Apps Market - Forecasting


This App forecasts your consumption and costs for the coming months and compares them with your goals.


Getting Started

Note: you should be superAdmin or Admin to configure the Forecasting App. Otherwise, contact the administrator of your account.

The first step is to configure your forecast and allow the system to do some calculations based on temporal series algorithms. To do so, you should go to Settings > Forecasting > and click on "New Single Forecast". Here you can define the device for which you would like to generate a forecast. Just select the associated location, the device and the corresponding parameter. Below you have an example:

The app is available for the following parameters:

  • Active Energy (402)
  • Gas Energy (420)
  • Water (901)
  • Exported Active Energy (452)
  • Thermal Energy (802)

Once you have configured a new forecasting, you will receive the forecast in a new datapoint in the chosen device, which you should accept to visualise the forecast on other screens (see this article on how to accept datapoints). Note that when the datapoint is accepted, you will be able to plot the forecast on any screen that allows the forecasted parameter.


Note. Take into account that 6 months of readings are assumed necessary to be able to accurately build the forecast. 


Add/Edit baselines

You may want to compare your forecast consumption against your goals or targets as a baseline. Here you can add up to 5 baselines to compare with your forecast consumption. The baselines will have a monthly resolution.


The model behind the Forecasting App

The model is based on artificial intelligence algorithms that use the consumption (or production) historical behaviour together with calendar parameters: month, day and time.

The algorithm calculates a consumption forecast for each future hour, based on similar periods considering similar days/hours/months from the last year of available historical data. The forecast is calculated for a given year (365 days).


The more historical data available, the more reliable the model's prediction will be.


Analysing the results

Once you have configured your forecast information, you should wait a bit. Forecasted data is calculated every day (03:00h GMT) for one year in advance. The system takes between 1 week and 3 months to provide accurate models. 
You can plot the forecast in any screen that allows the Forecasted parameters, but you can also check the Forecast in the app's screen:

In this screen, we can see the following elements:

  1. Energy source selector. Select the energy source you want to analyse
  2. Dates range picker. Select the period of time for the analysis. You can use the presets available to move quickly (today, yesterday, last 7 days, last month...)
  3. Location. Select the location for analysis.
  4. Device. Select the device you want to analyse
  5. Frequency. Select the frequency between daily, weekly and monthly
  6. Compare against these Goals and Budgets. Here you can select the baselines previously configured to compare them against the forecasted data.
  7. Cost. Convert the forecasting directly into Euros. We take the average price for the selected period to do the calculations
  8. Export data to Excel. Export the selected data to an Excel file.
  9. Bars chart. View your data in bars
  10. Accumulated view. Analyse your forecasting aggregating all the consumption for the selected period.
  11. Summary table. Comparison between the real consumption, the forecast, and all the baselines selected if any. Here you can see the total consumption and the difference with the real consumption.

Forecasting Reliability

What is the forecasting reliability and how can we see it in the platform?

The Forecasting Reliability Metric allows you to assess how much you can rely on the forecast. It consists of a tag with 5 possible values. The main factor that determines the forecast reliability is the data quality of the device consumption curve. No data gaps, an understandable pattern and lack of outliers are factors that would increase the data quality and ensure better forecasting results.
To make it more comprehensive, when you hover your mouse over the tag, a tooltip pops up with information relative to the resulting metric. The possible values are:
  • High
  • Medium
  • Low
  • None
  • Error

The difference between your forecast and your data is less than 10%. You can rely on this forecast.


The difference between your forecast and your data is between 10% and 20%. It is an acceptable forecast. To increase the reliability the main recommendations are to include more historical data and to check outliers in the consumption.

The difference between your forecast and your data is greater than 20%. This means that the data used to train the model did not allow us to obtain good results. Before working with this forecast we would advise you to check if the data quality can be improved.
There is not enough data to calculate your forecast or your profile has too much variability.

Something went wrong during the calculation of the forecast. Please, contact your support team.


How is the forecasting reliability metric calculated?

The Forecasting Reliability Metric is based on statistical indicators that assess the quality of your data. Some of the analysed variables are:

  • The amount of zero values in the historical consumption
  • The amount of missing values in the historical consumption
  • Chaotic pattern changes
  • Outliers

The better the quality of your data is, the more grounded the forecasts will be. The score obtained determines the value of the Forecasting Reliability Metric.

For reliable predictions, it is recommended to have 1 year of historical data. This way, the algorithm will have the reference of the behaviour in summer and winter. With fewer data, it is possible that the prediction does not correspond to the specific behaviour in those periods. 



Was this article helpful?