Predicting energy consumption, a number of days in advance

Predicting Small to Medium Enterprises consumption, a number of days in advance to increase Planet 9’s revenue.

The Challenge

Planet 9 energy needed a model that could be integrated into their existing energy supply business to predict customer's consumption requirements a few days in advance. This would allow them to buy and sell energy from the market at the optimal times.

The Solution

The particular customers of Planet 9 had usage profiles that did not align with the typical profiles for their particular industry resulting in it being very challenging for Planet 9 to earn revenue. A number of predictive models for the half-hourly energy consumption profiles for UK SMEs were developed using machine learning, based on their historic usage. The models were tested against each other to determine with performed the best. The algorithm gave Planet 9 the ability to make more accurate forecasts related to their customers usage a few days in advance, this meant Planet 9 could minimise the costs involved hedging and thereby maximise their profit.

The Results

  • Planet 9 can now target customers that other energy suppliers generally avoid due to uncertainties associated with their usage
  • This is a win win situation for both Planet 9 and their customers

Contact the ESB Energy Insights Team Today, email,