Purpose The study aims to maximize the efficiency of the process under a given current condition by changing the geometry of the coil. This optimization is economically justified by reducing the cost of the process. Design/methodology/approach The paper presents the author’s optimization process for a case requiring long computational time. The presented optimization is based on a 3D simulation model of an electromagnetic levitation melting (ELM) inductor. Findings The result of the work is to find a suboptimal inductor geometry for ELM. Research limitations/implications To solve the presented problem, a procedure using an evolutionary algorithm was relied on. As for all global search algorithms, it is possible to find a local optimum instead of a global one. Practical implications The new inductor geometry for ELM, thanks to its higher process efficiency for its class of inductors, can lead to the reduction of the costs of the process by using this type of equipment. Originality/value The novelty of the article is a proprietary optimization algorithm and the use of an advanced 3D simulation model which was necessary due to the lack of symmetry of the ELM inductor.