LEAST COST FULFILLMENT OPTIMIZER

THE CHALLENGE

E-commerce companies are striving to their head above water when it comes to profit. Our client was having to drop pricing due to competitive pressures, thus having a negative impact on margins. The client wanted to explore ways to reduce cost, thereby increasing profitability.Was there way to bring the overall cost down to compete with other larger competitors on price?

OUR APPROACH

Our client sources merchandise from various suppliers across the United States, each having a significant number of warehouses, as well as a range of price levels for their product lines. The Tatras Data scientists suggested applying the principle of Least Cost Management (LCM), factoring in the supplier product offering, product pricing, rebate incentives, delivery costs, delivery service levels and delivery location.

THE METHODOLOGY

• The orders were pulled in real-time from the order management system.
• Items within the order were matched with data from supplier web-services using factors such as stocks availability, price and incentives offered, warehouse locations vs. order location, cost of shipping, time taken for delivery, etc.
• An exhaustive evaluation of all possible ways to fulfill the order is computationally not feasible. Hence Tatras developed an intelligent algorithm that started from a random population of potential solutions and iteratively improved those solutions based on the principles of evolutionary computing.
• The strategy produced by the algorithm splits orders into an optimal number of shipments, each shipment being allocated to a different supplier warehouse. The total cost of fulfillment being minimized and service levels optimized.

TECHNIQUES, TECHNOLOGIES, TOOLS

Evolutionary Programming and Java

RESULT/IMPACT

Developed a platform in 3 months for least cost order fulfilment. It was integrated with suppliers and logistics partners that minimizes the time and cost of fulfilling an order hence maximizing profit by $3 to each order.