1. Leveraging Advanced Meters For Leak Detection
- Renee Davis, Fort Collins Utilities

2. Single family residential Advanced meter infrastructure (AMI) tools
Intro
Single family residential
Advanced meter infrastructure (AMI) tools
Customer service
More to come as presentation develops…

3. What we know about leaks – end use data
Big leaks matter – 2% of homes leaked at 120 gal/day
AMI can help us find those leaks
Courtesy of Aquacraft, Inc.

4. How we’re using advanced meters to find leaking homes:
Query reports list of homes without zero use.
Includes customer data such as account number and contact information
Ranked by average hourly use
We contact only customers who average more than 60 gal/hour.

5. Finding and fixing leaks is their responsibility.
Making the phone call
What we say:
What we convey:
We see something odd so we’re giving you a call.
It’s a service.
We’re here to help.
It’s all carrot, no stick.
Let’s talk about the meter.
Data is how we’re going to help.
What do you think?
Get to sleuthing.
Finding and fixing leaks is their responsibility.

6. Toilet handle stuck What we’re seeing
This is the daily use from a home that had stuck toilet handle. Each bar represents an hour of data. The flow is 150 to 200 gal/hour (the saw tooth is due to meter resolution).
This ran for a few days. Because this house was otherwise unoccupied, we can get a clear measure of how much water was lost: 7kgal. A second incident occurred where the handle jammed again.

7. Toilet handle stuck What we’re seeing
This is the daily use from a home that had stuck toilet handle. Each bar represents an hour of data. The flow is 150 to 200 gal/hour (the saw tooth is due to meter resolution).
This ran for a few days. Because this house was otherwise unoccupied, we can get a clear measure of how much water was lost: 7kgal. A second incident occurred where the handle jammed again.

8. Home inspector left water running
What we’re seeing
This is the daily use from a home that was sitting empty waiting to be sold. No one was living here. The time it started correlated to a home inspection by a prospective buyer. It was shut off by the time the home owner got our call and got to the house.
The peak flow use was 8500 gal/day for a total of about 26kgal. It works out to just about 6 gal/min so my guess is it was a bath tub.
This homeowner was understandably upset and worried about his bill. We gave him an estimate of how much this amount of water would cost. That information seemed to help too.

9. Broken sprinkler line
What we’re seeing

10. 1262 customers contacted since start.
Program reach
1262 customers contacted since start.
Typically we contact about 7 customers per day in the summer and about 4 in the winter.
A typical day has 96 premises on the list – but this includes low leakage rates.
Here are few stats about the reach of the program.
We’ve contacted 1626 folks since January The list as an average of 96 names everyday, but some are repeats. We don’t go after repeats; we move on to the next customer.
This is summer peaking program. The count on the list jumped up this spring as irrigation systems were turned on (and started leaking). Likewise, the number of premises rising into the highest range – the range we call – also increased in the summer. At some points this summer, we couldn’t contact everybody on the day’s list during that day. We’d start at the top and work our way down and then start again at the top the next day.

11. Data tracking, surveys and feedback
Improved data tracking
Limited survey responses so far
Feedback - Only first quarter of the year
Mostly reporting toilets
We’re getting survey data but it’s slow.
We’re actually getting more information from folks calling back to say thanks.
The sample is really small (31) and it’s seasonal.
And from this, most (78%) report that the toilet was at fault.
But hose bibs and sprinkler systems showed up even in the winter.

12. Customer responses
We really appreciate that call because we called a plumber and found the problem and we wouldn’t even have known about it if the city hadn’t called us. So thank you. I think it’s a wonderful service.
Our approach emphasizing that it’s a service results in generally good vibes. People appreciate the heads up. One of our next steps will be to survey customers to get reactions and information about how issues were resolved.
I have gotten feedback from people who didn’t realize I was part of the program. They just mentioned in casual conversation that they thought our organization was doing something cool.
We’ve also avoided continuing contact with people who tend to stay on the list. Future efforts might be more assertive with this set of continuous users.
Folks who are out of town really appreciate the call. IF they can’t get ahold of a neighbor to shut off their water, we still do offer curb shut off.
The customer interactions from this program are in stark contrast to customers that call us. When customers see something they don’t like in their billing data, tends to start as a more negative conversation.

13. Estimated water volume – data from a sample of homes analyzed
Looked at 25 homes from a two-week period in October
Compared to previous year.
Found that homes had an average increase of 10 kgal of use for that month.
At 95% confidence, the range is ±4.45kgal.

14. Lessons learned Staff time is not trivial – the calls take time.
About a 0.5 FTE
Meter questions
Meter shop was over-loaded.
We sometimes function as AMI quality control.
Staff time is not insignificant, and the program has changed to reduce this. Initially when I started staff was calling and if they were not able to reach the customer after several ties, they would visit the home. This has since changed and now we limit our outreach to phone calls.
Still, between data collection and calls and callbacks, the time per case is about a half hour. It works out to about a 0.5 FTE. Next summer we might have a higher cut-off for average flow rate. Raising the average flow rate that triggers the call could keep the workload under control.
At one point we were offering curbside shut-offs to folks as part of this service. The meter shop was doing these and frankly, these had to stop. Requests for meter checks also increased through this program. The burden on the meter shop was too much. We also got to be a great deal more careful about talking to folks about their meter. We spend time explaining meters so we don’t have meter checks.

15. questions?
Didn’t make the cut