How tech firm hopes to raise yields using data

17th Aug, 2019
How tech firm hopes to raise yields using data
Rather than use a mobile system, the tractor tracking system uses satellite technology.

Unfortunately, farmers on the African continent continue to lag behind due to lack of mechanised technology.

According to recent statistics, more than 60 per cent of crops in the continent are managed manually, with less than 20 per cent managed by tractors and other machinery.

Sadly, this creates an unsustainable model as food demand increases due to population growth that averages 11 million per year.

In addition, up to 50 per cent of farmers suffer post-harvest losses annually due to poor planting practices.

To stem this tide, one global technology firm is partnering with farming app developers in applying technologies seeking to transform local agriculture by shortening the time it takes to till the land, plant and harvest.

In a recent interview at their Nairobi offices, software engineer Andrew Kinai says one of the key partnerships involves corroboration with Hello Tractors, an Uber-like application that links tractor owners in different parts of the country with smallholder farmers.

Together with Hello Tractor’s developers, Kinai says the partnership will apply several technologies including the Watson Decision Platform for Agriculture, Blockchain, the Internet of Things (IoT) and IBM Cloud. 

Value chain

“The magic behind the idea is what we call an agriculture digital wallet, a blockchain-enabled and AI-based decision support platform that enables capturing, tracking, and instant sharing of data, while creating end-to-end trust and transparency for all the parties involved across the agribusiness value chain,” said Kinai.

When operational, the system will help farmers predict crop yields by using satellite-enabled remote sensing data and Internet of Things.

Kinai says data from tractors will be incorporated into the app to help small farmers know when to cultivate, the quality of their farm cultivation, what to plant, and what fertiliser to use.

In the future, he adds, IBM AgroPad technology developed at the firm’s lab in Brazil will also be incorporated to determine soil quality.

IBM’s agriculture interventions will not only be a boost to farmers but to tractor owners as well. Kinai says fleet owners will be able to view and manage tractor utilisation and predictive maintenance.  

“The system will forecast future tractor utilisations based on history, real-time weather data and remote sensing satellite data. Using a five-star rating system, tractor operators will be ranked and utilised based on their training in different areas such as plowing, deep ripping, harrowing, fertilising,” he says.

New tractors

In addition, historical data collected from the tractors will open up financing opportunities for maintenance and buying new tractors and other implements. 

Rather than use a mobile system, the tractor tracking system uses satellite technology, which is less problematic in terms of connectivity.

Instead of having farmers use the app directly, the system will utilise community-based agents who will book the tractors on their behalf.

“Initially, Hello Tractor tried to take reservations via SMS, but the farmers didn’t like to transact large amounts of money via SMS and this is how the agent model was developed,” says Kinai.

According to Kinai, the same value chain will help tractor dealers improve tractor repair and servicing, after-sales support, spare part inventory planning and credit administration.

In the next phase of the project, IBM researchers and Hello Tractor engineers will try to use machine learning with image recognition to predict the quality of cultivation.

Weather data

In the end, such remote sensing data combined with weather data could be used to predict the next harvest and thus boost food security.  

Apart from Kenya and Nigeria, plans are underway to expand the platform to Mozambique, Senegal, Tanzania, Pakistan and Bangladesh. 


our partners