A method of optimizing the cost and reliability of a system represented by a fault tree is proposed which is useful in choosing the technology options for basic events that are available to the system designer. The problem of optimization is to minimize the overall system cost by exercising technology choices that will yield a prescribed probability of failure of the top event of the system. The well known non-dominated sorting genetic algorithm (NSGA-II) has been used for optimization. Both single and two objective optimization problems have been explored and their numerical convergence features discussed. The application problem treated is one of leakage in a subsea production system. © 2018 Elsevier Ltd