A. Need for a Standard
1. Initial Positions of Parties
PG&E represents that it has extended significant efforts and made substantial investments during the past five years to improve access, capacity, service and quality for normal operations and emergency operations. PG&E does not feel that further investment is warranted or economical. PG&E believes that its current capacity and capability are more than satisfactory and that additional call center standards are not necessary. PG&E believes that existing regulation and requirements promote quality customer service and emergency preparedness.
PG&E also believes that performance standards are unworkable. In actual events, it cannot control such things as the unavailability of the telephone network, the inability to reach off-duty employees, the inability of off duty employees to reach call center locations, etc. To judge a utility's performance and effectiveness without a comprehensive assessment, considering all applicable factors and variables, can result in arbitrary and unreasonable conclusions regarding the adequacy of its response. PG&E recommends a post-event review approach.
PG&E represents that it has exceeded Commission-mandated performance standards for its call centers. It has met or exceeded the standard for monthly average speed of answer (20 seconds) for 40 consecutive months. It has achieved the standard for monthly busy percent performance (less that 1 percent busies) for 33 consecutive months.
PG&E believes that its call centers are state-of-the-art. PG&E states that in the last five years it has reorganized the customer telephone service function, improved functionality and capacity, added features and technology and developed a comprehensive emergency plan.
PG&E points out that G.O. 166 requires it to file a plan describing how the call centers will be used to communicate with the public during a major outage. G.O. 166 also specifies information to be made available to the public by the call centers.
PG&E claims that the existing standards, average speed of answer and busies percent, will not work to evaluate performance during a major outage due to the uniqueness of emergencies and disasters. Call center performance can be affected by the following regardless of the call center design and planning:
· Time of day (staffing levels are less at night).
· Call volume (volume is likely to be higher when the public is not aware of the cause of the outage).
· Ability of off-duty staff to get to the call center (road conditions, etc.).
· Condition of the telephone network.
· Vendor problems regarding call center equipment.
SCE's position mirrored that of PG&E. SCE believes that additional standards are not required but, if adopted, they should reflect the following principles:
· They must respond to specific needs.
· They must be feasible and achievable.
· They must be demonstrated to be cost-effective.
· They must be fair.
SCE also believes that the Commission must provide for recovery of any resulting costs.
SCE estimates that the proposed standards would cost $17 million in incremental expenditures and $10 million in facilities costs.
SCE believes that the purpose of a prescriptive standard such as that proposed in D.98-03-036 should be to describe, for a given set of circumstances, a minimum acceptable end result, all other things being equal. Since major outages are by nature, unpredictable, such a standard cannot be developed.
SCE argues that Pub. Util. Code § 364(d) requires an after-the-fact review and the exercise of discretion. Self-executing penalties relieves the Commission of the obligation to review and determine fault. This would be contrary to the spirit and language of the statute.
SCE's current service level goal for its customer call center is 75% of weekly calls answered within 50 seconds for 90% of the weeks of the year. This goal was explicitly recognized by the Commission in D.98-07-077 and includes the impact of any major outages. SCE has exceeded the goal in 1997, 1998 and 1999 (year-to-date).
SCE states that its current call center procedures and processes already allow customers to report hazardous conditions with the highest priority. Most customers experiencing hazardous or life threatening conditions not related to electric service or equipment do not call SCE.
SDG&E believes that call center standards during emergency events are not necessary. It believes that such standards will not increase responsiveness to customer inquiries. It further believes that standards would not be cost effective, particularly given the improvements SDG&E has made to its call center.
SDG&E's current call center standard is 80% of all calls answered within 60 seconds. Additionally, SDG&E annually reports call center performance on the following criteria:
· Convenience of the timeframe arranged for an appointment.
· Average response time to electric emergencies.
· Average response time to gas emergencies.
· Level of busies in the call center.
· Number of abandoned calls in the call center.
· Average number of minutes (for a new caller) between the call connection to the first menu and the menu choice for a customer service representative.
· Time for completion of gas service.
· Time waiting for electric service.
SDG&E believes that the utilities call center should not be intended to be the primary source of restoration information during an unplanned catastrophic event. The customers should be able to receive such information through public forums such as public service announcements.
