How do you calculate SAIDI

SESB power supply reliability issues

Tomorrow is the final day for the English Premier League current season. If you love watching live football broadcast like I do, you wouldn't want another statewide blackout tomorrow at 11.00pm Malaysian time, or at any time for that matter. 

However, Sabah's SESB is notoriously known for power failure during crucial times. Reliability is not in their dictionary, they are far from being reliable.

Typical main intake substation (PMU): Do you think SESB is a reliable power provider?

Since SESB takes pride of itself for having improved SAIDI significantly, one would be interested to know if the lower SAIDI figure (lower is better) is translated into better quality in actual daily experience. SESB's past 6 years SAIDI record is as follows (source -SESB website):

2006 - 4,030 minutes per customer per year
2007 - 1,986
2008 - 1,855
2009 - 2,867
2010 - 687
2011 - 494
2012 - 362 (as at April)

Supposing the 2012 figure did not take into account the statewide power failures on 30 April - 1 May, this year's SAIDI could probably surpass the 2011 figure.

System Average Interruption Duration Index (SAIDI)

This index measures the total duration of an interruption for the average customer during a given time period. This is the most often used performance measurement for a sustained interruption and is normally calculated on either monthly or yearly basis.

SAIDI calculation

To calculate SAIDI, each interruption during the time period is multiplied by the duration of the interruption to find the customer-minutes of interruption. The customer-minutes of all interruptions are then summed to determine the total customer-minutes. To find the SAIDI value, the customer-minutes are divided by the total customers served.

Formula:

SAIDI = Σ(r_{i *} N_i ) / N_T

where:
.
SAIDI  = System average interruption duration index, (minutes)
r_i          = Restoration time, (minutes)
N_i        = Total number of customers interrupted
N_T       = Total number of customers served


Example:

What is the SAIDI for 30th April and 1st May given the following data? 

Table 1 shows each power outage, the duration of the outage, and the customer-hours. As of August 2011, SESB has 456,406 customers.

Table 1: Calculating customer-hours:

Date	Time	Customers	Duration (minutes)		Customer-hours
April 30, 2012	03:23	456,000	0.33 hrs	or   20 min	151,985
April 30, 2012	03:23	100,000	8 hrs	or  600 min	800,000
April 30, 2012	03:23	10,000	13 hrs	or  300 min	130,000
May 1, 2012	12:05	60,000	2 hrs	or  120 min	120,000
					1,201,985

From the table, the first outage was at 3:23 in the morning and 456,000 (approx.) customers were out of service for 20 minutes (0.33 hours). The 20 minutes was based on the statement made by SESB that the utility started normalising supply in stages as early as 3.40am that day. 

From the Table, customer-hours = 456,000 * 0.3333     (for the first power outage on 30th April)
                                               = 151,985 hours

Total customer-hours                 = 1,201,985 hours or    
                                               = 72,119,088 customer-minutes


Therefore, SAIDI = 72,119,088 / 456,000
                         = 158 minutes

This means that the average customer was out for 158 minutes on the 30th April - 1st May 2012.


For Malaysians in Sabah, SAIDI tells nothing but a number

1) Lower SAIDI recorded in recent years look nice on paper but did little to a better power supply experience in general.

2) SESB sets SAIDI target of 600 minutes in 2012. Take for example these states in West Malaysia, they achieved SAIDI of less than 80 minutes per customer per year in 2011:

State	Target ( 2011 )	Achievement  (minutes per customer per year)
Sabah	700	494
Penang	99	76
Kelantan	75	72
Selangor	74	61
Johor	99	78

3) If you are living in Sabah, last year you would have experience 494 minutes of power outage, or slightly over 8 hours for the whole year. Take this year's 30th April statewide blackout into account and you are already more than 8 hours without power supply. Expect double of that figure by year end.

Parting shot: Who verify your figures?

We now know how SAIDI is calculated. What we need next is transparency in terms of data verification. Someone has to look into the utility's method of calculating it, just like what is done in financial audit or quality audit.