Related Experiment Videos
Agentic AI-enhanced digital twins for Smart City civil infrastructure: A secure, autonomous and auditable management
Toqeer Ali Syed1, Ali Akarma1,2, Ali Alatify3
1AI Center, Faculty of Computer and Information Systems, Islamic University of Madinah, Madinah, Saudi Arabia.
Plos One
|July 17, 2026
Summary
This study introduces an AI-powered Digital Twin framework for smart city infrastructure, significantly reducing detection latency and improving mitigation success. The system enhances accountability through blockchain integration for auditable interventions.
Area of Science:
- Artificial Intelligence
- Civil Engineering
- Computer Science
Background:
- Smart city initiatives face challenges in bridging the gap between anomaly detection and effective intervention in civil infrastructure.
- Existing systems lack integrated monitoring, action, and auditability for infrastructure management.
Purpose of the Study:
- To propose and evaluate an Agentic AI-supported Digital Twin framework for smart city civil infrastructure management.
- To enhance intervention safety, timeliness, and accountability through an integrated, auditable system.
Main Methods:
- Developed a Digital Twin framework integrating multi-stream telemetry for asset and network modeling (bridges, roads, water infrastructure).
- Implemented an agent-based Perception-Conceptualization-Action workflow using LangChain and LangGraph for cross-domain reasoning and mitigation planning.
- Utilized a permissioned blockchain to ensure provenance, governance, and tamper evidence for observations and interventions.
- Conducted 18,000 incident simulations across various configurations and complexity levels to evaluate framework performance.
Main Results:
- The agentic system demonstrated a significant reduction in mean detection latency (3,197s vs. 39,374s) compared to a rule-based baseline.
- Mitigation success rate improved to 66.2% from 45.5%, with 71.8% blockchain-anchored decision justification.
- Operator workload was reduced by 91.7% in the agentic system.
- Ablation analysis confirmed multi-agent orchestration drives latency and mitigation gains, while blockchain ensures auditability.
Conclusions:
- The proposed framework measurably improves response efficiency, recommendation quality, and action accountability in smart city infrastructure management.
- Combining simulation-enabled digital twins with governance-aware agentic orchestration offers a robust solution for operational gaps.
- The study highlights the potential of AI and blockchain for creating safer, more responsive, and accountable smart city systems.
