Developing new types of steel is an ongoing challenge in the metallurgical sector, as researchers are constantly seeking to create lighter and stronger steel vari- ants adapted to the various industrial demands. This study uses a diversity archive, with an update rule according to Shannon entropy to increase the genotypic diversity of steels and increase the possibility of choices for decision-makers. The problem is multi-objective and requires the optimization of three mechanical properties, so a multi-objective evolutionary algorithm called RVEA was used. Two configurations were tested: C1, where only one solution archive is used to store the steels, and C2, where two archives are used, solutions and diversity, and solutions from the diversity archive are reintroduced into the population. For each configuration, the diversity of the population was evaluated using entropy, checking whether there was an increase in the distribution of the input variable ranges. In addition, tests were carried out to find the ideal value for the parameter for reintroducing the solutions from the diversity archive. The results show that using the diversity archive governed by Shannon’s entropy effectively increases genotypic diversity, leading to more distinct configurations.