Decision-Ready Composite Performance Index for Raw Water Supply Systems: PLS-SEM and Generalized Reduced Gradient

Open

Ussy Andawayanti, Ery Suhartanto, Rahmah Dara Lufira, Hari Siswoyo, Sri Utami Sudiarti, Rizki Ramadhani Pratama

2026 Jurnal Presipitasi Vol. 23 Issue 1 Article Cited by 0 Quartile

Abstract

Reliable raw water service depends on asset condition, institutional capability, and watershed context; existing checklists in Indonesia fail to produce a validated, decision-ready performance score. This study develops a composite performance indicator for raw water infrastructure that incorporates technical (Tk), institutional/non-technical (NT), and environmental (Li) dimensions. Data were collected from 21 schemes in Lombok–Sumbawa, West Nusa Tenggara Province, Indonesia (NTB), with 160 respondents, using field assessments and 1–4 scale questionnaires. Estimated reflective formative partial least squares structural equation modeling was then applied, and GRG calibration was used to minimize deviation from field scores under non-negativity and unit sum constraints for interpretability and portability. All pillars contribute positively and significantly to the composite index, which exhibits high explanatory power (R² = 0.997). The calibrated index is PIRWSS = 0.440 PITk + 0.340 PINT + 0.220 PILi, with SSR ≈ 83.412, RMSE ≈ 0.522, MSE ≈ 5.721, and ≈ 99.70% accuracy relative to field benchmarks. A cross-site analysis shows higher performance in Lombok than in Sumbawa, reflecting hydroclimatic conditions and conveyance configurations. The index provides utilities and regulators with a transparent, reproducible framework for benchmarking and prioritizing operations, maintenance, rehabilitation, and source water protection. © 2026 by Authors,.

Affiliations

Department of Water Resources Engineering, Universitas Brawijaya, Jalan MT. Haryono, Malang, 65145, Indonesia; Department of Civil and Environmental Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, 464-8601, Japan