Shiraz Hota, Managing Director for Emerging Africa at SAP, asserts that solving post-harvest losses requires more than just physical infrastructure. He examines how artificial intelligence and related data can help overcome the food loss crisis in Africa, bridge the agricultural information gap, and increase farmers' profits.
Every year, a significant portion of crops grown by African farmers does not reach the consumer. Grain losses in Sub-Saharan African countries account for about 20%, while for fruits, vegetables, and tubers, this figure reaches 40–50%. The exact figures vary widely depending on the crop, country, and measurement method, which itself is telling: in many markets, it is unclear where exactly the losses occur and what intervention measures will have the greatest impact.
This post-harvest crisis is often presented as an infrastructure problem. Indeed, Africa needs additional warehouses, refrigerated transport, processing facilities, and reliable power supply, but beneath these physical limitations lies another, less obvious problem—a lack of information.
Under the Kampala Declaration, African leaders set a goal to reduce post-harvest losses by 50% by 2035. Achieving this goal will require continuous investment in storage, energy, transportation, and processing. However, building additional infrastructure without improving the flow of information risks reproducing today's inefficiencies on a larger scale.
Agricultural value chains often suffer from the absence of timely and accurate data on harvested crops, storage locations, delivery urgency to the market, and demand. Without such transparency, enterprises cannot coordinate transport, plan production capacity, or intervene before products spoil. Infrastructure is important, but it cannot realize its full potential if decisions are made based on incomplete or outdated information.
What we call food loss is also a consequence of disconnected decisions. A processor may not know about the early arrival of a harvest. A logistics operator may lack information to allocate refrigerated transport, and a warehouse operator may only detect a temperature breach after the cargo has spoiled.
Farmers who do not have access to market signals may sell their produce immediately after harvest when prices are lowest, as they cannot assess whether storing it would yield better profits.
Artificial intelligence can eliminate these gaps by transforming disparate operational data into coordinated actions. Yield forecasts allow processors and distributors to foresee expected supply in advance. Connected sensors can monitor temperature and humidity during storage and transport. Predictive models help identify spoilage risks, and intelligent planning systems can match products with available warehouse capacity, transport, processing facilities, and buyers.
However, the greatest potential is unlocked by integrating these capabilities across the entire value chain. Temperature alerts have limited value if they do not trigger a logistical response. A yield forecast cannot prevent waste if procurement, labor, and processing plans remain unchanged. Traceability data will not grant access to premium export markets if it is not integrated into quality management, compliance reporting, and financial systems.
Enterprise technologies play a crucial role by linking procurement, inventory, production, quality, logistics, finance, traceability, and intelligent business systems to form a unified operational picture from farm to retailer. AI then helps organizations predict failures, coordinate responses, and continuously adjust plans as conditions change.
These systems can support practical solutions: redirecting vulnerable produce to the nearest processor, prioritizing refrigerated transport, adjusting warehouse staff before delivery, or identifying quality risks before goods reach exporters.
Over time, this transforms agricultural supply chains from being reactive to losses to being preventative.
Elements of this model are already emerging. Solar-powered cold rooms with a pay-per-storage system make cooling more accessible for small producers in Nigeria. Digital supplier platforms help integrate farmers into formal value chains; one initiative implemented with SAP was designed to connect 850,000 maize producers in Nigeria with buyers and services.
The next step is to integrate infrastructure, farm-level information, and enterprise operations into a single connected system. Local conditions and infrastructure must be considered. Many smallholder farmers do not own smartphones or have stable connectivity. Therefore, successful solutions must include offline capabilities, simple mobile interfaces, and trusted intermediaries who can collect information and provide local insights.
Technologies must align with the realities of Africa's agricultural sector, rather than assuming every farmer is constantly connected to the network.
Greater transparency in agricultural value chains benefits not only food security. Traceable and well-managed supply chains can strengthen access to financing, support sustainability reporting, and help agribusiness comply with the quality and regulatory requirements of high-yield export markets. Reducing losses improves margins without the need to increase land, water, or agricultural resources, providing both economic and environmental opportunities.
Africa needs an intelligent, connected value chain capable of early problem detection and coordinating action across organizational boundaries. When physical infrastructure is combined with reliable data and AI-driven operations, post-harvest loss becomes a preventable failure that can be addressed jointly by businesses, governments, and technology partners.


