This article describes the use of genetic algorithms in real-time systems that employ the imprecise computation paradigm. In real-time systems, the focus is on ensuring that a set of tasks each complete within their deadlines. Faults may occur in the computation or the environment that can cause missed deadlines. That is why the idea of using partial results when exact ones cannot be produced within the deadline has been introduced. This idea has been formalized using the concepts of anytime algorithms and imprecise computation and specific techniques have been developed for designing programs which can produce partial results and for developing systems that can support imprecise computation techniques. Genetic algorithms are methods that can be, without any adaptation, used in an imprecise computation system. They produce a solution that bears a certain measure of reliability. During the process of their execution, this solution is constantly improving. They can be used as a part of a real-time system, especially for optimizing tasks where the classical algorithms are not applicable or its computational time proves to be too expensive.