5.4 ERMUSR 05-14-20244r.;,
Elk Riv
Municipal Utilities UTILITIES COMMISSION MEETING
TO:
FROM:
ERMU Commission
Mike Tietz —Technical Services Superintendent
MEETING DATE:
AGENDA ITEM NUMBER:
May 14, 2024
5.4
SUBJECT:
2023 Annual Reliability Report
ACTION REQUESTED:
Receive the 2023 Annual Reliability Report
BACKGROUND:
Minnesota Rules Chapter 7826 Public Utilities Commission Electric Utility Standards cover
safety, reliability, service, and reporting requirements. Per 7826.0100(A), municipal utilities are
exempt from these requirements. However, the Elk River Municipal Utilities Commission
adopted several parts of this chapter as a Distribution Reliability Standard policy requiring
annual reporting on system reliability.
DISCUSSION:
In 2023, our reliability index numbers remain very good. These reliability index numbers reflect
the condition of our robust electrical system as well as the superb response time from our local
line crews. This also reflects the local accountability, long term visioning on system design, and
ongoing system maintenance. The table below reflects the number of customer outage minutes
for the last 10 years.
Year
# of Customers
# of Outage Minutes
# of Outages
2014
9426
243,965
44
2015
9449
225,337
41
2016
10862
432,310
59
2017
11489
354,625
40
2018
12158
117,055
51
2019
12463
312,885*
59
2020
12604
197,884
51
2021
13038
628,670
61
2022
13228
174,500
50
2023
13543
187,469
51
*118,978 oft hese minutes are due to two separate transmission outages.
Page 1 of 3
185
Listed below are several reliability indices that are used within the electric industry to make it
easier to compare performance among utilities along with ERMU's 2022 and 2023 statistics.
Average Service Availability Index (ASAI)
ASAI is a measure of the average availability of the sub -transmission and distribution systems
to serve customers. It is the ratio of the total customer minutes that service was available to
the total customer minutes available in a time period. This is normally expressed as a
percentage.
ERMU's 2023 ASAI is 99.9973% Availability ERMU's 2022 ASAI is 99.9974% Availability
Customer Average Interruption Duration Index (CAIDI)
CAIDI is defined as the average duration (in minutes) of an interruption experienced by
customers during a specific time frame. It is calculated by summing the customer minutes off
during each sustained' interruption in the time period and dividing this by the number of
customers experiencing one or more sustained' interruptions during the time period. The
resulting unit is minutes. The index enables utilities to report the average duration of a
customer outage for those customers affected.
ERMU's 2023 CAIDI is 84.827 minutes ERMU's 2022 CAIDI is 90.146 minutes
System Average Interruption Duration Index (SAIDI)
SAIDI is defined as the average interruption duration (in minutes) for customers served
during a specified time period. It is determined by summing the customer minutes off for each
sustained' interruption during a specified time period and dividing that sum by the average
number of customers served during that period. The resulting unit is minutes. This index
enables the utility to report how many minutes' customers would have been out of service if all
customers were out at one time.
ERMU's 2023 SAIDI is 14.077 minutes ERMU's 2022 SAIDI was 13.289 minutes
System Average Interruption Frequency Index (SAIFI)
SAIFI is defined as the average number of times that a customer is interrupted during a
specified time period. It is determined by dividing the total number of sustained' interrupted
customers in a time period by the average number of customers served. The resulting unit is
"interruptions per customer."
ERMU's 2023 SAIFI is 0.166 ERMU's 2022 SAIFI was 0.147
' Only outages lasting longer than five minutes are included in the calculations, as defined by IEEE 1366.
Page 2 of 3
Staff reviews these statistics in detail to find ways to continually improve our system. These
statistics are very good when compared with other utilities. As indicated by the following pie
chart and statistics list, you can see the causes for the outages experienced in 2023.
Lightning
4%
Equipment Damage
6%
Underground
s%
Contraclor-Dig-in
B%
Other
1 A -A
Squirrel
Equipment Worn Out 18%
8%
Outage Cause
Count
View
17
0
Squirrel
9
0
Equipment Worn Out
4
0
ree
3% Contractor -Dig -In
4
0
Underground
4
0
Equipment Damage
3
0
Lightning
2
0
ure
0
Equipment
2
0
Wind
1
0
Vehicle Accident
1
0
Wildlife
1
0
Vegetation
1
0
Total
51
Many investor -owned and cooperative run utilities "storm normalize "their reliability numbers
by removing "major outages" due to storms from their reliability index number calculations.
