Failure Probability Analysis of Small Ductile DI Pipe

Posted by Paul Carpenter on Nov 9, 2018 7:00:00 AM


Purdue University and Louisiana Tech University partnered with the Water Research Foundation to study the “Practical Condition Assessment and Failure Probability Analysis of Small Diameter Ductile Iron Pipe.” This effort interviewed and surveyed the water pipe industry, as well as utility and condition assessment practitioners. DI pipe represents a little less than one third of the potable waterpipe market in the United States and the focus was on small diameter DI pipe less than 12 inches.

The research indicated that corrosion-related failures in DI pipe are one of the main concerns of many water utilities, considering that failures lead to higher costs and water losses. A key factor contributing to failures is that DI pipes with a diameter less than or equal to 12 inches have a thinner pipe wall thickness than the larger diameter DI pipe. This significant difference means that the small diameter DI pipe can be considered a different cohort and should not be rolled into the analysis of larger diameter DI pipe, nor should the characteristics of large diameter DI pipe be applied to DI pipe 12 inches or a less.

AWWA Specification Thickness Reductions 

The findings concluded with as much as a 75 percent difference in wall thickness the time taken for corrosion to penetrate DI pipe is reduced, contributing to their shorter useful life, as compared to CI pipes. It also concluded that the increased number of small diameter breaks could make the cost comparable to large diameter pipes with a higher consequence of failure.

The survey concluded that the average total number of failures per year per 100 miles of small diameter DIP was 15.1 failures/year/100 miles with the average 10-year direct cost of failure was approximately $12,600 per occurrence and the average indirect cost was approximately $5,600 per occurrence.

For water utilities with small diameter DI pipe, the typical model of using an age-based approach no longer works. Neither does applying the decay curve from CI pipe analysis.

As a result, a fast, accurate and low cost method of Condition Assessment (CA) is needed. Many traditional CA techniques and methodologies as part of an overall asset management program can be expensive and time consuming.

But when asset management practices and condition assessment activities are infused with Machine Learning for water pipe, a new solution of aligning maintenance, repair and replacement strategies to better allocate limited resources can be discovered.

Machine Learning provides a fast, accurate and cost-effective approach for condition assessment tools to help address the water industry’s most pressing concerns of producing an accurate prediction of water main Likelihood of Failure (LoF). Fracta produces an accurate five year model using Machine Learning to help direct leak detection efforts, focus preventative maintenance crews, validate capital plans and align master planning efforts.

Fracta’s Machine Learning algorithms take into consideration hundreds of variables contributing to thinner walled pipe and does not rely on a purely age-based analysis of small diameter DI pipes.

Data acquisition, assessment and cleaning for any Machine Learning process is roughly 60-80 percent of the work, also known as pre-processing or data wrangling, with the remaining percentage being the Machine Learning itself. Once the data is assessed, cleaned and imputed where needed, it is ready to be fed into a machine learning model where it is subsequently ‘trained’ to learn the patterns of the small diameter DI pipe that predict breakage events.

The more data a model contains, the more robust the model. As utilities are constantly collecting data such as new breaks, that data can continually be fed into a machine learning model. New pipe data strengthens the predictive power of a Machine Learning algorithm. Machine Learning can also benefit utilities with limited asset or breakage data by filling in the gaps from an extensive database.

Fracta provides the LoF analysis so water utilities can get ahead of the risks associated with smaller DI pipe issues. This can help in preventing costly breaks that have direct and indirect costs of over $18,000 per failure.

Topics: Condition Assessment, Leak Detection

Subscribe for more news & updates from Fracta