SDG&E believes that a standard based on the number of calls receiving a busy signal during a major outage would not be appropriate. This is because, during such an event, telecommunications providers may restrict use of the phone system to insure adequate infrastructure for emergency services. Also a busy signal may be received from some portion of the telephone system, a central office for example, rather than the electric utility. Additionally, due to the redial feature on most phones, individual callers may generate many busy signals thus inflating the number of busy signals received.
SDG&E states that a standard based on a percentage of calls answered within a specified time would not be reasonable. Since there is no way to predict the number of calls that would be received, it would be expensive to plan for such events. Also, satisfying such a standard would not mean that service would be restored any faster. SDG&E also believes that a standard based on hang ups would not be appropriate, because the hang ups may not be due to the utility's actions or failures to act.
SDG&E estimates that to respond to 10% of its customers calling simultaneously would cost $1 million annually for a recorded message only. The cost would be over $4 million annually to allow a customer to opt out to a live agent.
SDG&E believes that a readiness evaluation check list, that establishes up-front expectations for call center management actions during a major outage, would insure that reasonable actions are taken in response to a major outage. Such a list would be a post-event review tool and would include such factors as:
· Was the mutual aid for call centers agreement invoked?
· Is there an on-going training program in place for responding to emergency situations?
· Were all trained utility employees contacted and asked to respond within a specified time frame?
· Were public service announcements provided in a timely manner?
Sierra believes that call center standards are unnecessary and would impose significant costs on customers. Sierra states that it is currently in the process of upgrading its company-wide telephone response systems. It estimates that it would cost $330,000 to provide a call center that would meet the standards proposed in D.98-03-036.
Sierra indicates that it is impossible for it to track calls by state of origin. At the same time, its electric distribution system is constructed in such a way that it is possible to have outages that interrupt both California and Nevada customers.
Sierra believes that if the Public Service Commission of Nevada (PUCN) were to adopt standards, it would be cost-effective to adopt such standards system wide. PUCN has not addressed this issue to date.
To meet the call center standard previously proposed by TURN and ORA, that the utilities be able to answer at least 50% of incoming calls during emergencies, the costs would be:
Annual Expense $124,000
e) SoCal Water
SoCal Water has spent $62,000 since March of 1997 upgrading its BVES telephone system. Its existing systems can not be upgraded to measure how long customers must wait to talk to a customer service representative. Such an upgrade would not be a reasonable investment. SoCal Water believes it should be exempted from any standards that may be adopted due to its small service territory and limited number of permanent residents. The risk of major outages is primarily outside the BVES system.
SoCal Water explains that BVES was solely an electric distribution utility prior to electric restructuring and remains so today. Electric restructuring has had no effect on BVES except the source of its purchase power.
PacifiCorp believes that imposition of call center standards would impose a hardship on PacifiCorp. PacifiCorp already operates two call centers in Portland, Oregon and Salt Lake City, Utah. PacifiCorp has already invested in facilities to allow communications 24 hours a day at both locations.
g) TURN and ORA
TURN and ORA believe that call center standards are needed for substantially the same reasons that restoration standards are needed.
B. Design of a Standard
1. Initial Positions of Parties
a) TURN and ORA
TURN and ORA jointly recommend as follows:
· The commission should adopt reasonable call center standards because such standards are both necessary and useful.
· A call center standard should be applicable in large (over 10%) events with the following characteristics:
· "Percent busies," registered at the utility's phone switch and that of its contractors, would be the measurement indicator for the standard. The indicator would be measured on a 24-hour basis from the day the event begins, and separately for each 24-hour period until the end of the event. During the Commission investigation which follows the event, the utility's response would be presumed reasonable if the percent of busies was less than level-1; would be presumed unreasonable if the percent of busies was more than level-2; no presumption would apply for values of percent busies equal to or between level-1 or level-2, but the Commission would undertake a critical investigation. These presumptions would be rebuttable in the same way as for the restoration standard.
· The standard would be applicable to all events affecting 10% or more of customers, regardless of the cause.
· The beginning and end of an event would be measured the same as for the restoration standard.
· Automatic penalties would not be applicable. However, the Commission, at its discretion, could assign penalties for poor performance, where appropriate.
· Level-1 would be defined as 30% busies over the day of the outage (midnight to midnight).
