April 24, 2024

Outage Management Proves Its Value During Hurricanes
- One hundred percent of a utility’s customers without power.

by By Terry Nielsen, SPL
That would have been an unthinkable nightmare just a few months ago. Today, we all know it’s a reality that more than one utility faced as a result of Hurricane Katrina.

Rebuilding the devastated infrastructure will take months. But reconnection has gone remarkably swiftly. Just 12 days after Katrina ended, for instance, Mississippi Power reported that all customers that could accept power had been reconnected.

That achievement results from many factors, including long-standing preparedness practices. Highly trained local crews, planning, drills, and crews from utilities outside the affected region all helped. But in the past decade, an additional factor has significantly speeded recovery from widespread power outages: implementation of outage management systems (OMS).

________________________
How OMS Works
Typically, outages are small and localized. A curious squirrel trips a breaker or a careless construction worker cuts a line. In this case:

• The outage management system (OMS) goes into action based on customer calls or on electronic notifications from an automatic meter reading and monitoring system.

• It pinpoints the problem location as closely as possible based on the pattern of calls or notifications. A controller dispatches a crew, using the data from the utility’s Geographic Information System (GIS). The crew reaches the problem and reports details.

• The OMS uses those reported details to analyze the number and type of customers affected. It identifies other parts of the system that might be involved. And it prioritizes restoration options based on factors like critical needs, available crews and equipment, and contractual penalties for extended outages. In this example of a localized outage, for instance, the OMS would identify the switches whose opening and closing will restore the most power to the highest priority and largest number of customers. The controller, however, makes the final choice of restoration plan.

• Now the OMS moves into modeling mode, tracking all changes made to a system. This tracking is a safety device, ensuring that work proceeds in ordered steps that protect workers from the consequences of unexpected power flow.

During major storms like Katrina, these steps undergo some modification. Outages are not treated as individual occurrences. They are instead amalgamated into an overall assessment of system damage. In this way, as the storm wanes and crews can be safely deployed, the highest priority customers get initial attention, and control centers can appropriately prioritize repairs over the entire distribution system.

Why Use OMS?

Not every utility uses OMS. But those that do report substantial improvements in customer service statistics and restoration costs. Here are some typical results:

• A utility in southeastern Canada reports a 25% drop in aborted switching operations.

• A southern US utility reports an 80% improvement in crew dispatch accuracy.

• A utility in a sparsely populated western state reports that average outage times have fallen from four hours to one.

• An eastern US utility was able to restore power to all customers within 8 days rather than a predicted 10 without OMS.

• A UK utility reports a 45 percent drop in customer minutes lost, a 10 percent drop in faults, a doubling of fault progression information, and a 20 percent reduction in repair costs.
• A recent survey shows OMS reduced staff needed for maintenance and support fell 0.5 to 1.5 FTEs (full-time equivalent employees) as compared with the staffing needed for legacy systems.

• A similar survey showed reductions in the System Average Interruption Duration Index (SAIDI) of 5 to 15 minutes.
________________________

OMS pays for itself during “ordinary” emergencies like wind and ice storms that impact large numbers of customers, where as many as 50 or 75 percent of a utility’s customers lose power. They are invaluable in:

• Keeping track of what is restored and what is not.
• Running damage-assessment routines that give utilities an accurate assessment of the resources required to manage the restoration.
• Prioritizing restoration activities based upon needs like medical facilities, critical
infrastructure such as 911 centers, police, life support customers, etc.
• Supplying information about the current status of the outage for customers, the media, and concerned friends and relatives outside the affected area.
• Providing estimated restoration times based on known damage and available resources.

OMS users have proved eager to share their experience with others. And the consequent body of knowledge and best practices that have accumulated over the past few years proved invaluable in dealing with the unprecedented emergencies of Hurricanes Katrina and Rita.

Preparation
Twenty years ago, hurricanes hit with only the most general warnings to affected areas.

Improvements in meteorology today track hurricane paths more accurately and provide close estimates of their likely strength. The OMS-using utilities of the Gulf region had time to put emergency preparedness plans into action.

First on the agenda was preparing the OMS for heavy use by doing some routine maintenance. These preparations tuned the system for the storm, much as a homeowner puts up storm shutters and makes that last-minute trip to the grocery store for batteries and bottled water. In the case of the OMS, preparations anticipate the larger than normal number of people who will be using the system and the large amount of data that will accumulate. Utilities typically make sure the database is properly tuned, deploy laptops to temporary users, and make sure everyone’s training is refreshed.

