WEAO 07 Conference:INFORMATION DRIVEN ASSET MANAGEMENT – LEVERAGING

This paper by Ross Homeniuk, UMA | AECOM and Hesham Osman, UMA | AECOM was the only interesting paper out of a whole day in the Asset Management track at London’s WEAO conference.

The paper uses NRC CCTV inspection codes (1 to 5 where 5 is collapse) to set up a failure rate model using Markov chains. The result when you recognize that about 3% to 6% of a population of pipes actually get worse enough to move up in the defect assessment codes is that some wastewater pipes have a statistical probability of reaching a useful life of 250 years.

This is a very interesting conclusion because it provides a mathematical method that verifies something that I’ve suspected for a long time. Ever since, that tour of a Roman aqeduct that was built 2000 years ago and would still function today if put into service. Indeed there are examples in the Middle East of infrastruture that is hundreds and thousands of years old that is still in service.

The difficulty is that the analysis of the markov chain only provides a decay curve for pipes that progress to a collapse failure. What about the class of pipe that decays but does not progress to a collapse failure? (Most of the sample data is based on the City of Winnepeg infrastructure). This analysis does not include those pipes that are considered to have failed because they leak like a sieve and cause downstream flooding problems. This a class of pipe failure that is related to an I&I problem. It may not result in a complete pipe replacement but in an in situ rehabilitation technology; however, the result of incurring a capital expense that is a betterman, (extension of service life) is the result.

So I suspect that the analysis may have be a bit too uniform in its treatment of pipe failure types. There may be other ways that a pipe can fail that do not fit in this analysis.  A really interesting bit of work though and very well presented. Kudos to the authors.

Discovering a pin hole leak

Our children’s caregiver lives with us during the week and last night she came upstairs to say that she could hear water dripping. We have an old house and this is exactly the kind of thing that gives me palpatations when I hear it.

Investigation revealed that we did have a water leak but I could not find from where. Two years ago, we had a pipe freeze and burst which revealed itself with a light fixture filling up with water. We were lucky that we were home and caught that one when we did.

Thankfully, this time the old line had an isolation valve on it and so with relative ease I was able to close the valve and confirm that the dripping was slowing down. So I had the right pipe at least. I turned the line back on and climbed up on a chair and reached in towards the pipe and found a pin hole leak in one of the elbows. Wow, what causes that to show up?

I’ve read some of the reports that have associated excess flux on copper pipe soldering with pin hole leaks and maybe that’s what’s going on here. It’s not an easy failure to diagnose as compared to say the burst elbow from pipe freezing. It looked like someone had shot a bullet out of the elbow from inside. The pin hole leak for this elbow was on the top outside part of the elbow and I believe the flux leakage corrosion points were all along the invert or lower portion of copper pipe and not the upper side.

It’s in a tricky position to fix but that’s what my afternoon is going to look like.

Debt-ridden M.W.R.A. to hike up water rates over next five years

HOLBROOK (AP) — A new survey says thousands of residents south of Boston will be faced with water and sewer rate increases of more than 40 percent over the next five years.

story is here.

Engineers, IT and robustness

From Spit and Bailing Wire comes a very nice little observation on why it is so tough to introduce IT frameworks to engineers.

Gotta love that title, eh. I have to say that I’ve seen this more than a few times in my own engineering career and it has lead to a great dislike of spreadsheets as a technique for problem solving. Spreadsheets lead to quick and dirty analysis that is almost impossible to check. The spreadsheet, I suspect, is a leading cause in the decline of peer review amongst engineers. When I began my engineering career, (now about 20 years ago), peer review of ones calculations was a common practice. The problem analysis was formulated on paper, you cited the problem statement and used references to specific texts for the empirical formulae being used in your solution. Anyone could follow your solution to its conclusion. Not so with a spreadsheet.

Yes, I have solutions to the spreadsheet problem and I have used them myself but “Try to get them to use one…I dare you.”