Our reliability numbers have not been storm normalized. Our numbers reflect these major
outages and represent what our customers experienced in 2023. Overall, our electric
distribution system is very resilient, and our system reliability remains excellent.
Staff requests that the Commission receive the attached American Public Power Association's
(APPA) eReliability Tracker Annual Report, which contains published averages that can be used
to better understand the performance of our electric system relative to other utilities nationally
and to those within our region or size class.
ATTACHMENT:
• APPA eReliability Tracker 2023 Annual Report for Elk River Municipal Utilities
Page 3 of 3
187
Elk River Municipal Utilities
6,# ANNUAL
CM BENCHMARKING
eRELIABILITV
IM REPORT
DEED
PUBLIC
'�►! . rR
RESEARCH 6 DEVELOPMENT
American Public Power Association
AMERICAN
PUBUC
P=! _. ER
ASSOCIATION
Powering Strong Communities
I. About This Report
This report focuses on distribution system reliability across the country and is customized to
each utility that participates in the American Public Power Association's eReliability Tracker
ffetyilm. APPA created the eReliability Tracker Annual Report to assist utilities in their e
understand and analyze their electric system. In 2012, APPA developed the eReliability Tracker
fG fe\m4uprrtefmoo[DBEM)emonstration of Energy & E
program.
fts dz0a time eReliability Tracker from January 1, 2023 to December 31, 2023.
tbtsjoualytilitogJtbth :ttgsriipyedycianot have a full year of
data in the system. The report includes data recorded as of March 15, 2024.
Redialbdttyhretoric and ongoing engineering investment decisions within a utility.
Proper use of reliability metrics ensures that a utility is performing its intended function and is
ffctividiingrsee.ice in a consistent and e
While the primary use of reliability statistics is for self -evaluation, you can use these statistics
ffaarmai !psrffutilities. However, di
§umation, ambient environment, weather conditions, and number of customers served
typically limit most utility -to -utility comparisons. Due to the diverse range of utilities that use
the eReliability Tracker, this report endeavors to improve comparative analyses by grouping
utilities by size and region.
Since this report contains data for all utilities that use the eReliability Tracker, it is important to
fiati6irha�b Gi ticbl�r�gldar e3sffnall utility can a
the issues associated with comparability, each utility's reliability statistics are weighted based
on customer count when aggregated. This means that all utilities are equally weighted, and all
individual statistics are developed on a per customer basis.
Etie260gregate statistics in this report are calculated from the 324 utilities with veri
outage data. Utilities that experienced no outages in 2023, or did not upload any data, will
6aJ1d,vD�eela�rir`iel'udas� in their report for utility-speci
the aggregate analysis. Also note that log -normal data with a z-score! greater than 3.25 may
fianilyclaiidt&tt itl-fdgjgregate statistics.
ffrAfzcmtledrmdemtegdrrdvwsroaucie,aaciDtagrreint di
of 3.25 indicates that the data point is three and one -quarter standard deviations from
the mean. A z-score of 0 indicates that the data point is identical to the mean.
Pj
f dlklgsClassi
This report separates utilities into groups according to geographic region and the number of
customers served. Table 1 shows the range of customer counts for utilities that use the
fiRdiaM4gTamaperdfi�approximately 105 utilities per group.
Your utility is in size class zbnd region 3
Table 1. Customer count range per size class
Customer Count Range
Class 1
[0,1518)
Class 2
[1518, 3480)
Class 3
[3480, 7325)
Class 4
[7325, 14489)
Class 5
[14489, 503649)
Each utility is also grouped with all other participating utilities within their region. Figure 1
shows the number of utilities using the eReliability Tracker in each region and Figure 2 shows
the states and territories included in each region.
