Last Mile Logistics


The Last Mile Logistics work stream projected a savings of 10% on the current direct delivery budget through savings on fuel and per diem costs associated with the optimized routes. These savings have the potential to be used for purchasing additional medicines. In some pilots, MSD reported improvements of up to 20% in medicine availability. In addition to the creation of the excel-based tool, the work resulted in the development of a standard operating procedure (SOP) and toolkits for rolling out direct delivery in other parts of the country. This SOP will allow area managers to focus on implementing the change, instead of recreating planning processes. The SOP and related toolkits can also lead to easier cost comparisons across zones, promoting learning for further cost savings.

Although the Excel templates provided a low cost and low-tech solution to route optimization, the enhanced technical capabilities of Llamasoft allowed the process to be automated. Data can be easily fed into the system to produce an optimized route, allowing the process to be repeated for additional routes. MSD and ADP are currently pilot testing to ensure that these software developed routes correspond accurately to actual driving distances. Two MSD employees have been trained as “Network Model Managers” to continue the optimization process.

The Last Mile team also evaluated several outsourcing models for generating additional cost savings allowing MSD to focus on serving the hardest to reach facilities.Coca-Cola Kwanza has shared their methodology and contractual templates of how they recruit, select and engage third-party partners within the supply chain. However, as a public sector entity, MSD faced many regulatory hurdles to contract third party distributors. Enusring timely feedback and a “proof of deliveries” also raised concerns regarding outsourcing last mile operations. Yet, amidst these hurdles, MSD has  secured a third party vendor and have received approval to outsource some of their distribution beginning Summer 2014. 

Key Metrics for Last Mile work stream:

  • Distribution Cost/Cost to serve
  • On Time In Full Orders to Health Facilities
  • Service Level Agreement achievement by distribution partners
  • Fleet Capacity Utilisation
  • Customer Satisfaction / Service Level

Moving Forward: 

Incorrect GIS coordinates posed a serious challenge to Llamosft supported route optimization. Pilot tests revealed that routes, distances and costs predicted by the software were quite different to the actual results. MSD has also been collaborating with John Snow Inc to collect GPS codes of the health facilities to generate accurately optimized routes through Llamasoft. MSD has also introduced a fleet management car tracking system, making the drivers more accountable to the delivery and location of their trucks.

However, optimization is not a one-time exercise but requires regular continuous optimization as vehicles become available, the number and location of functioning health facilities changes, and road conditions evolve. Drivers need additional training in logistics and incentives to continue the successful operations. Trucks also require constant maintenance, adding to the overall costs of the direct delivery process. Funding continues to pose serious challenges with MSD simply not having enough vehicles or financial resources to support direct delivery to all health facilities.

Moreover, the goal of direct delivery is to decrease lead times and stock outs at the facility level, but this level of analysis is not possible with the current data infrastructure in the country. Additionally, barriers in the ordering process are likely to limit the impact of the improvements on stock levels in the health facilities. For stock out rates to improve, facility orders must reflect an accurate forecast of need, be received on time, be filled completely based on both MSD stock and facility budget, and, finally, be efficiently delivered. Most of these factors will require additional resources and trainings at both the central and zonal levels.


 Background Resources

Tanzania: 2020 Supply Chain Modeling: Forecasting Demand from 2020–2024

Click below to view the ADP presentation on accomplishments and challenges throughout Phase II.

 Partners' Perspectives 

“I shared with them how we go through the process of selecting our partners, how we vet them, how we support and invest, how do we do our assessments, how do we evaluate them and how do we sustain the continuity of them trading with usCoca Cola Kwanza Official