When you talk about "creating an optimization algorithm," do you mean that you want to invent a novel optimization algorithm that has never been seen, heard, or published before? If you don't know where and how to start, then I suggest you look up the design frameworks of popular nature- or metaphor-inspired metaheuristics because it is a very active area of research.
However, it is crucial to understand that asserting the novelty of your proposed optimization algorithm does not inherently guarantee its actual novelty. Numerous algorithms that are claimed to be "novel" often turn out to be mere repackaging of pre-existing ideas, with existing concepts being renamed. A notable instance of this debate revolves around the Intelligent Water Drops algorithm. If you are interested, you can find further information on this topic here:
In general, it is important to ensure that your work aligns with the criteria established in the Policies on Heuristic Search by the Journal of Heuristics, as these criteria play a significant role in contributing to the advancement of the field. 
