@RISK Monte Carlo Simulation Balance Inventory with Overseas Shipments by Air and Sea to Optimize Global Supply Chain
Published: Nov. 17, 2021
Updated: Oct. 12, 2022
The long global supply chain for technology companies in the United States combined with the need for timely delivery of products to customers has brought challenges to planning. The global pandemic added even more complexity to meet demand for timely delivery.
Dell Technologies adopted a hybrid replenishment strategy using both ocean and air shipping to ensure inventory and service levels while reducing logistics costs. The global supply chain strategy team evaluated the static initial inventory policy to identify an optimal solution that would minimize total fulfillment cost.
Francisco Erize, a supply chain digital automation product owner at Dell Technologies, has more than 15 years of experience at Dell in operations research and business improvement. Erize is focused on global supply chain technical solutions to improve demand forecasting and strategic supply planning.
“Safety stock” inventory provides a buffer that enables the company to achieve a desired service level despite uncertain demand. Weeks before the final replenishment decision is made to match the target inventory, replenishment planners have the option to place long lead time orders.
The lower cost, slow replenishment mode can be ocean or rail depending on the geographic region. For example, for North America this long lead time is the seven weeks from China. The objective of placing these early inventory replenishment orders is to save freight costs by avoiding the need to replenish by air, however this decision carries the risk of increasing the inventory level beyond the required target.
“The hybrid replenishment strategy does not currently have a closed solution in the operation research field to identify optimal safety stock targets while minimizing total fulfillment costs,” Erize said. “While we are working with an academic research team to achieve a closed solution, we have created an empirical solution based on Monte Carlo simulation models for inventory replenishment using @RISK.”
Erize said conventional, inventory models would not solve his needs, and he rejected the option to purchase proprietary, pre-packaged inventory simulations. “I wanted to build a simulation that I could test and be confident in the results,” he said.
He developed and tested optimal replenishment models using actual historical results before framing a policy that achieved less than 10 percent deviation from actual results. “@RISK is so easy to develop models,” he said. “By using @RISK, I was able to pressure test and insert boundary conditions to identify mistakes in the model. This is the best way to find practical solutions. The improvements have been dramatic with the amount of money we saved.”