· Level-2 would be defined as 50% busies over the day of the outage, and more than 50% busies in each of 6 hours.
· Small utilities (under 150,000 California customers) should be exempt from the call center emergency standards.
· The Commission should require utilities to gather additional data, for each event where over 10% of the customers are out, on system damage, number of customers restored during designated time periods, and weather conditions.
Level-1 is defined as 30% busies over the day of the outage. This level was chosen because it could be met on a daily basis, by PG&E, SCE and SDG&E during a major outage.
Level-2 is defined as 50% busies over the day of the outage and more than 50% busies for any six one-hour periods (need not be consecutive). This level was chosen because at this level, customers would have a reasonable chance of getting through to the call center over the course of several hours.
TURN and ORA believe that their proposed standard could be met with the utilities' existing capacity. Therefore, no additional costs would be incurred. In addition, the utilities should not have to unduly minimize the time spent on each call.
PG&E opposes the standard proposed by TURN and ORA. TURN and ORA claim that, since their proposed standard is based on historical data, it represents levels of emergency preparedness that ratepayers have already paid for. Therefore, no additional costs are necessary to meet the standard. From this, TURN and ORA conclude that their proposed standard is cost effective. PG&E points out that in its 1999 GRC, ORA is recommending a substantially lower level of call center funding than proposed by PG&E. PG&E alleges that its GRC proposal is necessary to maintain the current level of service. PG&E concludes that the TURN and ORA proposal has not been shown to be cost effective and is inconsistent with ORA's recommendation in PG&E's 1999 GRC proceeding.
PacifiCorp agrees with the TURN and ORA proposal for small utilities.
C. Joint Proposal
TURN, ORA, SDG&E, and SCE sponsored a joint proposal for a call center performance standard. It is included in Attachment 1.
The joint proposal can be summarized as follows:
· The two benchmarks, are:
· Level-1 is defined is 30% busies over the day of the outage.
· Level-2 is defined as 50% busies over the day of the outage plus at least 50% busies in each of six one-hour increments.
· The utility's call center performance is presumed reasonable if the percent busies is below level-1.
· The utility's call center performance is presumed unreasonable if the percent busies is above level-2.
· The presumptions are rebuttable.
· Only busies encountered by customers at the utility's switch and that of its contractors are included.
· Utilities with fewer than 150,000 electric customers are exempted.
1. Positions of Parties
a) TURN and ORA
TURN and ORA believe that the standard is needed because the call centers are an important part of a utility's response to a major outage. A call center standard would counter cost-cutting pressures that could adversely affect performance. Existing regulatory and PBR mechanisms do not cover the number of busies that occur during major outages.
TURN and ORA believe that the standard included in the joint proposal will help ensure that customers have a reasonable opportunity to get through to the utility to give and receive information during a major outage.
TURN and ORA point out that the design of the standard has been found reasonable by all parties, assuming in PG&E's case that there would be a standard at all.
TURN and ORA allege that the standard is cost effective because it will provide benefits at little or no cost relative to current spending levels. They further represent that the 30% benchmark level has been shown by modeling to be achievable by most utilities under the assumed circumstances.
TURN points out that in PG&E's 1999 GRC, TURN opposed recovery of costs associated with its Outage Information System (OIS). The OIS is used to facilitate reporting and restoration of outages. TURN maintained that the OIS was needed, but TURN opposed cost recovery as a penalty for PG&E's failure to replace its previous outdated system in a timely manner. TURN believes that its position in this proceeding is entirely consistent.
SDG&E believes that the joint proposal for call center standards represents a compromise acceptable to all of the major parties to the proceeding. The standard focuses on busies at the utility's telephone switch. It is a useful indicator because it does the following things at once:
· It measures how easily customers can get through to inform the utility about the hazards that they are facing.
· It measures how easy it is for the customers to get information that they need in order to plan their response to the emergency.
· It is a surrogate for the performance of the utility on how well it's able to get the information out through other media to inform people about the status of the emergency and how the restoration is going.
· It captures how accurate the information is that the customer receives on the first call because if they don't get accurate information, they will call again and keep overwhelming the switch.
· To a certain extent, it captures how the utility's outage information system is working.
SDG&E represents that the call center standard is reasonable because it sets an objective benchmark for measuring performance. It also provides reasonable assurance that customers can get through to the utility to receive information or report trouble.
