As our industry already knows, moving gas or electrons from one point to another inherently carries some risk and there isn’t anything you can do to fully eliminate risk on your system.
Since we can never eliminate all risk, the traditional approach to risk has been to rank order risk relative to other assets: “Circuit A has worse SAIFI than Circuit B, so let’s invest in improving Circuit A until it’s performance looks more like the average.” As the session title says: This. Approach. Is. Wrong.
With data science you can fundamentally change the way you approach the entire concept of risk. By leveraging advances in data science, you can predict the elements of risk that are within your control versus the elements of risk that are innate to the system at specific locations. Instead of focusing on risk, you should be focusing on the risk-spend efficiency. In other words: you have limited budget, how do you get the most return on your investment in risk mitigation.
Once this concept of data science-based risk-spend efficiency is understood, you’ll be able to apply this approach to a variety of use cases including: vegetation management, circuit reliability, wildfire mitigation, gas pipeline O&M, storm outage prediction, capital deployment, and a dozen other focus areas.