Today, there is a need to manage composite patents based on emerging technologies in business. The research related to ranking candidate patents and selecting optimization strategies does not consider the constraints of non-functional properties. This paper propose a method for incorporating the GA and rough set theory that can (i) solve problems that can be decomposed into functional requirements, and (ii) improve GA performance by reducing the range of the initial population and constrained crossover using rough set theory. Based on our experimental result, this approach has shown great promise and operates effectively.