Research
#7 Research meet-up: Daily route optimisation cut unnecessary medical sample trips by 55% in Malawi
A new system for transporting medical samples in Malawi reduced unnecessary courier trips by 55% and shortened average transportation delays by an estimated 25%. It combines low-cost data sharing through feature phones with daily route optimisation for couriers.
A new system for transporting medical samples in Malawi reduced unnecessary courier trips by 55% and shortened average transportation delays by an estimated 25%. It combines low-cost data sharing through feature phones with daily route optimisation for couriers.
Why does sample transportation matter for diagnosis?
Centralised diagnostic networks rely on samples such as blood and sputum being transported from healthcare facilities to laboratories.
When transportation is inefficient, samples can take longer to reach laboratories and diagnostic results can take longer to return to patients. These delays can affect how quickly clinical decisions are made.
The project focused on a national sample-transportation system operated by Riders For Health in Malawi.
Why is this a prescriptive analytics problem?
Courier routes need to respond to the number of samples waiting at different healthcare facilities.
The project therefore combines two components. A low-cost feature-phone platform allows facilities to share information about current sample volumes. An optimisation model then uses that information to generate efficient routes for sample-transport couriers each day.
Rather than relying on fixed routes regardless of demand, the system can adjust transport decisions using current information.
What changed after implementation?
The system has operated continuously in three districts in Malawi since August 2019.
Since implementation:
unnecessary courier trips have fallen by 55%;
average sample-transportation delays have been shortened by an estimated 25%.
The practical significance is that fewer courier journeys are made when they are not needed, while samples can move through the diagnostic network more quickly.
Who developed the research?
The webinar was presented by Emma Gibson, a PhD candidate at the MIT Operations Research Centre. Her research covers optimisation, logistics and machine learning, with a focus on applying mathematical models to healthcare delivery in settings with constrained resources.
Before joining MIT, she completed a BSc (Hons) in Mathematical Sciences at the University of the Witwatersrand and an MSc in Logistics at Stellenbosch University.
More updates
Programs
The Impact-First ABW Flagship Program Begins
Some of the most important work directly impacting underserved communities is being done by non-profit organisations that have rich on-the-ground knowledge, but who are also in the early stages of their data maturity journeys. In 2026, Analytics for a Better World took a step toward accelerating the impact non-profit organisations can have.
Read more
Programs
Accelerating nonprofit impact with AI
What happens when the world’s best data and AI experts apply their skills and share their tools with nonprofits to solve pressing global challenges? They accelerate progress. This is the promise of Analytics for a Better World (ABW), the nonprofit institute founded three years ago by ORTEC and the University of Amsterdam Business School. Its mission: to bridge the gap between advanced AI applications and the nonprofit sector.
Read more
News
A founder's farewell with Robert Monné
Analytics for a Better World (ABW) co-founder Robert Monné is stepping back from his role, marking the end of the organization's founding chapter and the beginning of an exciting new one. Robert was essential to the successful ecosystem that ABW is today, contributing to the transformation of a bold idea into a thriving global community of 40+ organisations united under the purpose of using analytics and AI equitably for good.
Read more