The Kleinfelder Blog

To support agencies with public water systems in meeting the Environmental Protection Agency (EPA)’s Lead and Copper Rule Revisions (LCRR) and the Lead Service Line Inventory (LSLI) submission requirements, Kleinfelder has adopted the Document Intelligence Platform (DIP). This artificial intelligence (AI)-powered tool, built on Microsoft Azure AI Services, significantly accelerates the data extraction process from record documents and increases productivity, helping to quickly achieve potable water compliance and public safety standards.

The EPA updated their 1991 Lead and Copper Rule (LCR) with the LCRR in 2021 and proposed further changes under the Lead and Copper Rule Improvements (LCRI) draft in 2023. These updates were driven by incidents of compromised potable water, such as with the contaminated water crisis in Flint, Michigan. The LCRR mandates that all community water systems submit an LSLI to catalog all existing lead service lines and thoroughly examine any service lines containing undocumented materials. The EPA plans to achieve full documentation of current water systems by October 16, 2024, and requires the replacement of lead service lines within 10 years of the proposed LCRI.

Kleinfelder’s Boston team began its efforts in the Town of Medway, Massachusetts, training the DIP using a small sample set of typed and handwritten documents to process over 5,500 records, which included tie cards and meter sheets. The analysis quickly indicated the absence of lead, allowing the team to reallocate resources to other aspects of the project. Using the DIP saved considerable time compared to manually reviewing each document individually.

To identify service line material information and collect it in an organized dataset, Kleinfelder employed the DIP to review documentation from Medway’s Department of Public Works and Water Department. The DIP uses Optical Character Recognition (OCR) to convert document images into machine-readable text and custom-trained Microsoft Document Intelligence AI models to understand the documents and identify key information. Kleinfelder experts then parse the text into a standardized dataset for efficient critical analysis. This approach accelerates the process of turning client asset data into knowledge that is usable in delivering an engineered solution.

“The great thing about this technology is that we can now complete record reviews much quicker and reallocate that time and brainpower to developing an effective plan of action,” said Kleinfelder Project Professional Ajay Sharma. “We’ll be looking for trends and mapping it all out so we can see the bigger picture of what’s going on.”

Kleinfelder has since used the DIP to successfully assist other municipalities in the New England area with their LSLI reviews, as well as utility clients across the country with reviewing well over hundreds of thousands of documents under varying scopes. With continued use of the DIP, Kleinfelder demonstrates its commitment to applying cutting-edge technologies within the engineering field and providing new ways to efficiently address emerging challenges.

“It’s the way we’ve been deploying the tool that allows our people’s expertise to really shine. By transforming client data into organized knowledge, we’ve significantly accelerated our problem-solving process,” Kleinfelder Vice President Evan Stark commented. “We bring in multi-disciplinary expertise from across the US that can solve our clients’ specific problems and we can figure out what to do a lot faster with this tool.”