How Utilities Use Data to Weather the Storm

Image courtesy of Luc Legay under Attribution-ShareAlike 2.0 Generic License, resized to 700 x 391 pixels.

Data is the lifeblood of storm assessment, response and recovery.  There’s no way around it, utilities cannot maximize reliability performance and outage restoration speed without good data.  There are many ways that utilities leverage information, and for today’s post I thought we could take a look at a few examples.

Examples of How Utility Companies Use Data to Weather the Proverbial Storm

According to this article from my friends at Utility Dive, the most efficient and effective utilities are the ones that leverage data to optimize decision-making.  Here are 2 relatively obvious examples:

• Before a storm, weather information is useful for obtaining an early warning sign of an impending threat.  This includes information around weather-related plant shutoffs, the impact of weather on customer demand, and more.  Much of this insight can be garnered by analyzing patterns of historical storm activity and looking at the data in the context of its probability of impact based on historical results.
• After the storm, electricity market data is useful to identify supply-purchase options when an outage occurs. It is also necessary to analyze the damage (i.e., damage assessment), optimize crew dispatching, communicate to customers, employees and regulators, and track progress.  Finally, analytics can be utilized to conduct post-mortems and identify lessons learned.

As I’m sure you know, these are just a few of many possible examples.  The thing about data is, the more you have the better it can be analyzed, but steps must be taken to ensure accuracy.  The old adage, “garbage in, garbage out” is an effective way to make my point on this.  Not only can limited information reduce your ability to see the big picture, worse yet, it could lead to poor decisions.

Therefore, the bottom line is that data in and of itself is not enough.  It must be comprehensive and accurate, and utilities must have the systems in place to analyze it properly.  If these characteristics are not in place, data is simply noise.