The Asbestos Beneath our Streets

Posted by Lars Stenstedt on Oct 1, 2018 4:26:54 PM


Much of the US and Canada experienced large population growth in the 1950s-1960s. At this time, the use of Asbestos Cement (“AC”) pipe was common for underground infrastructure use. According to an HDR Engineering study, AC pipe can contain as much as 12% asbestos fiber content, and was a popular choice of engineers for potable water, sanitary sewer and storm drain pipelines. AC was touted because of its lightweight and low coefficient of friction. HDR estimates that over 600,000 miles of AC pipe is installed in the US and Canada, with a typical design life of 50 years.

In the US, potable water market alone there are approximately 1 million miles of pipe (of all materials) that need to be replaced in the next 30-50 years – a liability of as much as $1 trillion for the 50,000 water utilities in the US. To date, however, water main systems replacement plans have focused on the oldest pipe with the highest break rate, cast iron. In fact, some water utility capital planning doesn’t include a “water main replacement program.” Instead, they have “cast iron replacement programs.”

While cast iron pipe break rates are high (the 2018 Utah State study showed US break rates for CI pipe at 35/100miles/year, an increase of 43% over the past 6 years), AC pipe break rates are also significant, and are increasing dramatically (the same Utah State study showed US break rates for AC of 10/100miles/year, an increase of 46% over the past 6 years). 

Further, engineers believe that AC pipe fails more catastrophically when it first breaks, while cast iron pipe failures can start small (with just a pin hole leak) and grow gradually to a full break, giving early warning before a catastrophic main break occurs. 

A key first question for AC pipe condition assessment is this - with traditional desktop condition assessment models built substantially on pipe break history, how does a utility find high-risk AC pipes before they break in catastrophic fashion?

A few years back, the American Water Works Association (AWWA) issued the Buried No Longer (BNL) report, which provided average life expectancy for AC pipe across the country. Here on the west coast where Fracta is headquartered, the average life of AC pipe is estimated at 65-105 years, depending on a “long service life” or “short service life” assumption.  

For utilities, this still leaves large questions with major impacts unanswered:

  • Are all my AC pipes in long or short service life environments?
  • What is the range of service life around that assumed average?
  • And most importantly, which individual pipes are at the low end of such assumed range, and which are at the high end so we can properly prioritize replacement or rehabilitation?

In the end, each individual AC pipe could have a useful life from as short as 50 years to as long as 150 years. With this information, how can a utility know which AC pipe needs to be replaced at 50 years and which can be left in service for 150?

There are four basic alternatives for thorough management of AC water mains long term:

  1. 100% pipe replacement. A typical cost for this is $1M/mile and can take 50-100 years.
  2. Engineering-driven, system-wide (all the miles of AC pipe) combination of desktop and physical condition assessment. This can take as much as 2-3 years to execute and cost approximately $3000-5000/mile ($1-1.5M for a 300 mile system).
  3. A pipe-level, detailed combination of desktop and physical condition assessment. This may only take a few months, but costs $30,000-$50,000/mile, and only provides analysis on the specific chosen pipe.
  4. Use aggregated data and machine learning for system-wide condition assessment. A sophisticated model based on Machine Learning provides the granularity of the detailed pipe level analysis in a fraction of the time (just 4-8 weeks) and at a fraction of the cost ($100/mile or less).

To summarize:


For some utilities with manageable amounts of AC pipe, the first approach can make sense – just replace all of the AC pipe. For other utilities with available time, funding and engineering resources, the second approach can make sense. The third alternative is best suited for large, transmission pipes that can’t be allowed to “run to failure.” 

But for most utilities, the fourth approach, the data driven Machine Learning approach, can be a very cost effective approach for a full, system wide analysis – which can then be used to prioritize the AC pipe replacements at the streets with the shortest amount of remaining life. 



  1. ASBESTOS CEMENT PIPE: WHAT IF IT NEEDS TO BE REPLACED? Eric Williams, P.E. Professional Associate/Vice President, HDR Engineering, Inc., Sunset Beach, NC Kent Von Aspern, P.E. Senior Project Manager, HDR Engineering, Inc., Walnut Creek, California
  2. American Water Works Association, Buried No Longer: Confronting America’s Water Infrastructure Challenge.” 
  3. Water Main Break Rates In the USA and Canada: A Comprehensive Study.  Utah State University, March 2018.

Topics: Risk Management

Subscribe for more news & updates from Fracta