Structural fuses have been used to bias and control failures in structural applications where predictability of the progressive failure or collapse response is important. Tailoring structural fuses by trial and error in large structures that have numerous possible load and failure paths is not possible because the optimum failure sequence is not known a priori. Using nondeterministic methods to tailor structural fuses is computationally expensive. A procedure for developing deterministic measures to optimize structural fuses is presented here. The progressive failure of composite laminates is used for demonstration. Structural fuses are optimized using a reliability optimization. The failure response characteristics of the laminate with optimum structural fuses are used to identify deterministic measures that correlate with high progressive failure predictability. The deterministic measures are validated by using them as surrogate design criteria in a deterministic optimization to optimize structural fuses that control failure and improve progressive failure predictability. The improvement in predictability of the deterministic optimum design achieved by using optimized structural fuses is better than that obtained by optimizing the ply angles of the laminate explicitly for predictability.