November 17 Data Science Community – Midterm Load Forecasting with Reviewed, Explainable Forecasts
Join the Data Science Community on Tuesday, November 17, 2026 at 1:00 PM CT for a presentation with Q&A/open discussion on, “Midterm Load Forecasting with Reviewed, Explainable Forecasts” led by Mehdi Ganjkhani, Ph.D., Senior Data Scientist, Analytics at The Energy Authority.
Session Description:
Midterm load forecasts, spanning weeks to a year ahead, support critical procurement, hedging, and maintenance decisions. However, they face two major challenges: uncertainty in future weather conditions and structural load growth that cannot be explained by weather alone.
This presentation introduces a midterm forecasting framework that addresses these challenges by separating long-term load growth from weather-driven demand. The approach combines growth estimation, gradient-boosting models, and weather forecasts from multiple horizons. It also incorporates strategies for handling extreme temperatures beyond historical observations.
Generative AI supports the framework by reviewing forecasting decisions within predefined constraints, while deterministic code handles calculations and quality checks. It also generates plain-language explanations that attribute forecast changes to growth, weather, and calendar effects, helping users understand forecast drivers and identify cases requiring further review.
Using an anonymized utility load zone, the session will explore the modeling methodology, backtesting results, the impact of weather uncertainty, and practical lessons learned.
*UAI Utility Members can register and access all nine of our Analytics Communities via our Community Page. Not a part of the Customer Analytics Community and want to join this session? Please use our direct link here to request access. You will find the Teams meeting credentials to the upcoming meeting at the top of the feed on this page. If you have any questions or run into any issues, please contact us at info@utilityanalytics.com.