Figure 1. Number of utilities subscribed to the eReliability Tracker by region
140
120
N
a 100
0
80
O
Q1 60
E
7 40
Z
20
0
1 2 3 4 5 6 7 8 9
Regions
190
3
Figure 2. Regions
J11111111-
AMERICAN GUAM NORTHERN PUERTO U.S. VIRGIN
SAMOA MARIANA RICO ISLANDS
ISLANDS
191
II. IEEE Statistics
I imithmInet;i=eelliBbility, the industry standard metrics are de
Electrical and Electronics Engineers' Guide for Electric Power Distribution Reliability Indices, or
IEEE 1366 guidelines. For each utility, the eReliability Tracker performs IEEE 1366 calculations
for System Average Interruption Duration Index (SAIDI), System Average Interruption
Frequency Index (SAIFI), Customer Average Interruption Duration Index (CAIDI), Momentary
Average Interruption Frequency Index (MAIFI) and Average Service Availability Index (ASAI).
It is important to note how major events (MEs) are calculated and used in this report. An
example of a ME includes severe weather, such as a tornado or hurricane, that leads to
unusually long outages in comparison to your distribution system's typical outage. This report
uses the ABl3ilh�ihdctly on the SAIDI for speci
rather than a daily SAIDI. The APPA ME threshold allows a utility to remove outages that
exceed the IEEE 2.5 beta threshold for outage events, which considers up to 10 years of the
utility's outage history. In the eReliability Tracker, if a utility does not have at least 36 outage
events prior to the year being analyzed, then no threshold is calculated. If this is the case for
9"aniilfbft tiabnj ,Wi mlhrbidveutNULL value in the following
MEs in the SAIDI, SAIFI, CAIDI, and ASAI sections of this report will be the same as the
calculations with MEs for your utility. More outage history will provide a better threshold for
your utility.
YugflhAtifity's APPA major event threshold is
For each of the reliability indices, this report displays your utility's metrics alongside the mean
values for all utilities using the eReliability Tracker and within the same class and region as
gk=btdi*tffiheeach of the following subsections allows you to better
understand the performance of your electric system relative to other utilities nationwide and
to those within your same region or size class. The second table breaks down the national
data into quartile ranges, a minimum value, and a maximum value.
All indices, except MAIFI, are calculated for outages with and without MEs. Furthermore, the
tables show indices for scheduled and unscheduled outages. Note that scheduled and
unscheduled calculations include MEs. Also note that wherever MEs are excluded, the
exclusion is based on the APPA ME threshold for your system.
192
IIA. System Average Interruption Duration Index
SAIDI is the average duration (in minutes) of an interruption per customer served by the utility
fl6ffiMfbcffpeci
flene-OrSMdX is a sustained interruption index, only outages lasting longer than
are included in the calculations. SAIDI is calculated by dividing the sum of all customer
minutes of interruptionlfidittti*etfraBpeby the average number of customers
served during that period. For example, a utility with 100 customer minutes of interruption
and 100 customers would have a SAIDI of 1.
Note that in the tables below, scheduled and unscheduled calculations include MEs. Also note
that wherever MEs are excluded, the exclusion is based on the APPA ME threshold for your
system.
TaMm4e SAIDI with and without MEs
In minutes
Your
Utilities that use the eReliability Tracker
Utilities in your region
Utilities in your size class
All
No MEs
Unscheduled
Scheduled
14.08
5.78
11.72
2.36
88.97
47
84.06
8.95
46.77
26.83
44.13
5.4
72.22
37.37
70.12
3.36
Tabte-ary SAIDI data from the eReliability Tracker
In minutes
Minimum
First Quartile
Median
Third Quartile
Maximum
All
No MEs
Unscheduled
Scheduled
0.2
0.04
0.04
0
18.2
11.89
15.9
0.19
50.54
27.57
45.7
1.04
111.3
58.28
106.4
4.44
1028.99
691.25
1028.75
480.88
193
1.1
Figure 3. Average SAIDI by region
175
N 150
125
F, 100
Q
U-) 75
N
50
U!
Q 25
Regions
1. Customer minutes of interruption is calculated by multiplying total customers
interrupted and total minutes of interruption.
IN
194
II.2. System Average Interruption Frequency Index
SAIFI is the average instances a customer on the utility system will experience a sustained
f mirn ,dDmduring a speci
%nn-drSM6t ire sustained interruption index, only outages lasting longer than
included in the calculations. SAIFI is calculated by dividing the total number of customers that
experienced sustained interruptions by the average number of customers served during that
period. For example, a utility with 150 customer interruptions and 200 customers would have
a SAIFI of 0.75.
