Page 238 - Water Loss Control

Contr olling Appar ent Losses—Data Transfer Err ors 211

Example Calculation of Data Transfer Error in County Water Company (CWC)
The manager compiling the water audit for county water company has suspicions that
customer accounts with many billing cycles of estimates are a potential source of data
transfer error. He determines to field investigate 50 customer accounts that have not had an
actual meter reading in over 2 years. Estimates have been used in billing these accounts
over this 2-year period.
The manager’s first step was to obtain access to the customer premises in order to
obtain current meter readings. After sending violation notices to the customers and making
contact during the first month of this effort, CWC was able to gain access and obtain current
meter readings in 38 of the 50 properties. For the 12 properties that could still not be
accessed, more aggressive steps by the water utility–such as service disconnection–will be
needed in order to force access to the customer meter. For the 38 accounts that were
accessed, updated meter readings found that these accounts had been collectively under-
billed by 360 thousand gallons (kgal) during the 2-year period, or an average of 180 kgal per
year. Billing records show that the water utility had a total of 487 accounts that went without
an actual meter reading over the past 2 years. Based upon the findings of the 38 accounts

487
Apparent loss (2-year missing reads) = (180 kgal) = 2,307 kgal
38
The value of 2307 kgal should be included in the water audit as one subcategory of
apparent loss due to data transfer error. Any other groups of suspicious accounts, such as
zero consumption accounts, should also be investigated and extrapolated over the customer
population to obtain a quantity of apparent loss due to data transfer error. Several
subcategories might be identified in any water utility and these should ultimately be included
in the water audit and totaled under apparent loss due to data transfer error.

FIGURE 13.10 Example calculation of data transfer error in County Water Company. (Source:
George Kunkel.)

two to as many as eight or more. Subcategories that might be considered for investiga-
tion can include
• Accounts without actual meter readings for one year or longer
• Accounts showing zero consumption for three or more billing cycles
• Accounts suddenly evidencing a significant drop in consumption after a stable
history of higher consumption levels
• Accounts with confirmed AMR equipment failures
• Accounts known to have suffered from manual meter reading inaccuracy from
one or more meter readers confirmed to be inattentive or dishonest
• Accounts known to have suffered data distortion in transferring data from
handheld meter reading devices into the customer billing system

These are but several possible causes of apparent loss due to data transfer error. It
is incumbent on each utility to determine a reasonable extent of cause and volumes of
this form of apparent water loss. The key bottom-up activities in this regard are analy-
sis of billing records for unusual consumption patterns or missing meter readings, and
auditing/investigation of samples of suspect accounts to confirm actual volumes of
apparent loss.