5 innovations powered by artificial intelligence that tackle world hunger
Here’s how innovations using artificial intelligence can add value to WFP’s humanitarian operations
By Gulia Rakhimova
One of the most promising technologies to emerge in recent years is artificial intelligence (AI) — computer programs that can execute tasks that normally require human intelligence, such as visual perception, speech recognition, and decision making. These AI capabilities are achieved by combining massive volumes of data, referred to as “big data,” with algorithms that extract patterns and insights from the data. AI is already powering smart technology advancements like digital assistants, chatbots, ride-hailing applications, and self-driving cars.
But how can humanitarian organizations create value with artificial intelligence?
The United Nations World Food Programme (WFP) is the world’s largest humanitarian organization tackling global hunger, serving 115.5 million people in more than 80 countries. From blockchain, drones, and robotic technology to innovative financing and new agri-business models, we have been using a breadth of innovative solutions to meet the challenges of an ever-changing world and serve people more effectively around the globe.
WFP explores responsible, sustainable, and inclusive uses of AI as part of its Frontier Innovations Programme, which seeks technological innovations that can enhance humanitarian work. AI can assist humanitarian response by optimizing resource allocation, digitizing data management systems, and using chatbots to improve services to affected people.
Let’s take a closer look at some of the key areas where AI is making an impact on WFP’s operations.
Making optimal choices — Optimus
AI can assist humanitarian operations by helping to better allocate scarce resources. WFP makes hundreds of decisions every day to ensure the right assistance is provided to those in need, with limited resources, and under growing challenges such as climate change, conflict and the COVID-19 pandemic. Optimus is an innovation achieving impact in WFP’s Scale-Up Enablement Programme, which uses advanced mathematics to recommend optimal operational plans whilst minimizing costs.
Optimus combines data from numerous sources — such as population size, transportation routes, and nutritional value of food — to provide vital insights regarding ideal food basket design, food sourcing techniques, and distribution networks. WFP operational teams can create specific scenarios or ask Optimus to find optimal plans, taking into account operational constraints such as lead times and funding. For example, Optimus has been used to calculate how the same budget can serve up to 20 percent more people, stretching donor support even further.
Joining the data points — HungerMap LIVE
In an era of continuously expanding data, ensuring that humanitarian responders on the ground have access to the relevant information is vital. WFP’s HungerMap LIVE is an AI-based hunger monitoring platform funded by the WFP Innovation Accelerator’s Sprint Programme. It tracks and predicts food security in near real-time, combining critical variables from many data sources, including weather, population size, conflict, hazards, nutrition and macroeconomic data. The results are shown on an interactive map that provides global, regional, and country-level insights.
In the wake of the COVID-19 pandemic, HungerMap LIVE has been upgraded to incorporate COVID-19 data into the analysis and expand near-real-time monitoring to new countries in WFP’s operational portfolio. Having access to this data is critical for humanitarian responders to better understand how food security evolves in a complex, ever-changing environment, and enables better informed and timely decision-making.
Reading data from satellite images to speed up emergency response — SKAI
Another promising application of AI for humanitarian operations relates to emergency response. Following disasters, we need to assess the extent of damage and the immediate needs on the ground. Satellite imaging is becoming an increasingly powerful tool in this process, providing unprecedented visual information on how disasters affect the built environment, infrastructure, and communities. However, getting a human team to assess the locations on foot or to look through images and find operational information takes several weeks. SKAI is an innovative project under the WFP Sprint Programme that aims to deploy an AI machine learning model to analyse satellite images and quickly assess building damage to inform humanitarian response within 24 hours.
Machine learning is a type of artificial intelligence that draws on data and patterns to develop new insights. SKAI’s machine learning algorithm uses computer vision to extract and compare satellite image data captured before and after disasters to determine damage. Well-trained machine learning models can detect damage in entire cities in minutes, allowing for faster and more efficient humanitarian response.
Keeping lines of communication open — Voice to Text AI
AI can also assist in improving communication with and accountability to the people we serve. The COVID-19 pandemic has hampered face-to-face surveys, which have been a typical technique to collect nutrition information from individuals living in remote locations in Ethiopia. This primary data is critical in implementing nutrition programmes such as WFP’s Fresh Food Voucher Programme that addresses stunting — impaired growth and development that children experience from poor nutrition — through diversifying diets for children and lactating women.
The Voice to Text AI, a WFP Sprint Programme innovation, allows people to complete nutrition surveys remotely from their phones. It automatically dials up and surveys respondents using Interactive Voice Response technology and speech recognition technology. The pilot aims to fine tune an open-source voice recognition machine learning model that may be used by people who speak languages that are underrepresented in current AI tools, such as Amharic and Somali. This AI solution is more efficient and less expensive than face-to-face and telephone surveying, allowing savings to be reinvested in nutrition programmes to support more people.
Turning paper records into digital data — MEZA
AI-enabled technologies have the potential to accelerate the digitization of data management systems, and open up new avenues for improving the services provided to those in need. For example, one of the challenges that humanitarian workers face is gathering data to inform malnutrition programming. It can take considerable time to get and review the health records of malnourished children because they are kept in paper-based booklets located in remote, under-resourced health clinics.
MEZA is a WFP Sprint Programme project that leverages AI to solve this issue. It uses Optical Character Recognition — an AI approach that reads text inside photographs and extracts data from photos of logbooks submitted by health clinic workers. WFP and governments can use MEZA to quickly and safely digitize handwritten data from paper-based clinical records to inform nutrition programmes, ensuring that children receive nutrition support that is tailored to their specific needs.
Emerging frontiers of humanitarian action
Innovations in artificial intelligence are contributing to the digital transformation of the humanitarian sector. AI-enabled technologies can help us analyze massive amounts of data quickly, predict emergencies, inform optimal decisions, and act faster with real-time awareness of the situation. The WFP Innovation Accelerator provides hands-on support and space for experimentation for innovators to successfully create, launch, and scale such transformational solutions, while keeping the people we serve at the centre of these efforts.
The WFP Innovation Accelerator sources, supports and scales high-potential solutions to end hunger worldwide. We provide WFP staff, entrepreneurs, start-ups, companies and non-governmental organizations with access to funding, mentorship, hands-on support and WFP operations.
The SKAI isn’t the limit: How WFP uses satellite imagery and machine learning in emergencies
The WFP Innovation Accelerator partnered with Google Research to set up the SKAI platform to automatically assess…
The benefits of machine learning in emergencies from river DEEP to mountain SKAI
Why automating analysis of aerial images is taking off for humanitarian response