1. A large technology company engaged a consulting firm to develop a flexible inventory model for about 18,000 SKUs with highly volatile demand for new technologies. 2. The goal was to create an algorithm that simulates demand and assigns optimal buffer inventory levels for each SKU, while ensuring the inventory budget is not exceeded. 3. The algorithm clusters SKUs, estimates demand distributions, and uses nonlinear programming to optimize flexible inventory levels to minimize total costs while meeting service level targets.