Note that in the tables below, scheduled and unscheduled calculations include MEs. Also note
that wherever MEs are excluded, the exclusion is based on the APPA ME threshold for your
system.
lFW3dKB4e SAIFI with and without MEs
In interruptions
All
No MEs
Unscheduled
Scheduled
Your utility
0.17
0.1
0.11
0.06
Utilities that use the eReliability Tracker
0.74
0.51
0.7
0.06
Utilities in your region
0.58
0.39
0.56
0.05
Utilities in your size class
0.7
0.481
0.671
0.03
Tabiui%ry SAIFI data from the eReliability Tracker
In interruptions
All
I No MEs
Unscheduled
Scheduled
Minimum
0
0
0
0
First Quartile
0.22
0.16
0.22
0
Median
0.56
0.37
0.52
0.01
Third Quartile
1.07
0.71
1.01
0.04
Maximum
3.47
2.86
3.47
1
195
Figure 4. Average SAIFI by region
1.4
0 1.2
Q
3 1.0
UJ
0.8
0.6
Q
U'1
N 0.4
j 0.2
Q
0.0
Regions
196
II.3. Customer Average Interruption Duration Index
CAIDI is the average duration (in minutes) of an interruption experienced by customers during a
fii&de frame.
&nadrCLkOl is a sustained interruption index, only outages lasting longer than
are included in the calculations. CAIDI is calculated by dividing the sum of all customer
minutes of interruption by the number of customers that experienced one or more
#mBrtlaptioer,aoerQngtbmL-pepapdr.iThis--metric re
(minutes of duration) during an outage.
Note that in the tables below, scheduled and unscheduled calculations include MEs. Also note
that wherever MEs are excluded, the exclusion is based on the APPA ME threshold for your
system.
VaMmge CAIDI with and without MEs
In minutes
All
No MEs
Unscheduled
Scheduled
Your utility
84.83
59.7
109.27
40.19
Utilities that use the eReliability Tracker
111.62
87.71
110.38
130.83
Utilities in your region
75.97
61.96
78.04
103.9
Utilities in your size class
j 107.49
j 81.49
j 107.94j
103.51
Tabief7ary CAIDI data from the eReliability Tracker
In minutes
All
No MEs
Unscheduled
Scheduled
Minimum
15.04
13.65
14.74
6.68
First Quartile
64.3
53.52
63.02
59.83
Median
88.91
75.24
89.51
85.93
Third Quartile
130.35
106.26
129.46
140.8
Maximum
716.84
482.79
777.49
1373.7
10
197
Figure S. Average CAIDI by region
175
Ln
150
C
E 125
100
Regions
W.
11
II.4. Momentary Average Interruption Frequency Index
MAIFI is the average number of momentary interruptions a utility customer will experience during a
jil&re frame.
i1Veti$tal3�i✓ith a duration of
MAIFI is calculated by dividing the total number of customers that experienced momentary
interruptions by the total number of customers served by the utility. For example, a utility with
20 momentary customer interruptions and 100 customers would have a MAIFI of 0.20.
ffi&terfitacy bntelmda tion9va mIaL,@nmietcbmated
outage management system might not log these interruptions; therefore, some utilities have
a MAIFI of zero.
RkWEa§.e MAIFI
In interruptions
Your utility
Utilities that use the eReli
Utilities in your region
Utilities in your size class
All
NULL
Tracker 0.41
0.58
0.42
Mabte%ry MAIFI data from the eReliability Tracker
In interruptions
All
Minimum
0
First Quartile
0
Median
0.06
Third Quartile
0.43
Maximum
4.45
199
12
Figure 6. Average MAIFI by region
° 0.6
Q
L 0s
w
0.4
Q 0.3
0.2
fu
L
j 0.1
Q
0.0
Regions
200
13
II.5. Average Service Availability Index
ASAI is the percentage of time the sub -transmission and distribution systems are available to serve
fiLtsrftDfeasr&.iring a speci
This load -based index represents the percentage availability of electric service to customers
within the period analyzed. It is calculated by dividing the total hours in which service is
available to customers by the total hours that service is demanded by the customers. For
example, an ASAI of 99.99% means that electric service was available for 99.99% of the time
during the given period. Note that the higher your ASAI value, the better the performance.
