Abstract

We introduce a moving mesh algorithm for simulation optimization across a continuous domain. During each iteration, the mesh movement is determined by allocating simulation runs to partitions of the domain that are critical in the process of identifying good solutions. The partition boundaries are then adjusted to equally distribute the allocation runs between each partition. To test the moving mesh algorithm, we implemented it using the OCBA method to allocate simulation runs to each partition. But, the simplicity of the procedure should provide flexibility for it to be used with other simulation optimization techniques. Results are presented for several numerical experiments and suggest that the technique has potential given further development.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call