Storm preparations also involve a check with the OMS vendor to ensure that support is ready. During emergencies, vendors can make sure that the experts in the system are on standby and are familiar with the systems that are most likely to be impacted. Vendors can discuss with utilities any changes to the system that might have been made recently. Vendor experts may also review preparations with the utility team and offer suggestions based on the experiences of other clients in similar situations.
In the past, vendors frequently sent OMS gurus to be on-site. But when a storm is predicted to affect a wide area, it may be more valuable to keep the gurus back in their home offices where they can support multiple utilities. Delaying deployment also permits vendors to send people to the sites of greatest need—locations it may be hard to predict accurately prior to the storm.

As systems become more robust, the need for on-site vendor gurus becomes less critical or even redundant. Our own experience during Katrina, for instance, was that none of our clients needed anything more than phone consultation.

• Bar chart of a UK utility’s rise in 24-hour restoration percentages (data in the right column) for successive storms (data in the left column).

Dec. '98 - 61
Feb. '01 - 94
Jan. '05 - 95
Feb. '05 - 100


Initial Assessment
As storms wane, the OMS typically focuses on damage assessment. How many poles are down? How many lines, transformers and switches are damaged. This knowledge helps determine the number of crews, the type of equipment, and the quantities of materials that restoration will require.

In parallel, utilities line up ‘mutual aid’ from other utilities whose crews are available to assist in the restoration efforts. Tentative assignments are made so that out-of-region crews have ready access to the knowledge and experience available only from local crews.

The first outside help will likely come from affiliate utilities in the same parent company or regional cooperative association—at least, those outside the areas affected by the storm. It’s particularly easy to use that help when it arrives from utilities using identical systems. In Katrina, for instance, Georgia Power easily provided Mississippi Power not only with crews but also with additional dispatch resources from Georgia Power using the IT and communication infrastructure of the parent, Southern Company. This was possible because both utilities use SPL OMS. As dispatch practices become more standardized, of course, this type of aid will likely be possible from additional utilities, not just those with the same parent holding company.

Phased Restoration
Once crews could safely begin work, the OMS provided a prioritized list of outages to be restored.
First on the list, of course, were sites where live downed wires presented safety hazards and where critical infrastructure had been damaged. An OMS handles this by presenting to the dispatch users a prioritized list of work-based rules established long before the storm, typically during system deployment.

Second are outages that affect large numbers of customers. Because the OMS contains an
accurate, real-time model of the utility’s distribution system, it can track how many customers are affected by the identified failures. As restoration proceeds, OMS users update the model to keep track of who is restored and who is not. This provides a good picture of what is going on and allows the utility to track the progress of restoration region by region. The OMS also provides accurate updates that utilities can pass on to the media and customers.

In parallel with the early restoration activity, field crews undertake detailed damage assessment, identifying the locations, types and severity of damage. This tedious survey covers every electrical circuit across the entire affected territory. At the best-equipped utilities, employees can enter survey results directly into the OMS using laptops or mobile computers in the field. Other utilities arm field crews with paper forms that are then keyed into the OMS each evening.

Putting the damage assessment information into the OMS helps remove the “fog of war”. Utilities can draw up better “battle plans” based on a more detailed picture of the work needed

The final phase of restoration addresses damage that affects only one or a few customers. In regions with light damage, this generally takes place within 24 hours. But in the case of Katrina, power restoration in many cases must be preceded by building repairs that could take months. Clearly, some customers will not rebuild, and there will be no restoration.

Retrospective Assessment
Because an unprecedented number of utility customers lost power during Katrina, the storm put outage systems through paces that laboratory exercises—no matter how extensive—can never fully provide. Snapshots of databases that utility employees took during restoration efforts will help vendors clear up glitches and make the OMS of the future even better. Comparison of the damage and actual restoration efforts against the early projections, for instance, can help the utility tune its prediction model and resource forecasting models.

Utilities are likely to find that changes need to be made in other IT applications, too, especially in asset management. Only a few utilities, for instance, use the electronic tagging that can pinpoint equipment location and speed return. Even fewer can readily change equipment maintenance cycles from those based on average use to those that keep equipment running when usage becomes 24 hours per day, seven days a week, for weeks on end.

Pulling Together
The most inspiring lessons of Katrina are those of the human spirit. Utility employees performed their jobs despite damaged facilities, long hours, and life-threatening conditions. Volunteers were plentiful; at Mississippi Power, for instance, more than 200 employees from other Southern Company utilities moved into the area to provide relief. Even more remarkable are the many who worked for the good of their communities when they themselves had lost all material possessions and in many cases were uncertain about the fate of family members.

But we also learned just how important OMS technology is in underpinning these efforts. A tried and tested outage management system, employees trained in its use and in the restoration process, and accurate OMS data can dramatically increase the efficiency of outage restoration.