In the tables below, scheduled and unscheduled calculations include MEs. Also note that
wherever MEs are excluded, the exclusion is based on the APPA ME threshold for your system.
Table 10. Average ASAI with and without MEs
In percentage
All
No MEs
Unscheduled
Scheduled
Your utility
99.9973
99.9989
99.9977
99.9995
Utilities that use the eReliability Tracker
99.9833
99.9912
99.9842
99.9982
Utilities in your region
99.9912
99.9949
99.9917
99.9989
Utilities in your size class
99.9864
99.9929
99.9868
99.9993
Table 11. Summary ASAI data from the eReliability Tracker
In percentage
All
Maximum 199.9999
No MEs
Unscheduled
Scheduled
99.9999
99.9999
99.9999
First Quartile
99.9965
99.9977
99.9969
99.9999
Median
99.9907
99.9947
99.9916
99.9998
Third Quartile
99.979
99.989
99.9804
99.9991
Minimum
99.8042
99.8684
99.8042
99.9085
14
201
Figure 7. Average ASAI by region
100.000
UJ 99.975
m
C 99.950
v
U
99.925
Q
Q99.900
v1
99.875
U1
ro 99.850
L
99.825
99.800
Regions
202
15
II.6. Energy Information Administration Form 861 Data
Form EIA-861 collects annual information on electric power industry participants involved in the
generation, transmission, distribution, and sale of electric energy in the United States and its
territories.
In 2014, EIA began publishing reliability statistics in Form EIA-861; therefore, APPA included
these statistics in this report for informational purposes. Please note that the following data
includes 174 investor -owned, 465 rural cooperative, and 324 public power utilities that were
Mcgo thrmfujhE11A116fle4wn-ecThe statistics do not include data
from utilities that complete the EIA 861-S form, which smaller entities complete. Note that the
fl&bp W"ugip4Mi;cpm Wealutilities include entities classi
subdivision, and state. In addition, since the collection and release of EIA form data lags by a
year, the data is based on 2022 data that was published October 5, 2023. Therefore, we
suggest you only use the aggregate statistics contained herein as an informational tool for
further comparison of reliability statistics.
In Form EIA-861, an entity provides SAIDI and SAIFI including and excluding ME days in
accordance with the IEEE 1366-2003 or IEEE 1366-2012 standard.
Although EIA collected other reliability -related data, the tables below only include SAIDI and
SAIFI data including and excluding ME days. You can download the full set of data at:
www.eia.gov/electricity/data/eia861 /.
Table 12. Your utility's SAIDI and SAIFI with and without IEEE ME days
SAIDI with IEEE ME SAIDI without IEEE SAIFI with IEEE ME days
days (minutes) ME days (minutes) (interruptions)
14.08 5.78 0.17
Table 13. Summary SAIDI data from Form EIA-861, 2022
In minutes
All
No MEs
Average
363.98
148.09
Minimum
First Quartile
0.08
79.8
0
54.64
Median
176.36
105.13
Third Quartile
369.21
178.96
Maximum
11949.11
1760.49
SAIFI without IEEE ME
days (interruptions)
0.1
203
i[:
Table 14. Summary SAIFI data from Form EIA-861, 2022
In interruptions
All
TNo MEs
Average
1.74
1.3
Minimum
0
0
First Quartile
0.85
0.64
Median
1.42
1.07
Third Quartile
2.25
1.66
Maximum
15.96
12.07
204
17
III. Outage Causes
Equipment failure, extreme weather events, wildlife, and vegetation are some of the most
common causes of electric system outages.The following pie chart shows the percentages of
the primary causes of outages for all utilities using the eReliability Tracker in 2023.
weather
wildlife 17 8%
18.7%
21.5%
equipment
4.4%
utility human error
power supply
public
9.9%
14.9°Io scheduled
11.5°/0
vegetation
unknown
Figure 8. Primary causes of outages in 2023
Certain factors, such as regional weather and animal/vegetation patterns, can make some
5apmaEpno6rAiptimmI'Ehefotlewftisection includes graphs
depicting common causes of outages for your utility, all utilities in your region, and all utilities
using the eReliability Tracker.
