Explainable AI with dual-layer validation framework for ensuring noise resilience and physical consistency in low-cost sustainable water quality monitoring
- 1Department of Informatics Engineering, Universitas Muria Kudus
Abstract
Background Ensuring sustainable water management (SDG 6) requires reliable IoT-based sensor networks, yet low-cost deployments are often hindered by environmental noise and the opaque nature of complex AI models. This research aims to develop a robust contaminant classification framework that prioritizes actionable transparency and physical consistency in environmental monitoring. Methods Time-series water quality data were analyzed using Random Forest (RF), LSTM, and Transformer models. Beyond standard performance metrics (Weighted F1-Score), this study introduces a Dual-Layer Validation (DLV) framework powered by SHAP-based Explainable AI (XAI) to verify the models’ alignment with hydrogeochemical principles. Results The RF model achieved the highest performance (F1 = 0.872), outperforming LSTM (0.852) and Transformer (0.817) (p < 0.05). Crucially, the XAI analysis revealed a strong cross-architecture consensus: both ensemble and attention-based models independently identified Turbidity and pH as the principal eco-chemical drivers of contamination. Conclusion The DLV framework demonstrates that transparent models can effectively learn genuine physical and hydrogeochemical fingerprints rather than statistical artifacts. This cross-validated resilience to sensor noise offers a critical pathway toward sustainable, trustworthy, and cost-effective IoT water quality management systems.
Related
- Retention forestry practices in India for biodiversity conservation, climate resilience, and sustainable forest management across diverse forest zones
- AI-driven environmental performance index using deep autoencoders for monitoring sustainable development goals
- Real-time intelligence framework for microplastic monitoring and coastal resilience
- Explainable spatiotemporal Graph Neural Networks: Multi-City PM2.5 forecasting in Bangladesh with geographic graph topology validation
- PhysMetrics.Weather: An Evaluation Framework for Physical Consistency in ML Weather Models
BibTeX
@article{sugiharto2026explainable,
title = {{Explainable AI with dual-layer validation framework for ensuring noise resilience and physical consistency in low-cost sustainable water quality monitoring}},
author = {Wibowo Harry Sugiharto and Rizkysari Meimaharani and Ahmad Abdul Chamid and Muhammad Imam Ghozali and Alif Catur Murti},
journal = {Discover Sustainability},
year = {2026},
month = jul,
doi = {10.1007/s43621-026-04210-9},
url = {https://doi.org/10.1007/s43621-026-04210-9},
note = {CC BY},
}
RIS
TY - JOUR
TI - Explainable AI with dual-layer validation framework for ensuring noise resilience and physical consistency in low-cost sustainable water quality monitoring
AU - Wibowo Harry Sugiharto
AU - Rizkysari Meimaharani
AU - Ahmad Abdul Chamid
AU - Muhammad Imam Ghozali
AU - Alif Catur Murti
PY - 2026
DA - 2026/07/01
JO - Discover Sustainability
DO - 10.1007/s43621-026-04210-9
UR - https://doi.org/10.1007/s43621-026-04210-9
AB - Background Ensuring sustainable water management (SDG 6) requires reliable IoT-based sensor networks, yet low-cost deployments are often hindered by environmental noise and the opaque nature of complex AI models. This research aims to develop a robust contaminant classification framework that prioritizes actionable transparency and physical consistency in environmental monitoring. Methods Time-series water quality data were analyzed using Random Forest (RF), LSTM, and Transformer models. Beyond standard performance metrics (Weighted F1-Score), this study introduces a Dual-Layer Validation (DLV) framework powered by SHAP-based Explainable AI (XAI) to verify the models’ alignment with hydrogeochemical principles. Results The RF model achieved the highest performance (F1 = 0.872), outperforming LSTM (0.852) and Transformer (0.817) (p < 0.05). Crucially, the XAI analysis revealed a strong cross-architecture consensus: both ensemble and attention-based models independently identified Turbidity and pH as the principal eco-chemical drivers of contamination. Conclusion The DLV framework demonstrates that transparent models can effectively learn genuine physical and hydrogeochemical fingerprints rather than statistical artifacts. This cross-validated resilience to sensor noise offers a critical pathway toward sustainable, trustworthy, and cost-effective IoT water quality management systems.
SN - 2662-9984
ER -