A public health challenge has led to a more efficient way to allocate all sorts of resources
Optimization models do an excellent job of ensuring products get where they are needed in the most efficient way possible, but those models require an impractical amount of computing power when dealing with large-scale problems—such as distributing vaccines to millions of people. Researchers have now demonstrated a way to drastically reduce the computing power needed to run these optimization…
Researchers have created an optimized method to distribute vaccines that cuts down on computing power while still achieving near-optimal results. Traditionally, optimization models for resource allocation like vaccine distribution use large amounts of computing power to handle millions of variables. Leila Hajibabai, an associate professor, explains that their team developed an optimization model to efficiently distribute vaccines, but applying it on a statewide level proved impractical due to the high computational requirements.
To address this, they combined machine learning with column generation, a technique that breaks down large problems into smaller, more manageable ones. Machine learning then predicts promising decisions based on previously solved instances, allowing the algorithm to focus on the most likely to improve the solution, thus reducing the computational effort needed.
As a result, the machine learning-guided column generation (ML-CG) approach decreased runtime by 79.1% compared to traditional column generation, while still providing solutions within 6% of the optimal outcome. This method's efficiency and effectiveness suggest its potential utility in various resource allocation scenarios, such as disaster relief, humanitarian logistics, transportation planning, and energy system design.
Written by urgent.news from Medical Xpress's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
