A risk and disaster management tool using artificial intelligence (AI) has been developed by Zambian agri-tech company, AgriPredict, to help smallholders overcome agricultural challenges, such as droughts, pests and diseases.
Using the AgriPredict web and mobile-based platform, Zambian farmers are able to take photos of a suspected diseased plant/animal in order to receive a diagnosis, options for treatment, if required, as well as location of the nearest agro-dealer. Through machine learning, the tool can also predict pest and disease occurrence, as well as forecast weather patterns, enabling users to take preventative measures to limit potential crop and livestock losses.
AgriPredict was founded as a result of Tuta absoluta, a tomato disease, and advent of Fall Armyworm that caused significant crop losses for Zambian farmers in 2016. “We noticed that there were no tools to help farmers mitigate, prevent or even counter these challenges,” says Mwila Kangwa, AgriPredict CEO.
The appeal of AgriPredict lies in how the company has tailor-made the platform to enable easy access to a wide range of farmers; the information is easily accessible via web services, basic mobile phones (USSD and SMS), smartphones (Android and iOS) and tablets, as well as via a social media chat bot (Facebook). “The platform is built by farmers, for farmers. Our vision for AgriPredict is to become the biggest data hub in Africa, providing proper insights to enable farmers to make better informed decisions,” says Kangwa. The platform is currently under further development to include both voice and visual features to overcome digital literacy barriers.
“AgriPredict has been cost saving; it tells me what disease is affecting my tomato. Before, I would have had to employ a full-time agronomist to help with the running of the farm, but ever since we started using AgriPredict, it cut down that cost completely,” says farmer Yunike Phiri.
Initial development of the platform was made possible with support from the SNV Netherlands Development Organisation for a pilot project focusing on Zambia’s tomato and maize value chains. The pilot, which was launched in October 2018, involves 22,000 farmers in the Eastern Province and included training sessions on app utilisation, site visits and gathering feedback from farmers. A social media campaign ‘#coolfarmers’ is also underway to promote the technology and increase interest among youths.