January 2026 Vol. 81 No. 1
Features
The digital transformation of underground infrastructure monitoring
Gordon Feller, Contributing Editor
Over the past few years, Silicon Valley labs and leading tech companies have developed innovations in the remote monitoring of underground infrastructure and pioneering new systems rooted in digital tools – especially using AI, machine learning (ML), the Internet of Things (IoT) and advanced sensor-based technologies. These advances could fundamentally change how cities, utilities and industrial operators manage the hidden network of pipes, tunnels and systems running below our communities.
The oldest challenge in remote underground infrastructure monitoring has been spotting issues before they erupt into environmental disasters, structural failures or public-health crises. The emergence of AI-enabled systems, such as PreView and SewerAI, demonstrates how computer vision and ML algorithms have shifted monitoring from passive, periodic inspection to active, real-time surveillance.
New cameras and sensors collect continuous visual and quantitative data streams. Advanced algorithms analyze these to flag anomalies – debris, illegal dumping, leaks, cracks, blockages – that human inspectors might miss. Notably, automated defect detection means managers receive actionable alerts within moments, allowing for timely intervention and preventive maintenance – long before catastrophic events unfold.
IoT devices and wireless sensor networks have proliferated across underground systems, providing eyes and ears deep below ground. Devices can track moisture, pressure, temperature, vibration, chemical composition, corrosion and more, transmitting pulses of data back to cloud-based platforms or local controllers.
Modern systems are designed for seamless communication and integration; many of these are designed expressly to support edge-computing. They run basic ML tasks locally to reduce data transmission and enable faster responses. These sensors work collaboratively, creating a digital twin – a living, up-to-date model of buried assets and environmental conditions.
Long-range fiber sensing
One major innovation emerging out of academic partnerships and Silicon Valley is long-range fiber-optic sensing technology. These systems use optical fibers to detect subtle vibrations and changes in the subsurface across vast distances. Unlike traditional point-based sensors, fiber-optic cables allow continuous, high-resolution monitoring over 6 miles (10 kilometers) or more, ideal for critical infrastructure like pipelines, levees and tunnels.
For example, at the UC Berkeley Richmond Field Station, collaborations with Japanese firm Kajima have demonstrated fiber-based distributed sensing capable of mapping S-wave velocities and identifying ground instabilities – crucial for disaster preparedness and post-event assessment.
Data collection alone is not enough. Silicon Valley start-ups and established players have heavily invested in advanced visualization and geospatial analytics platforms. These digital dashboards aggregate and harmonize information from AI-driven image processing, IoT sensors and fiber-optic cable monitors.
Using these unified platforms, operators can view real-time infrastructure health status, drill into detailed reports and simulate what-if scenarios using predictive analytics. This potent blend of big data and ML enables a move from reactive repairs toward predictive, risk-based maintenance – optimizing asset lifespans and budgets.
Modern monitoring systems now integrate directly with GIS and Computerized Maintenance Management Systems (CMMS). This connectivity ensures that defect locations, risk scores and recommended interventions are instantly mapped against utility layouts, work orders and historical records. It also facilitates faster regulatory compliance reporting and more intelligent work scheduling across distributed teams and contractors.
The result is greater operational efficiency and a consistent framework for decision-making – capabilities highlighted as industry standards by companies like SewerAI.
Portability, modularity, and non-disruptive deployment have become key goals for engineers. Recent technologies emphasize the high value of both portability and field-upgradability. The lightweight TREKK360 system, for example, weighs only 8 pounds and uses off-the-shelf 360-degree cameras that can be easily replaced or upgraded in the field.
This design principle minimizes system downtime and deployment costs, while improving safety, by reducing the need for dangerous confined-space entry. These modular systems allow crews to perform significantly more inspections and even extend monitoring to unexpected applications, such as industrial tank inspections.
Predictive maintenance
Applications that enable predictive maintenance and risk mitigation are now emerging – and they could be game-changers. AI, IoT and sensor-based advances are especially valuable as North American cities confront aging infrastructure, workforce retirements, and tight budgets. A 2025 Black & Veatch survey found nearly 60 percent of utilities now prioritize asset monitoring and management as their top digital investment.
However, most still lag in harnessing existing data effectively. Adopting new AI-powered platforms is a strategic solution to “do more with less,” transforming infrastructure management from reactive troubleshooting to proactive care.
The use of powerful 3D modeling software and project delivery toolkits has become increasingly popular for underground infrastructure planning and monitoring. These platforms not only visualize subterranean networks but also virtually simulate construction conflicts, enabling spatially contained urban projects to proceed faster, safer, and more cost-effectively. The interconnection of AI-driven monitoring with digital twins – complete, real-time replicas of physical systems – facilitates what-if analyses, risk simulations and long-term planning.
The shift to digital remote monitoring yields dramatic improvements in project risk management, overall efficiency, and public safety. For example, advanced sensor systems like Sika Sensor's DuraMon and CiDRA SmartHatch offer real-time data and predictive analytics for corrosion detection and concrete quality monitoring. These tools make proactive interventions possible before safety compromises occur, extending lifespans and lowering costs.
Despite immense progress, challenges remain. Data quality and accessibility – especially for older assets – must improve for AI/ML models to reach their full potential. Collaboration between private sector innovators, consultants, government agencies and research institutions is vital to building comprehensive repositories and driving interoperability. Funding gaps also persist, with the 2025 ASCE Infrastructure Report Card rating U.S. wastewater systems at D+, underscoring urgent investment needs in digital transformation.
The next few years will likely see wider adoption of these integrated platforms, advances in edge computing for underground monitoring, new applications of quantum sensors and photonic technologies, and increased use of remote monitoring in climate resilience, disaster response and smart city initiatives. Silicon Valley’s leadership in software, analytics and IoT continues to push the field forward, ensuring that underground infrastructure is managed with unprecedented intelligence and reliability.
The digital transformation of underground infrastructure monitoring is now taking giant leaps, thanks to a host of innovative solutions: AI/ML-powered defect detection, ubiquitous IoT sensor networks, fiber-optic distributed sensing, real-time data visualization platforms and seamless integration with GIS/CMMS systems. Tech labs inside universities and companies are increasingly focused on innovations that can enable smarter, safer and more resilient infrastructure.
FOR MORE INFORMATION:
Sewer AI, 1 (855) 952 2200, sewerai.com/contact-us
TREKK360, (816) 313-8539, trekkllc.com
Sika DuraMon, sika.com/en/solutions-for-projects/water.html
CiDRA SmartHatch, cidraconcrete.com
Comments