Over 2 billion people live with unreliable electricity.
More than 210,000 Mini-Grids are required to achieve the number 7 Sustainable Development Goal: Ensure access to affordable, reliable, sustainable and modern energy for all.
Identifying remote villages for off-grid electrification and gathering useful information about them is a key barrier.
Surveys are slow, costly and imprecise. This leads to long project development timelines, low operational margins and restricts access to finance.
Scaling off-grid electrification means finding portfolios of commercially viable sites.
We need a reliable, fast and scalable method to identify portfolios of commercially viable sites and to make the information available to development organizations, government, donors and energy companies.
VIDA: Village Data Analytics
A software-enabled service that automatically identifies remote villages and determines its suitability for a mini-grid installation. VIDA uses machine-learning algorithms to predict socio-economic health of a village. This data-driven knowledge de-risks projects, reduces time and costs for viable off-grid planning and investment at scale.
How does it work?
Machine Learning Algorithms in VIDA
appliedAI supports TFE Energy in developing the AI capabilities that power the software's automated analysis and ranking. The algorithms behind VIDA are made up of a combination of state-of-the-art supervised and unsupervised Machine Learning methods. The technical setup of VIDA allows for retraining of algorithms when new data arrives into the system over time.
Based on an area of interest in a remote region, VIDA's custom AI algorithms identify villages that are off the electrical grid. These algorithms make use of satellite imagery and other publicly available data sets.
A second set of Machine Learning algorithms then use the identified village boundaries and predict village characteristics such as size, density and socioeconomic makeup. The algorithms are trained with a combination of satellite imagery and custom VIDA datasets that are based on proprietary data.
Finally, a ranking of villages is computed based on village-level characteristics. Users can filter for specific criteria and highlight or save villages that are interesting to them for further analysis.
We could not have asked for a better partner to develop Village Data Analytics with than appliedAI. appliedAI is a fast, solution-driven, creative, and structured partner. They are not only clearly at the forefront of what AI can do, but also have very practical ideas of how to use it to create value for Village Data Analytics. It is a pleasure to work with the team.Tobias Engelmeier, Founder and Managing Director of TFE Energy