When you think about food aid programs in developing countries what images come to mind? Teeming masses of hungry people gathered to receive their giant bags of rice?

Well, it turns out that this is a very inefficient way to distribute aid. Using this distribution model, up to 70% of food aid budgets is spent on logistics and the United Nations World Food Programme is only able to satisfy about 10% of worldwide demand. This model has also been criticized for creating a dependency on aid programs and negatively affecting local markets by flooding the region with free food.

In some situations, the use of cash or food vouchers may yield better results for the beneficiaries, local retailers and food aid agencies. However, vouchers require access to well-stocked markets, and distributing cash presents its own set of problems - so how should an agency go about selecting a distribution model?

That’s the question SFU Beedie Professor Feyza Sahinyazan seeks to answer in her paper Food Aid Modality Selection Problem, published in the Production and Operations Management Journal. In the paper, Sahinyazan and her coauthors provide a mathematical formula for selecting the most efficient food distribution model that accounts for the interests of aid agencies, beneficiaries and non-beneficiaries.

The current decision-making processes are mostly guidelines, and focus on a cost-reduction perspective for the aid agency. The easiest way to reduce costs is to distribute cash - but what about the beneficiaries? The whole point of distributing aid is to increase their level of nutrition, and they are not always familiar with food’s nutritional value. If you distribute cash, beneficiaries may end up maximizing value by buying cheap food with sub-optimal nutritional values - or worse - non-nutritious items like alcohol, tobacco and soft drinks.

And what about the local non-beneficiaries? It is well documented that distributing aid in a remote area can affect the economy of the surrounding region. Whether you’re injecting cash or thousands of kilograms of food, it’s often enough to shift prices for those who are not receiving aid, and most current models don’t account for this. 

“We want to make sure that we incorporate the people who are not receiving food aid as well,” Professor Sahinyazan says. “By including populations beyond the beneficiaries and agencies, we can have a total social welfare perspective when optimizing this distribution problem.” 

This model realizes that there is no one-size-fits-all solution, and serves as a jumping off point for future researchers to work on more efficient ways of distributing aid to beneficiaries in order to lift more people out of hunger.