We study a hybrid strategy that uses both process flexibility and finished goods inventory for supply chain risk mitigation. The interplay between process flexibility and inventory is modeled as a two-stage robust optimization problem. In the first stage, the firm allocates inventory before disruption happens; in the second stage, after a disruption happens, the firm determines production quantities at each plant to minimize demand loss. Our robust optimization model can be solved efficiently using constraint generation, and under some stylized assumptions, can be solved in closed form. For a canonical family of flexibility designs known as the K-chains, we provide an analytical expression for the optimal inventory solution, which allows us to study the effectiveness of different degrees of flexibilities. Moreover, we find that firms should allocate more inventory to high variability products when its level of flexibility is low, but as flexibility increases, the inventory allocation pattern “flips” and firms should allocate more inventory to low variability products. These observations are further verified through a numerical case study of an automobile supply chain. Finally, we extend our robust optimization model to the time-to-survive metric, a metric that computes the longest time a supply chain can guarantee a predetermined service level under disruption.