What is unknown management, exactly?
Water service lines (also called pipes) are made of different materials, from copper to PVC to lead. These lead service lines (LSLs) are old and pose a major health risk, especially for children and the elderly.
But most water systems have not kept complete historical records that show the material each pipe is made of. This leaves these water systems with service lines of unknown material, often simply called ‘unknowns.’ Identifying the correct material each of these service lines is made of is called unknown management.
There are essentially 2 ways to approach unknown management. First, a water system can dig up every unknown line to identify and record the materials. However, this approach is too time and resource intensive for most water systems, and particularly those with a lot of unknowns. For most water systems, predictive modeling or statistical analysis offer an opportunity to quickly, accurately, and affordably identify LSLs.
But why does this even matter? Unknown management is important for 2 main reasons: LSL replacement and compliance.
LSL replacement is critical for public health and water infrastructure modernization. And unknown management is the foundation for LSL replacement. Without a clear understanding of the exact volume and location of every lead pipe, it is impossible for water systems to successfully plan for, fund, and replace lead pipes. As of today, there are more than 9 million unknowns across the United States. Some water systems still don’t even know if they have lead pipes in their system or not!
Beyond the public health and modern infrastructure needs, water systems also face multiple compliance requirements for unknown management and LSL replacement via the Lead and Copper Rule Improvements (LCRI). There are stringent annual customer notification requirements, as well as upcoming reporting requirements, for service lines made of lead or unknown materials. And the LCRI requires identification of all unknowns, as well as replacement of all lead pipes, in a 10 year timeframe, from 2027-2037.
These public health and regulatory pressures, along with the need for replacement efficiency and infrastructure modernization, are creating a sense of urgency for water systems to quickly and accurately identify the materials of their service lines currently listed as unknown. With the LCRI officially going into effect in 2027, water systems must start classifying their unknowns as soon as possible.
Thankfully, technology is on hand to help, with new AI and software tools available to support unknown management and LSL replacement efficiency. Software breaks down data silos, updates dynamically as data changes, and gives water systems access to best-in-class operational, planning, and data analytics support tools. With tightening budgets and an aging workforce, software tools are the best option for water systems to stay ahead of unknown management, LSL replacement, and other key system needs.
In this blog post, we’ll break down the role software plays in unknown management, key features and benefits of using software for unknown management, and getting started with the right solution.
Key features of software for unknown management
Not all software is created equal and finding the best software solution to identify, manage, and classify unknown service lines is critical. While water systems may have some differing needs, we will outline a few key features to look for in a software solution.
Data centralization
Streamlined management means having your data in one place, rather than disparate silos. The best software tools make sure you have all relevant data at your fingertips by integrating multiple data sources into a single location for use.
Dynamic integrations
Water systems already use a lot of software tools in their day-to-day operations. Data shouldn’t just be centralized, it should be dynamically integrated to ensure you’re always looking at the most up-to-date information. For example, if you’re service line material data is connected to your work order management system, you can easily ask workers to record service line materials when they encounter them in the field. When this process happens automatically, it is an easy way to reduce your unknowns with no extra time or cost needed.
Integration of predictive modeling and statistical methods
For water systems with a lot of unknowns, it is not cost or time efficient to physically verify the service line materials for each one. Software tools that include predictive modeling or statistical assessment offer another accurate and significantly more efficient option for unknown management.
When it comes to predictive modeling, the best software tools enrich the baseline dataset and provide high-accuracy tailored models based on the system’s unique data with minimal requirements for digging. For example, BlueConduit predictive modeling customers save 95% compared to physical verification, typically only requiring ~100 (or less) physical inspections to enable lead likelihood predictions.
Dashboards and reports
Effectively managing your unknowns means having detailed information easily on hand to assess progress, lead lines identified, outstanding unknowns, and other metrics. Integration within existing systems, such as Esri ArcGIS or CityWorks, is also key. For example, BlueConduit’s predictive modeling solution is integrated with the Esri LSLI solution, including Standardized Inventory Data Management, BlueConduit Inspection Manager, (including recommended inspection lists), and the BlueConduit Predictions Dashboard.
