Case Study

Using Simulation to Evaluate Cost Effectiveness of Global Health Supply Chain Designs and Policies


WDI aims to improve access to medicines in Nigeria by extending current supply chain analyses to new demand and geography scenarios


LLamasoft Supply Chain Guru simulation


WDI was able to reduce risk and support decision-making by testing supply chain changes in a digital environment

Download PDF

Related Resource


The William Davidson Institute (WDI) was established at the University of Michigan in 1992 and is an independent, non-profit research and educational organization focused on providing private-sector solutions in emerging markets.

Global health supply chains face daily challenges such as:

  • Limited resources and fragmented funding sources
  • Poor data infrastructure, especially in rural clinics
  • Politically-oriented supply chain designs based on government administrative structure
  • Growing global demand for medical supplies and products
  • These challenges make it imperative that networks in developing areas are designed to be cost-effective, reliable and sustainable. 

Furthermore, given the high cost in time, money, and political capital needed to pilot supply chain design changes, the global health community faces a critical need for tools that enable rapid testing and modeling of supply chain designs.


Using LLamasoft Supply Chain Guru discrete event simulation, WDI aims to improve access to life-saving medicines by analyzing supply chain performance across a variety of parameters and design configurations. The result is a data-backed evidence base for supply chain policy decisions and the ability to explicitly capture performance and cost trade-offs between different supply chain designs.   

WDI simulated cost and service levels of four common distribution strategies under different demand, geographic and design scenarios. This can be used to help regions select a strategy that suits their particular geography and available resources.

How was simulation used to address specific supply chain challenges?

Resource Allocation: To improve allocation of resources at each level of the supply chain, through better understanding of supply chain cost drivers.

Pilot new features: To test new design adaptations without physically implementing a pilot, saving money and political capital.

Questioning and stress-testing existing designs: Understanding how specific supply chain designs respond to changes in demand, delivery frequency and delivery capacity. Challenging pre-conceived ideas about which supply chain designs work and which don’t. Moving away from a one-size-fits-all design, and instead helping to adopt a design which is adapted to a country’s particular market or region.

WDI used simulation to capture key design behavior across four models:

  1. Classic vendor-managed inventory (VMI) design, perceived as expensive, but high performing
  2. Direct delivery and information capture: A variant of the VMI design featuring separate trips for ordering and product delivery at the clinic level;  perceived as optimal for high volume systems
  3. Review and resupply: Buyers and sellers converge in one location to review and fill orders; perceived as low-performing and low-cost
  4. Review and direct delivery: Buyers and sellers converge in one location to order. Delivery occurs after 1-2 weeks

Capturing these models in Supply Chain Guru informs real-world design decisions, specifically addressing if the model results align with previously-held beliefs. Identifying the optimal design, simulating varying model factors (e.g. demand, geography, and population density) and testing where each model reaches its cost and performance limits.


WDI was able to reduce the risk and cost of supply chain design improvement by testing design changes in a digital environment before implementing them. Initial analysis highlighted preconceived processes that need to be re-considered:

  • The perception that vendor-managed inventory systems are expensive
  • The belief that scheduled collection models are inherently poor performers; in reality they may simply require more thoughtfully-designed management and incentives

The results also suggested a need to consider hybrid designs that would better match models to market conditions. For example, the “review and resupply” model is cheaper in high-density urban areas, whereas the “review and direct delivery” model does better in low-density rural areas. A hybrid design could combine these two distribution strategies, each in their areas of relative strength.

In addition, by capturing accurate data of the supply chain it was easier for WDI to understand the impacts of the various inputs and outputs. WDI is hard at work continuing to analyze and simulate alternate potential delivery models in order to reduce cost and risk and improve access to medicines in many settings throughout the developing world.

Ready to learn more? Click to request a live demo with the LLamasoft team.

Request A Demo