Intelligent cloud computing lifts villages out of water poverty
Articles,  Blog

Intelligent cloud computing lifts villages out of water poverty


>>:: I grew up in rural parts of Kenya, which have quite high rates of poverty. In our family, we used to get on our bicycles, each of us carrying three jugs, like three or four people, me, my uncles and my cousins. We would travel, like, 20 kilometers on a bicycle that day to go look for water.>>:: Four out of five people in the world don’t have access to safe drinking water live in rural areas. This situation is particularly compounded in Africa where hand pumps and rural water supplies often fail. And one of the newest technologies that’s now emerged in terms of mobile solutions and cloud-based computing is allowing us to advance with innovative ways to address this problem. The Reach program is aiming to make five million poor people water-secure in Africa and Asia.>>:: Groundwater is one of the safest supplies of water in rural regions of Kenya. So your being able to track the depth of the aquifers is critically important in being able to understand how healthy that water system is. We may be familiar with the idea of using one’s smartphone or smart devices to carry around with us and to use to monitor our own health. And the idea here was: Can’t we put some of those mobile health devices into the handle of a pump.>>:: So the data is really interesting. When you see the accelerometry parts, you kind of get these curves of the actual motion of someone pumping. When the pump is deeper, the weight of the water that you’re lifting is much larger. So there will be more vibrations; whereas, when the pump is shallower, you’re lifting less water, so there will be smaller vibrations.>>:: Imagine you have multiple intelligent nodes that are all transmitting data. You have to integrate data in a cloud-based system from data nodes across an entire region. Tens of thousands of pumps in our case that needs to be done in a cloud-based situation where one has a huge amount of computing resource to be able to perform the heavyweight machine learning algorithms on the integrated data. We like the inclusion of the Azure machine learning framework which allowed us to port our existing R and Python-based machine learning tools directly into a safe cloud-based system.>>:: Azure ML makes fitting machine learning models much faster as I can explore the parameters base much quicker on the cloud than just on my own computer.>>:: I think making data usable and effective to decision makers is absolutely critical. Policymakers and decision makers have the responsibility to take very difficult and onerous decisions. And there’s a huge information deficit in terms of how they should proceed. So the ability to collect data at scale in a way that’s usable and appropriate for them is extremely powerful.>>:: Initially, it’s used to take more than 30 days to repair a hand pump. But now, because we have this information, when a hand pump breaks down, we can repair them within less than three days or sometimes even 48 hours. That enables the villages to have better access to water. My hope and aspiration is to see this system making my village and other villages back in Kenya water-secure and moving them out of poverty. [Music]

Leave a Reply

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