State-specific inventory templates
The LCRI’s unknown management requirement includes annual service line material inventory updates to be submitted to the state regulatory agencies (who, in turn, submit that data to the EPA). A high quality unknown management tool must not only track remaining unknowns and newly classified service lines, it must also provide that data in a state-compliant inventory format. With unique formats for each state, this is a critical need for software solutions; without this feature, you’ll be forced to spend manual time understanding the nuances of state guidance and adjusting your data to meet state-specific needs.
Benefits of using software for unknown management
Water systems reap many benefits when they use software for unknown management vs manual tracking and/or physically verifying every unknown line. Let’s break down these benefits further.
Cost efficiency
It’s no secret that manual management of unknowns and physically verifying each unknown line can cost upwards of thousands of dollars per line in time, labor and materials. Multiply that by thousands of unknown lines and water systems are easily facing millions of dollars to physically verify each line. Using software and predictive modeling to manage unknowns drastically reduces this cost – by using algorithms and statistical methods to classify large quantities of unknown lines at once – water systems save time and money.
Meeting regulatory compliance
While state guidance on exact methodologies and requirements varies, most states accept predictive modeling and statistical methods for the purposes of inventory classification. A full understanding of the number of lead service lines in the system leads to better planning for replacements and any customer notifications. Software enables dynamic predictive modeling and eases the process of regulatory compliance.
Actionable and informed decision making
Using software to manage and classify unknown service lines gives water systems the ability to make key decisions about replacement planning and resource allocation more effectively than simply going address-by-address to verify and potentially replace lines. Software should be able to provide insights not just at the address level, but at the neighborhood or census block level for more effective replacement planning. Water systems will need to coordinate with other agencies for LSL replacements and there are socioeconomic, environmental, demographic, political, and infrastructure challenges to consider when it comes to identifying unknown service lines.
Public trust and transparency
People like things they can see. Software makes it easy to communicate and explain your plans for unknown identification and replacement (where necessary), which gives the public confidence and trust in their water system. In addition, there are LCRI compliance requirements around communication strategies as it relates to replacement planning. Having a full picture of the scope of unknowns to identify and replacements to complete helps with communicating these plans to the public.
Scalability and future-readiness
Water infrastructure consists of more than just service lines. The ideal software solution will allow for scalability into other areas, such as water mains, connectors or valves where there is opportunity for improved data insight and more efficient, optimized planning.
A real-world example of using software to manage unknowns
Rather than just talking about how important and beneficial software is for unknown management, let’s see it in action. For Des Moines Water Works, using predictive modeling from BlueConduit enabled them to quickly classify and reduce their unknowns by more than 75%.
Customer Overview
The utility (DMWW), founded in 1871, serves Des Moines as well as the surrounding suburbs, for a total of about 500,000 people. It is publicly owned and governed by a board of trustees. It’s also unique in that its customers own the entire service line; that is, the pipe leading from the house to the box, and also from the box to the curb.
The Problem
When they began to compile a service line inventory, DMWW had no data on service line materials. This meant that leading up to the Oct. 16, 2024 LCRR inventory deadline nearly their entire system – almost 100,000 service lines – would be classified as “unknown.” As a result, the utility was facing a particularly heavy burden to meet the updated LCRI compliance requirements around identification and replacement as well as the annual customer notifications still in effect from LCRR.
DMWW lacked the staff and resources needed to dig up every line, and they knew that lead was more likely to be found under the street than in the house. They needed a fast, accurate way to classify the lines in their inventory, with minimal information from their existing records.
The Solution
DMWW decided to use predictive modeling to help manage unknowns and selected BlueConduit’s LSL Predictions solution for this work. Since 2016, BlueConduit has developed innovative technology tools and processes to use AI and machine learning to help utilities and engineering firms predict the likely volume and locations of lead service lines.
