Skip to main content


Electricity demand forecasting in agriculture – harvesting the synergies of machine learning and survey data for electrification planning in Ethiopia

This project assesses how electricity utilities in developing countries can develop effective electricity demand forecasts and stimulation techniques for productive uses in agriculture, using the example of Ethiopia. For demand forecasting, we will apply an approach that combines innovative machine learning techniques with classical on-the-ground’ surveys among households, enterprises, and communities. The aim is to develop a scalable cost-effective approach that uses relatively ‘expensive to collect’ information from on-the-ground surveys in a limited number of regions to train machine algorithms how to extrapolate this in-depth information to the entire country. In an additional component, we will use the collected and generated data from the demand forecasts to derive data-driven suggestions for demand stimulation interventions. Embedding the knowledge of local policy agencies, we will map the demand for new irrigation and agro-processing potentials with an eye on most promising value chains. We thereby intend to contribute to the growing literature on techno-economic planning tools to design electrification strategies. Effective planning tools are pivotal to define the most suitable electrification strategy for different sub-national regions, comparing the extension of the national grid to the establishment of a mini-grid or other off-grid systems. This research is part of the Applied Research Programme on Energy for Economic Growth (EEG) led by Oxford Policy Management.
The programme is funded by the UK Government, through UK Aid.

Applied Research Programme on Energy for Economic Growth (EEG)

Project start:
01. March 2019

Project end:
31. December 2021

Project staff:
Prof. Dr. Jörg Peters, Dr. Gunther Bensch

Project partners:
University of Massachusetts Amherst, Policy Studies Institute

UK Aid