Hybrid Cuckoo Search for Accelerating Modeling and Design of High-Entropy Alloys

High-entropy alloys in competing crystal structures have a huge design spaces for unique chemical and mechanical properties. To enable computational design, we use a metaheuristic Hybrid Cuckoo Search (CS) to construct models on-the-fly having targeted atomic site and pair probabilities on arbitrary crystal lattices, configurations represented by Super-Cell Random APproximates (SCRAPs). Our Hybrid CS permits ultrafast global solutions for large, discrete combinatorial optimization that scale linearly in system size and in number of parallel processors (strong scaling). For example, a small system given by 4-element, 128-site SCRAP is found in seconds -- a reduction of 13,000+ over current strategies. With model-generation eliminated as a bottleneck, computational alloy design can be performed that was formerly impossible or impractical. We showcase the method for real alloys with varying atomic short-range order.