BlueConduit began working with Des Moines in January 2024. At that time, DMWW’s initial inventory was almost entirely classified as unknown, with only 2,000 lines being verified. After reviewing these existing datasets, BlueConduit compiled a list of recommended field inspection sites to achieve a representative, randomized sample for the model.
Des Moines conducted their field inspections in two batches, with BlueConduit’s LSL Predictions tool providing model-generated predictions after each batch. BlueConduit data scientists evaluated DMWW’s models across a range of data science metrics including precision, recall, specificity and false negative rate to recommend thresholds, or limits, to apply to all lines that fell under that threshold. For example, any lines greater than 70 percent likelihood of being lead would be classified in one category (e.g., “lead,” or “require physical inspection due to high likelihood of lead”). This process helped the water system determine which service lines to count as lead, galvanized steel or non-lead in the inventory.
The Results
These thresholds helped DMWW classify more than 70,000 service lines – eliminating their unknowns by nearly 75% in about six months.
Customer Testimonial
“This entire process has helped us realize that we really need the data to tell the story,” said Jenny Puffer, Director of Water Distribution at DMWW. “It’s one thing to think you know where the lead is, to have a gut feeling, but it’s completely different to actually see the results in the model and see the data and the best practices behind the decisions we have to make as a water system. We are trying to do what’s in the best interest of our customers and our community, and do it quickly but effectively. BlueConduit has really helped us to do that.”
Getting started with the right software management solution
In order to use software to effectively manage your unknowns (and lead service line replacement program, if you’ve got lead/GRR), there are a few things that need to be considered first.
Assessment of needs
There’s no one-size-fits-all solution for unknown management. When assessing your water system or engineering firm’s need for unknown management software, several key factors must be considered. First, evaluate the budget and cost-effectiveness of the software in comparison to manual methods or physical verification. Second, determine the number of unknowns and the resources required to manage them. Third, assess the software’s ability to classify unknowns according to state regulatory and inventory requirements. Fourth, consider how the software tool integrates into the existing data and asset management ecosystem. Finally, explore any additional features that might be needed, such as predictive modeling, replacement prioritization and planning support, reporting, or stakeholder communication support.
Key considerations
Utilizing predictive modeling software for unknown management start with ensuring high accuracy predictions. The software must be able to deliver reliable results, minimizing false positives and negatives, to avoid unnecessary costs and disruptions. Integration with existing tools, such as GIS systems or work order management software, is crucial for streamlining workflows and data sharing. Furthermore, the unknown management approach must directly connect with and inform the LSL replacement planning process. Predictive models should identify high-risk areas and enable you to prioritize replacements based on the likelihood of lead presence, demographics, presence of children, and other relevant factors. By aligning unknown management with replacement strategies, water systems can optimize resources, expedite LSL removal, and improve public health outcomes efficiently.
Implementation timing
Keep in mind that using predictive modeling software for unknown management saves time, but still takes a little time. Water systems must plan backward to ensure they allocate enough time for initial system setup and the necessary representative inspections. While predictive modeling drastically reduces the number of physical digs compared to manual verification, some targeted inspections are still essential to train the model and guarantee high-accuracy results. Properly conducted inspections provide crucial data that enhances the model’s reliability, allowing for confident classifications and informed LSL replacement planning. Rushing this process can lead to inaccurate predictions and undermine the effectiveness of the entire initiative.
Conclusion
Water systems face a number of challenges when it comes to service lines classified as unknown in their inventory, including LCRI compliance requirements and ensuring effective (and compliant) replacement plans are in place. Software and statistical modeling to identify and replace these unknown lines is the most cost effective, efficient approach to unknown management, giving communities and customers confidence and trust in their water system.
For more information on our software solutions and to see predictive modeling in action for yourself, contact BlueConduit. Our team are the leading experts in using data science and predictive modeling to help water systems efficiently locate and remove lead service lines.
You can also visit our resources page to read more case studies, blogs, and helpful guides to managing unknown service lines.
