Weather Company's Machine-Learning Analysis Model Helps Brands Prep, Respond
Outage Prediction model will aid utility companies to prepare and respond to weather-related events
The Weather Company
The Weather Company, an IBM Business, has officially launched an Outage Prediction model that will aid utility companies to prepare and respond to weather-related outages.
The model, which is uniquely tailored to each utility company's response plan, helps predict the number of outages based on incoming weather forecasts, as well as, the level of mobilization necessary to respond.
This kind of model will enable utility operation teams to begin planning for outages in advance, as much as 72-hours ahead of an anticipated weather event, which will not only help companies control costs but improve consumer's restoration time.
“In the past, utility companies were forced to respond within hours to a weather event, forcing them to use a considerable amount of cost and a lack in restoration time,” says Maia Sisk, director, offering management, location and new markets at The Weather Company. “However, with our machine learning analysis of historical information and up-to-date forecasts, the Outage Prediction model can help companies plan and prepare to respond and restore.”
By combining the information of historical events and current forecasts, the utility company can see what areas are predicted to be hit the hardest, which is critical in deciding where and when to pre-stage restoration crews and equipment.
“If a utility company is able to pre-stage or pre-plan for a weather event, it gives them a greater advantage in restoring the services quicker and cutting costs on the response,” says Sisk.
For utility companies using the model, the interface will provide a three-pronged approach:
- A real-time, automated storm prediction
- Historical storm searches, whereby companies can compare current forecasts with past outages
- Plan for scenarios, using a variety of weather factors
“This doesn't just go for severe weather events, but also less severe weather, such as high winds and downed trees, both of which can disrupt a consumer's utilities and have a cost impact for the company,” says Sisk.
The model will also account for parameters such as atmospheric pressure, soil moisture, foliage, types of vegetation, weather regime, and system design.
The Outage Prediction model is in line with The Weather Company's recent push to provide event-based data to companies.
The collaboration also delivered the Weather Company's real-time data on LiveRamp's IdentityLink Data Store to give advertisers an opportunity to apply data as a tier for decision making, such as using environmental conditions, severe weather, and relative weather conditions to improve their personalized marketing.
The result of the LiveRamp partnership gave brands and marketers the opportunity to develop better campaign outcomes, as well as, ensure the right message reaches the right customer at the right moment.