Advanced forecasting tools reduce risk, improve bottom line


Once primarily known for meter data collection and automatic meter reading, Itron has taken its data collection skills – its knowledge – and expanded them across the entire workforce (field and office), sharing that knowledge with other market participants for other energy solutions.

The ability to manage and minimise the risks associated with forecasting energy takes accurate data to meet the needs of customers on both a near-term and long-term basis. Here is how one European company, Luminus, used Itron solutions to better predict their energy needs.

Luminus is an energy retailer serving residential, commercial and industrial customers in Belgium. The company is well positioned to take advantage of the growth opportunities presented by Belgium’s liberalised energy market and is the second largest supplier of gas and electricity in the Belgian marketplace.

Liberalisation of the retail energy market allows consumers to choose their energy supplier. The choice depends on a variety of reasons, including the supplier with the most competitive price, the best service, or one that offers a supply portfolio of environmentally favourable power.

Since the beginning of 2002, Luminus has been supplying electricity to about 550 large commercial companies in Belgium and as from July 2003, to 850,000 residential customers. A select number of energy suppliers have been designated as ‘standard suppliers’ to ensure that the energy consumer receives a reliable, regulated supply of affordably priced power. Luminus is one of the ‘standard suppliers’.


As a retailer operating in a liberalised market, Luminus must consistently manage and minimise the risks associated with an unpredictable wholesale energy market. To avoid costly errors, it is crucial for Luminus to accurately forecast the energy required to meet the needs of its customers on both a near-term and longer-term basis.

Luminus serves customers whose energy consumption is measured by interval meters, as well as non-interval-metered consumers, and thus requires flexible forecasting methodologies to meet the requirements of the different customer segments. Non-interval-metered consumers – most often residential and small commercial customers – are settled using load profiles approved by the Flemish (regional) regulator VREG. For these loads, Luminus needed a method that accurately forecasts the load profiles that will be used to settle customer bills. For interval metered customers, Luminus must generate forecasts based on their actual individual loads.

To meet these critical business objectives, Luminus selected Itron to deliver a sophisticated energy forecasting approach that was flexible, modular and configurable – one that would enhance its forecasting capabilities and reduce imbalance charges. In choosing Itron, Luminus found a strategic partner that offered both state-of-the-art software tools and industry-leading forecasting knowledge to develop and implement the forecasting solution necessary to comply with new market directives.

Luminus chose to implement Itron’s Retail Forecasting System. Itron’s automated system is configured around MetrixIDR™ and its advanced statistical modelling engine MetrixND®. The system provides short-term load forecasts for Luminus’s electricity and gas customers.

“Itron’s system provides us with load forecasts of how much energy we need today, the next day, and ten days into the future,” said Gunter Walmagh, Luminus’ Operations Manager, Energy Management Group. “We use that load forecast in several ways, but foremost, it helps us know how much energy to buy. The accuracy of the forecast directly affects our bottom line.”


Luminus also selected Itron’s forecasting solution because it could meet requirements for both natural gas and electricity. Itron’s solution, which uses a combination of neural network and regression models, provided Luminus with an automated forecasting system configured around Itron’s advanced statistical modeling engine MetrixND.

Itron’s retail forecasting system automatically integrates the forecasting models with weather forecast data, calendar data, customer portfolio data and interval metered data to generate the load forecasts. The load forecasts are then presented to the Luminus forecast analysts for their review. The system also allows users to import different weather data for analysis purposes.

Itron’s forecasting consultants worked closely with Luminus’s analytical and IT departments to clarify the inputs to and the outputs from the forecasting system.
“Luminus is a young, savvy, growth-oriented company that recognised early on the value of having an accurate forecasting system,” said Frank Monforte, Itron’s vice president of forecasting. “We’ve shared our knowledge from working with energy retailers around the world to help Luminus in an energy market where energy companies need to have a ‘heads-up’ to survive and thrive.”

As a result of the collaboration, Luminus is able to implement an accurate forecasting system that meets their short-term forecasting requirements for both electricity and gas customers.