SDG&E states that the standard was designed to be low cost (or no cost) and is, therefore, cost-effective. It believes that there are no cost recovery issues for this proceeding.
SCE is a sponsor of the joint proposal. SCE emphasizes that the standard is a compromise among the parties. It supports the standard for the following additional reasons:
· It provides an objective benchmark against which performance is measured.
· It is reasonably feasible, achievable and practical to implement.
· It is fair because it limits application of the standard to matters within the utility's control.
Sierra does not object to the joint proposal or to PG&E's proposal because utilities with fewer than 150,000 electric customers would be exempted.
e) SoCal Water
SoCal Water believes that it should be exempted from the standards for the reasons set out in its initial recommendation.
PG&E believes that standards are not necessary for the reasons set out in its initial recommendation. However, it takes the position that it could support the joint proposal with two caveats.
First, since the standards are based on historical performance, the Commission should maintain the utility's current level of resources for the call center function within customer services. PG&E states that it has been spending considerably more than it was authorized in its last GRC. PG&E takes the position that it should be granted its requested amount in its 1999 GRC if the Commission adopts the joint proposal.
Second, PG&E states that the standard should not be applied to transmission outages.
A call center standard is needed for the same reasons as the restoration time standard. This is an important benchmark to have in place in order to judge utilities' responses to emergencies and major outages. Additionally, no party, including PG&E, has objected to the call center standard language in the joint recommendation. PG&E's proposal is identical.
The sponsors of the joint recommendation state that the proposed call center standard is feasible, achievable and practical to implement. No party except PG&E has said otherwise. PG&E's primary concern is related to funding in the 1999 GRC, which will be addressed in that proceeding. PG&E's arguments concerning inclusion of transmission outages were addressed in the discussion of the restoration time standard.
The sponsors of the joint recommendation state that it is cost effective and is likely to achieve its intended effect. No party has stated that it is not. The intended effect is that it will motivate the utilities to maintain or improve call center performance, and provide a benchmark against which performance can be measured.
We do not expect that the joint proposal will result in significant future costs to the utilities. However, in the future if the utilities find that there are significant costs, they will not be precluded from filing for recovery prospectively. Likewise, other parties will be free to oppose such requests.
For the above reasons, we conclude that the call center standard in the joint proposal is reasonable, cost-effective, and likely to have its intended effect.
For all of the above reasons we will adopt the call center standard included in the joint proposal.
Findings of Fact
1. The joint proposal regarding restoration performance benchmark and a call center benchmark is sponsored by TURN, ORA, SDG&E, and SCE.
2. PG&E's proposal is identical to the joint proposal except that it excludes transmission.
3. The restoration standard benchmark in the joint proposal is based on the available historical data.
4. The restoration standard benchmark in PG&E's proposal is based only on PG&E's historical data.
5. The cost of PG&E's proposed restoration standard is unknown for SDG&E and SCE.
6. Funding considerations raised in PG&E's 1999 GRC are properly addressed in that proceeding.
7. The restoration time standard in the joint proposal will focus management's attention on achieving reasonable restoration time without providing perverse incentives and will provide a benchmark against which performance can be measured.
8. The call center standard in the joint proposal will motivate the utilities to maintain or improve call center performance, and provide a benchmark against which performance can be measured.
9. The joint proposal focuses on facilities within the utility's control.
10. The joint proposal will have no significant immediate implementation costs.
11. The joint proposal will facilitate the Commission's investigation of utility performance during major outages.
12. Imposition of the restoration standard included in the joint proposal does not contradict any FERC-approved tariff, rate, term or condition of service.
Conclusions of Law
1. The joint recommendation is reasonable, cost-effective, and likely to have its intended effect.
2. PG&E's proposal is not reasonable or cost-effective.
3. The issue of the Commission's jurisdiction over transmission was decided in D.99-08-028. In that decision, we determined that we share concurrent jurisdiction with CAISO over elements of the transmission system and transmission reliability and we retain extensive jurisdiction over transmission and reliability including undisputed jurisdiction over safety.
4. In the future, if the utilities believe that the joint proposal will create significant implementation costs, they are not prohibited from seeking prospective cost recovery in future proceedings. Other parties may challenge the recovery requests.
