Supply Chain Frontiers issue #37
Cross-docking, a method routinely used by retailers to improve the flow of goods, could save countless lives in developing countries by streamlining the supply of lifesaving drugs to health care clinics.
The method is being tested in a number of districts in Zambia by a team from the World Bank, the MIT-Zaragoza International Logistics Program, and project partners, as part of groundbreaking research into pharmaceutical supply chains in Africa. If successful, the cross-docks could be rolled out nationally and save the lives of an estimated 27,000 children over the next five years by increasing the availability of malaria medicines. Even more lives could be saved if the supply chain for other essential medicines is improved. Moreover, the lessons learned from the research could fundamentally change the way medicines are distributed in other countries.
The Zambia Access to Artemisinin-Based Combination Therapy (ACT) Initiative is funded by the World Bank, USAID, and the United Kingdom’s Department for International Development. The project team is led by Dr. Monique Vledder, senior health specialist, and Jed Friedman, senior economist, at the World Bank research group responsible for the evaluation. Dr. Prashant Yadav, professor of supply chain management at the MIT-Zaragoza International Logistics Program, provided technical support to the team.
The project team studied the way malaria medications are supplied to public health care clinics in 16 districts in Zambia. Inefficiencies in these supply chains can have life-or-death consequences. Approximately 77,000 children five years of age and younger die in Zambia every year, and an estimated 20% of these deaths are caused by malaria. Access to effective treatment is a major challenge; only about 7% of Zambian children in rural areas receive malaria medicines within 24 hours of developing fever.
The research is unique in the way it compared and rigorously evaluated different supply chain configurations, according to Vledder. “This is the first time in the Africa region that it has happened,” she says.
The research findings produced some surprises. “Analytical models show that for products with very high demand variability, putting stock in local districts is the best option,” Yadav says. Locating inventory close to the point of service enables suppliers to respond quickly to order changes. “Where the demand for a product is more or less stable, you can have a cross-dock and deliver direct from there,” he continued. Even though the demand for malaria medicines is unstable, in theory stockpiling drugs locally—rather than delivering the medicines from more distant cross-docks—is the best distribution option. “But what we actually observed was that, irrespective of whether the product has a highly variable demand, the cross-dock option always worked best,” says Yadav.
The reason for this counter-intuitive result can be traced to staff behavior at the clinics. If, say, there is a policy to maintain at least two months worth of stock, having inventory readily available in the locale is a distinct advantage because stocks can be replenished quickly when levels fall below the two-month threshold. In reality, however, managers at the clinics tend not to adhere closely to replenishment policies. “In fact, they can become complacent knowing that inventory is not far away,” says Yadav. In addition, holding local inventory at each district requires smooth stock issuing and order picking at each district, which may not always be the case. As a result, the advantages of positioning inventory locally can be largely negated by lax ordering practices and the inefficiencies of decentralized order picking.
As Vledder explains, in the most successful supply chain configuration being tested, drug shipments are packaged and sealed by specialists at the national level. The districts function as cross-docks for final delivery of these shipments to health care clinics. In this supply chain, the district cross-docks are only responsible for transportation. Under the traditional system, district managers and their staff are also responsible for assembling the shipments. “To have specialized people doing the assembly and packaging specifically for the clinics has really made a difference,” she says.
The results are indeed impressive. In the trial districts where the supply chain improvements were introduced, pediatric malaria drugs are now available 345 out of 365 days, with an average downtime of only 20 days a year. The availability in control districts was just 247 days; the availability of other drugs also increased in the pilot areas. Amoxicillin, a lifesaving antibiotic that cures lower respiratory infections and other infections caused by HIV/AIDS, was available 92% of the time in the districts with the enhanced supply chain, compared to 63% of the time in other districts.
The pilot project covers eight control districts, eight with local inventory, and eight served by the cross-dock distribution model, and this configuration will operate for at least one year. If deemed successful, the cross-dock option will be rolled out nationally.
There is a power-sharing issue that could hamper such a rollout, however. “If you own inventory, you own power,” says Yadav, because managers can allocate drugs according to their own priorities. The cross-dock configuration effectively takes inventory away from these individuals, and there are concerns that this could deter them from participating. She says the key question is: “Will the people at the district level accept this as a good technical intervention and continue to achieve the efficiencies we have seen in the pilot, or will they see their role being marginalized and become less diligent about transmitting order information?”
If these obstacles can be overcome, cross-dock distribution methods could be adopted in other countries. “Most developing countries currently hold inventory at the district level except for HIV medicines,” Yadav explains. “Why not apply something like a cross-dock—which is a cornerstone of supply chain management in retailing and other industries—in the pharmaceutical supply chain in developing countries?”
The project in Zambia could provide the evidence needed to persuade other countries and government agencies to make the switch. “A large-scale, controlled experiment like this makes the case somewhat easier,” says Yadav.
Vledder agrees that the research project has generated a great deal of interest in other countries. “Each country has its own logistics system so it might not be possible to transfer the intervention exactly, but the concept is very applicable across Africa,” she says.
The project has also demonstrated how private-sector supply chain methods can achieve remarkable results in the public sector. “In my view, that is really exciting,” says Vledder. “We calculated that this impacts 40% of all the child deaths in Zambia if we do it nationwide. That kind of impact is really dramatic.”
The team is now looking at what other best practices can be included in the design that will be scaled up nationally. Dr. Jérémie Gallien from the MIT Sloan School of Management is working with the team to investigate better inventory control policies and the optimal design of “pre-packs,” which are commonly used in many apparel retailing environments.
For more information on the Zambia pilot project, contact Dr. Prashant Yadav or Dr. Monique Vledder at mvledder@worldbank.org .