup analisis co-occurrence kata kunci, network graph antar-node, dan density visualization untuk mengidentifikasi klaster topik dominan. Hasil analisis menunjukkan dominasi penelitian material aspal dan beban lalu lintas, dengan penerapan Machine Learning (ML) dan Artificial Intelligence (AI) yang masih sangat terbatas. Lima research gap utama teridentifikasi, mencakup minimnya integrasi AI–IoT untuk pemantauan real-time dan absennya model prediktif berbasis ML untuk kondisi iklim tropis Indonesia. Penelitian ini menghasilkan roadmap komputasional sebagai fondasi pengembangan Decision Support System (DSS) berbasis data untuk manajemen pemeliharaan infrastruktur jalan di Indonesia.
Road pavements face complex multi-load challenges including traffic loads, extreme climate, and Urban Heat Island (UHI) phenomena simultaneously. This study applies a Systematic Literature Review (SLR) approach based on the PICO framework, integrated with bibliometric Data Mining techniques using VOSviewer and R-Studio, to map research trends and patterns from 50 Scopus-indexed articles (2020–2026). Computational methods include keyword co-occurrence analysis, node network graphs, and density visualization to identify dominant topic clusters. Results reveal dominance of asphalt material and traffic load research, with very limited application of Machine Learning (ML) and Artificial Intelligence (AI). Five major research gaps were identified, including minimal AI–IoT integration for real-time monitoring and the absence of ML-based predictive models for Indonesia’s tropical climate conditions. This study provides a computational roadmap as a foundation for developing a data-driven Decision Support System (DSS) for road infrastructure maintenance management in Indonesia.
