Leveraging AI for Sewer Inspection QA/QC: A Houston Case Study
Eric Sullivan of SewerAI writes for the NASSCO Infrastructure Condition Assessment Committee in Trenchless Technology magazine, detailing how the City of Houston used AI to achieve a 55% reduction in contractor data submittal failures and over $1 million in cumulative savings.
Originally published in Trenchless Technology magazine, April 2025, as part of the NASSCO Report — Infrastructure Condition Assessment Committee, by Eric Sullivan of SewerAI.
The Scale of Houston's Wastewater Challenge
The City of Houston faces one of the most formidable wastewater infrastructure challenges in the United States. Serving 2.2 million residents across a vast urban footprint, the city operates more than 6,000 miles of pipes, 133,000 manholes, and 39 treatment plants. Maintaining this system while meeting aggressive federal compliance targets demands not just resources — it demands innovation.
Houston's flat terrain and exposure to frequent tropical storms have historically strained its wastewater system, contributing to sanitary sewer overflows (SSOs) during heavy rain events. In response, the EPA and the State of Texas issued a federal consent decree in 2021, requiring the city to undertake extensive infrastructure upgrades — including repairing or replacing 150 miles of sewer pipes annually and cleaning 275 miles of small-diameter pipes in high-priority areas.
Turning to AI for Condition Assessment
Houston began integrating AI into its condition assessment processes in 2021. By June 2023, the city had significantly expanded its use of AI-powered tools to streamline quality assurance and quality control (QA/QC) for sewer inspections. The system combines computer vision, cloud-based data management, and inspection software to evaluate CCTV inspections against project specifications — automatically and at scale.
Rather than relying on manual review of thousands of inspection videos, Houston's AI QA/QC platform performs comprehensive automated checks, flags issues in real time, and enables faster, more accurate contractor invoicing. The result is a smarter, more efficient pipeline from field inspection to data-driven decision-making.
Key Features of the AI QA/QC System
The AI-powered QA/QC platform deployed by Houston incorporates several capabilities designed to improve inspection data quality and operational efficiency:
- Comprehensive Automated Checks: The system validates NASSCO PACP survey headers, verifies operator certifications, and checks crawler distance counters, camera speed, and video playback compliance.
- Customizable Parameters: Users can define 'hard' and 'soft' no-pass criteria, including specific PACP condition codes, to tailor the QA/QC process to project requirements.
- Optical Quality Analysis: The platform assesses video clarity and verifies compliance with standards for camera speed and water levels, ensuring inspection footage meets quality thresholds.
- Dashboard Insights: A real-time dashboard provides continuous updates on inspection status, corrections needed, and quality issues — giving project managers immediate visibility into field performance.
- Collaborative Annotations: Stakeholders can be tagged with notes to evaluate edge cases, improving transparency and communication across teams.
- Automated Critical Defect Alerts: Following AI review by PACP-certified individuals, the system automatically issues alerts for major defects such as pipe collapses, enabling rapid response to urgent infrastructure issues.
Measurable Results: 55% Fewer Failures, $1M+ in Savings
The impact of Houston's AI-powered QA/QC program has been substantial. An analysis of 28,000 CCTV inspections — drawn from CSV data exported from the PIONEER platform over a one-year period — revealed a 55% reduction in contractor data submittal failures. Monthly failed surveys dropped from nearly 50 to an average of just 22, reflecting both improved contractor performance and significantly greater data accuracy.
Beyond data quality, the AI-powered process has driven meaningful operational and financial gains. Overhead costs have decreased, contractor invoicing has become more accurate, and accounts payable processes have accelerated. Cumulatively, these improvements have yielded more than $1 million in savings for the city.
In Their Own Words
"The data is readily available both in the app & via an API, making it easier for operations to make critical decisions required to maintain the collection system. [This system] reduced the internal labor hours needed to code & review, & the data is more precise."
— Gregy Eyerly, Senior Assistant Director, Houston Public Works
Part of a Broader Modernization Strategy
AI-powered QA/QC is just one component of Houston's broader effort to modernize its wastewater management infrastructure. The city has also deployed digital monitoring sensors throughout its collection system and is exploring additional technologies, including 360-degree cameras for manhole inspections. Together, these initiatives reflect a commitment to data-driven operations at every level of the system.
Houston's experience demonstrates that AI is not just a theoretical solution for large-scale infrastructure challenges — it is a practical, proven tool delivering measurable results in the field today.
A Model for Municipalities Nationwide
As cities across the country grapple with aging infrastructure, federal compliance mandates, and constrained budgets, Houston's approach offers a compelling blueprint. By embracing AI-driven condition assessment and QA/QC, municipalities can achieve compliance more efficiently, reduce costs, improve contractor accountability, and build long-term resilience into their wastewater systems.
The technology is available. The results are proven. Houston's story shows what's possible when a city commits to innovation in service of its infrastructure — and its residents.
