Most of today’s supply chain applications were developed in the 1990s. By necessity, given the complexity of supply chain optimization and the scarcity and expense of computing power at the time, traditional applications have limited analytical capabilities and only solve for individual parts of the overall problem. While this optimizes the specific operation being studied, the overall solution is suboptimal. Yet even with this piecemeal approach, companies saw significant benefits when these applications were first implemented. Today, however, companies struggle to squeeze additional value out of their supply chains.
LogicBlox-based applications are once again driving substantial value from optimizing supply chains. Our model-driven solutions, the use of parallel computing techniques, and the elastic scalability of the cloud enable us to model entire supply chains at a granular level and to solve very large problems efficiently and economically.
For one company, the results include a 10-30% reduction in inventory while meeting stringent fill rates. In this case, a LogicBlox-based replenishment system computes nightly store orders, vendor orders, and inventory in-motion, at distribution centers and at stores, by solving a massive problem to maximize profit. The number of nodes in the model is nearly 2 billion, and the number of variables is nearly 800 million. The LogicBlox-based approach simultaneously trades off complex relationships that are often in conflict, such as service level, sales, inventory productivity, transportation, and many other factors.
For another company, LogicBlox-based optimization models result in an 8% reduction in transportation costs. This second company has entrusted LogicBlox to re-engineer their entire supply chain. Using over 150 million transportation records and purchase order data, this LogicBlox-based solution makes adjustments to the assignment of stores to distribution centers and refinements to inbound and outbound flows by adding consolidation centers and pooling points and deleting others.