@ihBrte!c rmtaining aggregate information are customer -weighted to account for di
utility size for a better analytical comparison. For example, a particularly large utility may have
a large number of outages compared to a small utility. To avoid skewing the data toward large
utilities, the number of cause occurrences is divided by customer size to account for the
205
€frences. In Figures 9 to 14, the data represent the number of occurrences for each group
of 1,000 customers. A customer -weighted occurrence rate of "1" means an average of one
outage from that cause occurred per 1,000 customers in 2023.
Note that the sustained outage cause analysis is more comprehensive than the momentary
outage cause analysis due to a larger and more robust sample size for sustained outages.
Regardless, tracking both sustained and momentary outages helps utilities understand and
reduce outages. To successfully use the outage information tracked by your utility, it is
imperative to classify and record outages in detail. The more information provided per
outage, the more conclusive and practical your analyses will be.
19
206
IIIA. Sustained Outage Causes
In general, sustained outages are the most commonly tracked outage type. In analyses of
sustained outages, utilities tend to exclude scheduled outages, partial power, customer -
related problems, and qualifying major events from their reliability indices calculations. While
this is a valid method for reporting, these outages should be included for internal review to
make utility -level decisions. In this section, we evaluate common causes of sustained outages
for your utility, corresponding region, and for all utilities that use the eReliability Tracker. It is
fidoorttziatiibportastbadtsgetathatilosaages are classi
fmffu§gL1W al�cffli_�
Figure 9.%mpauses of sustained outages for all utilities that use the eReliability Tracker
2.5
L 1.0
O 0.5
0.0
Outage Cause Types
€tegpm*@.of sustained outages for your utility 1
1.25
1.2
O
O 1.0
O
v 0.8
0.66
v 0.6
C
4!
UU 0.4 I
0.29 0.29 0.29
u
0 0.2
0.0
Tree Squirrel Contractor -Dig -In Equipment Worn Out Underground
Outage Cause Types
207
Pi
ftpHLesti.of sustained outages in your region
32.0
0
a
N 1.5
a
v
V
C 1.0
W
7
U
O 0.5
0.0
Outage Cause Types
1. The number of occurrences for each cause is divided by the utility's customer count (in
ffireataibidt�to create an occurrence rate that can be compared across di
sizes.
21
1:
III.2. Momentary Outage Causes
MIt ebW1i 0eitaAlIk omsmem*aiyysUutagitwtan be di
due to the hazard they pose for electronic equipment, it is important to track and analyze the
causes of momentary outages. This section evaluates the common causes of momentary
outages for your utility, region, and size class as well as common causes for all utilities that
fdeeAhraaitEwbability Tracker. Please note that only outages lasting less than
it 6*WWy,j&hWyures 12
number of occurrences for each cause is divided by that utility's customer count (in
€fireataibidtjYtq�zemate an occurrence rate that can be compared across di
€kgpRms*2.of momentary outages for all utilities that use the eReliability
Tracker
10
10.22
�
O
O 8
O
L
a 6
v
V
C
LJ 4
L
U
U
U
2.06
O 2
1.58
0
Power Supply Storm Non -Payment
Outage Cause Types
fkpimesH.of momentary outages for your utility
0.04
O
O
O
rl 0.02
N
a
0.00
V
C
N
L
—0.02
V
U
O
—0.04
0.75 0.68
Unknown Utility Maintenance and Repairs
0.00 0.00 0.00 0.00 0.00
None None None None None
Outage Cause Types
209
PA
€kjpatesU.of momentary outages in your region
0.8
0
0
O
0.6
W
a
OV 0.2
0.0
Outage Cause Types
210
23
Thank you for your active participation in the eReliability Tracker service. We hope this report
is useful to your utility in analyzing your system. If you have any questions regarding the
material provided in this report, please contact:
APPA's Reliability Team
Paul Zummo
Ji Yoon Lee
Matthew Atienza
Reliability@PublicPower.or
American Public Power Association
2451 Crystal Drive, Suite 1000
Arlington, VA 22202
For more information on reliability, visit www.PublicPower.ora/Reliability..
AMERICAN
1r
PUB-.01`_ .r
PWv Wr �RTM
ASSOCIATION
Powering Strong Communities
2451 Crystal Drive
Suite loon
Arlington, VA 22202-4804
vwvw. P u b l i c P owe r. o rg
Copyright 2023 by the American Public Power Association. All rights reserved.
P011
211