Health supply chain managers face limited resources and increasing responsibility to ensure the right medicines are in the right place at the right time for patients. Mathematical optimization allows them to make the best decisions possible with limited resources.
Across industries, companies work tirelessly to optimize their supply chains to get customers what they want, when they want it, and spend as little money as possible in the process. From agribusiness to e-commerce and information technology, companies strive for better data, stronger foresight, and breakthrough innovations that allow them to deliver high-quality goods faster and more efficiently while making a profit.
But, optimization takes on new meaning in global health supply chains when the end users are patients in need of affordable and reliable antiretroviral treatments, bed nets to protect against malaria, or other health commodities. Stakeholders involved in the public health supply chain — procurement agents, ministers of health, warehouse managers, clinic workers — must make decisions continually, from warehousing and storage to distribution, delivery routes, and beyond. And there are costs at every turn. Supply chain logistics costs alone account for five to 50 percent of a final health product’s price. Reduced costs throughout the health supply chain could mean a country could buy more health products for its communities, strengthen its health systems, or reduce donor dependency and achieve sustainability of its operations. But with such complex operations, how do supply chain managers make decisions and determine where to achieve supply chain efficiencies and optimize their operations without sacrificing quality? That’s where the math comes in.
Applying Mathematical Optimization in Supply Chain Management
Mathematical optimization is a branch of applied mathematics. It is used to select the best solution from a set of viable options to maximize or minimize a given mathematical function. For example, a football player may want to find the best play from a set of options to maximize his running yards but keep his likelihood of fumbling the ball low. His ultimate goal is scoring as many touchdowns as possible. The options he is choosing from correspond to the decision he must make on which play to use; the viability of those options is the constraints he’s operating within, such as number of players he can use for a given play; and the mathematical function corresponds to how he evaluates his various options to make the decision. This concept is not new. Mathematical optimization has its roots in World War II when the need for military operation planning and decision-making gave birth to a brand-new discipline called operations research. With the explosion of data and computing power, optimization is experiencing a renaissance.
Mathematical optimization is part of prescriptive analytics, which is referred to as the final frontier of analytics capability. More commonly encountered today, descriptive analytics looks at historical data to gain a better understanding of performance; predictive analytics focuses on the likely outcomes and their chances in the future; and prescriptive analytics takes advantage of the results of descriptive and predictive analytics to find the best course of action to achieve a desired goal. For example, when you’re driving a car, you look at the instrument dashboard and rearview mirror to get descriptive analytics results on where you were and how the car is running; predictive analytics allows you to look through the windshield to see where your car is heading if no change is made; and prescriptive analytics is the steering wheel and brake pedals that allow you the ability to determine the best way to get where you want to go.
In public health supply chains, mathematical optimization and prescriptive analytics are particularly useful to allow us to make the best use of the resources available to ensure health commodities reach those who need them most. At the strategic level, mathematical optimization can be used to redesign our supply chain network, set up new distribution hubs, or change the service areas aligned with the hubs. At the operational level, it can be used to better manage the replenishment of the warehouse and supply planning. And at the tactical level, optimization can be used to create dynamic delivery routes based on available trucks and orders submitted.
Application on the USAID Global Health Supply Chain Program – Procurement and Supply Management (GHSC-PSM) Project
In the last three years, the GHSC-PSM project has utilized mathematical optimization in several areas:
1. Regional distribution network optimization. In 2016, the project conducted an optimization analysis of our network of regional distribution centers worldwide and proposed a new structure — a consolidation of five distribution centers to three — with projected savings of more than $6 million per year. The project is currently accumulating savings at a faster pace compared to those original estimates.
2. In-country optimization. In Lesotho, GHSC-PSM has applied network and route optimization to assess the network structural changes to reduce the impact of increasing the delivery frequency to twice a month. In Ghana and Ethiopia, GHSC-PSM proposed a new optimized warehouse hub design for both countries leading to cost savings and better service levels. And in Côte d’Ivoire, GHSC-PSM designed and developed a dynamic routing tool prototype that would allow the central medical store to create dynamic routes based on the order received to achieve better delivery performance and truck utilization.
3. Annual procurement for shipping lanes. One key way GHSC-PSM is optimizing the supply chain is bidding out as many aspects of our supply chain as possible, including shipping lanes, ensuring that we can deliver more health supplies for the same dollar. On an annual basis, GHSC-PSM conducts a transparent procurement process to select third-party logistics providers and uses a mathematical equation to determine best value, including a combination of technical capability, price, and transit times, rather than basing awards on price alone. Because of the large volume of data and need for analysis, GHSC-PSM uses simulation techniques and modeling to understand how demand variability — volume and transit times — could impact the selection.
Public health supply chains face limited resources and ever-increasing responsibility to ensure the right medicines are in the right place at right time for the patients who need them. Applying mathematical optimization and prescriptive analytics allows health supply chain stakeholders the ability to make the best decisions possible with the limited resources at hand. What could be cooler than that?
To learn more about GHSC-PSM’s optimization efforts, visit our sessions at the upcoming Global Health Supply Chain Summit in Zambia or visit https://www.ghsupplychain.org.
Posts on the Chemonics blog represent the views of the authors and do not necessarily represent the views of Chemonics.