5. The Commission has the jurisdiction necessary to adopt the joint proposal.
6. The joint proposal should be adopted.
IT IS ORDERED that:
1. The joint proposal, included as Attachment 1, is adopted.
2. This proceeding is closed.
This order is effective today.
Dated May 4, 2000 , at San Francisco, California.
LORETTA M. LYNCH
HENRY M. DUQUE
JOSIAH L. NEEPER
RICHARD A. BILAS
CARL W. WOOD
ADDITIONAL PROVISIONS TO G.O. 166
Addition to "Definitions"
Measured Event: A Measured Event is a Major Outage (as defined herein), resulting from non-earthquake, weather-related causes, affecting between 10% (simultaneous) and 40% (cumulative) of a utility's electric customer base. A Measured Event is deemed to begin at 12:00 a.m. on the day when more than one percent (simultaneous) of the utility's electric customers experience sustained interruptions. A Measured Event is deemed to end when fewer than one percent (simultaneous)of the utility's customers experience sustained interruptions in two consecutive 24-hour periods (12:00 a.m. to 11:59 p.m.); and the end of the Measured Event in 11:59 p.m. of that 48-hour period.
Transmission Facilities: Transmission facilities are those facilities subject to control by the Independent System Operator pursuant to Federal Energy Regulatory Commission orders.
Standard 12: Restoration Performance Benchmark For A Measured Event
The Commission shall perform a review of utility performance following every Major Outage. This standard sets a benchmark for the Commission to use in reviewing utility restoration performance only during Measured Events.
A utility's restoration performance during a Measure Event shall be presumed reasonable if the CAIDI is 570 or below, and presumed unreasonable if the CAIDI is above 570. These presumptions are rebuttable.
B. CAIDI Calculation
CAIDI stands for Customer Average Interruption Duration Index and is computed using the following equation:
total customer minutes of interruption
total number of customer interruptions
If a single customer experiences more than one sustained interruption during a Measured Event, each interruption shall count as a separate customer interruption. CAIDI shall be measured from the beginning of the Measured Event and shall continue until all customers experiencing interruptions during the Measured Event have been restored.
C. Transmission Outages
Customer minutes of interruption caused by outages of Transmission Facilities owned by the utility during a Measured Event are included in the calculation of CAIDI for purposes of this standard.
Customer minutes of interruption attributable to utility compliance with ISO directives, including its protocols, tariffs, transmission agreements or other written or verbal instructions specific to the event, which prevent the utility from restoring service it is otherwise able to provide shall be excluded in the calculation of CAIDI for purposes of this standard.
Utilities with fewer than 150,000 electric customers are exempted from application of this standard.
Standard 13 - Call Center Benchmark For A Measured Event
The Commission shall perform a review of utility performance following every Major Outage. This standard sets a benchmark for the Commission to use in reviewing utility call center performance only during Measured Events.
A utility's call center performance during a Measured Event shall be presumed reasonable if the percent busies calculation is lower than Level-1, and presumed unreasonable if the percent busies calculation is greater than Level-2. These presumptions are rebuttable. Performance equal to or between Level-1 and Level-2 is subject to no presumption.
Level-1 is defined as 30% busies over the day of the outage (12:00 a.m. to 11:59 p.m.).
Level-2 is defined as 50% busies over the day of the outage (12:00 a.m. to 11:59 p.m.) plus at least 50% busies in each of six one-hour increments (these increments need not be consecutive).
B. Percent Busies Calculation
Percent busies calculation measures the levels of busy signals encountered by customers at the utility's switch and that of its contractors. Mutual aid partners are not considered "contractors" for purposes of this standard, and busies encountered as a result of mutual aid assistance are not included in measurements to which this standard applies.
Percent busies indicator is measured on a 24-hour basis for outage-related calls (on energy outage and general call lines) from the time the Measured Event begins (12:00 a.m. to 11:59 p.m.), and separately for each 24-hour period until the Measured Event ends.
Either of the following methods for calculating percent busies is acceptable:
· Percent of call attempts reaching the utility which receive a busy signal
· Percent of time that trunk line capacity is exhausted.
Utilities with fewer than 150,000 electric customers are exempted from application of this standard.
(END OF ATTACHMENT 1)