BigDAPESI

Big-data analysis and forecasting of energy consumption and refurbishment costs for real estate

BigDAPESI

The BigDAPESI project aims to develop a big-data concept for the analysis and forecasting of energy consumption and refurbishment costs in real estate while maintaining a high level of protection of individual user data.

Name BigDAPESI
Funding Bavarian Ministry of Economic Affairs, Regional Development and Energy
Project Management VDI/VDE
Funding number according to notification IUK-1606-0002, IUK491/001
Start September 2016
Ende August 2018
Project duration ​​2 years
Project lead Prof. Dr. Gilbert Fridgen

Prof. Dr. Gilbert Fridgen

PayPal-FNR PEARL Chair in Digital Financial Services

Basics

The analysis and forecast of energy consumption at BigDAPESI is based on the processing of large amounts of building physics and consumption-oriented, structured and unstructured data. For this reason, in addition to data storage and data access, a large part of the research efforts will be focused on data processing and visualisation. Furthermore, in order to guarantee the protection of privacy, existing methods for the analysis of large amounts of data must be adapted in such a way that they can be used in combination with privacy reserving methods.

In summary, the innovation of the project is based on the data-driven determination of energy consumption. The corresponding IT support aims to significantly outperform established methods of building physics in speed and accuracy and to undercut them in terms of cost. This enables an improvement of the conventional method and offers a completely new approach in the real estate industry.

Key Aspects of the Project

  • Increased speed and precision while reducing the cost of energy consumption forecasts compared to existing methods
  • Protection of privacy and anonymity as a central and innovative feature
  • Near real-time energy consumption forecasting to ensure high service levels and user acceptance
  • Processing of structured and unstructured data from different data sources