Abstract

Many low- and middle-income countries, including Zambia, suffer from unreliable distribution of health commodities leading to high variation in service levels across health facilities. Our work investigates how transshipment can improve system-wide service levels, equity across facilities, and average inventory levels. We focus on the distribution of malaria medicines in Zambia’s public pharmaceutical supply chain, which is heavily impacted by the rainy season leading to seasonality and uncertainty in demand and lead times. We use the more advanced deep reinforcement learning method Proximal Policy Optimization to develop transshipment policies and compare their performance with currently available, easy-to-use heuristics. To ensure that the model applies to settings of a realistic scale, we adopt a policy architecture that effectively decouples the policy’s complexity from the problem dimensions. We find that deep reinforcement learning is mainly useful in resource-constrained environments where system-wide inventory is scarce. When sufficient inventory is available, transshipment heuristics are more appealing from an overall cost-effectiveness perspective. Based on our numerical experiments, we formulate policy insights that can support policymakers in a humanitarian health context.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.