ACET Uses Artificial Intelligence to Predict Future African Infrastructure Needs

March 3, 2021
ACET, working with partner Omdena and nearly 40 data scientists and machine learning experts from around the globe, recently completed the continent’s first Artificial Intelligence Challenge to help predict what infrastructure Africa will need in the future.

The exercise sought to identify machine learning tools and approaches that can inform policy decisions. The data scientists created models and designed methodologies that could help determine what infrastructure to build, where to build that infrastructure and what factors will impact long-term economic impacts.

The AI Challenge lasted 10 weeks and focused in particular on the use of machine learning techniques with computer vision, natural language processing, and exploratory data analysis (EDA). These tools allow new ways to source and utilize data on scales not available in the past. For example, EDA is used to identify what model better fits the data and whether any data cleansing might be required before putting it through machine learning and artificial intelligence algorithms.

As part of the challenge, ACET and Omdena used these tools to explore application for water infrastructure modeling and also for “distance to something” modeling for transport links to service provision such as schools and hospitals. This innovative work complements ACET’s other infrastructure-related programs, including support to the G20 Compact with Africa initiative to accelerate infrastructure investment in Africa and a collaboration with the African Union Development Agency (AUDA-NEPAD) and the OECD to improve infrastructure project cycles.

The data scientists looked at a myriad of data sources such as satellite images, socio-economic data, climate and topological data, population and demographic data, Google Trends, Google business data, and social media data too understand aspirations, needs, and sentiments of people living in the region.

They sought to build one or more models for the future infrastructure needs of Africa, while providing recommendations regarding verification approaches and networks to help scale across Africa.

The rationale for the challenge was grounded in the severe fiscal restraints faced by African governments, particularly in the context of the economic impacts from COVID-19. African governments are using significant portions of public budgets to finance infrastructure, but that infrastructure often responds to past or current needs, not future needs based on expected changes related to climate change, migration, or urbanization, among other factors.

Using artificial intelligence can be an important first step to addressing Africa’s infrastructure gap by allowing policymakers to benefit from having better information at their disposal to make more informed decisions about what infrastructure to build where and when—and to ensure that current projects are able to serve Africa’s growing population for the next 50-100 years.

ACET’s partner, Omdena is a global platform bridging mission-driven organizations with artificial intelligence engineers, data scientists, and domain experts from diverse backgrounds to solve real-world problems. They provide streamlined collaborative environments to build innovative, ethical, and efficient AI and data science solutions by hosting AI projects on their platform and building  an interdisciplinary and collaborative team of changemakers to develop solution.

Share this page


  1. Raluca Pauna

    can you tell me who are the data scientists and researchers who were involved in this project? I argue that there are many of them based in Africa and ready to be used in these kind of projects.

    It is important to use local human resources for the sake of inclusivity and for training the large number of university graduates.

    • John Osei

      Dear Raluca,
      We were very lucky to have nearly 50 data scientists from around the world working on the project. To do so we used the platform Omdena.

      Omdena is a platform to collaboratively address critical AI challenges. You are correct, we had a number of the data scientist from Africa, as well as Mexico, Philippines, Switzerland, USA, etc. We agree fully that it is important to use local skills and human resources when possible. Thank you for your comment.

  2. Dennis Mwighusa

    Wonderful, Using local experts has the advantage in the context of local content, i.e., understanding the local environments and develop solutions that will better serve the public. However the foreign expertise can not be neglected since they provide inspiration and guidance.


Submit a Comment

Your email address will not be published. Required fields are marked *