Robotic assembly line balancing (RALB) is specific for robotic assembly, where different robots require different assembly times to perform the same task because of their specialization. The problem includes optimal assignment of robots to line stations and balanced distribution of work between different stations. It aims at maximizing the production rate of the line. This problem is solved by a genetic algorithm (GA). Different procedures for adapting the GA to the RALB problem are introduced, including a local optimization (hill climbing) work-piece exchange procedure. The best combination of these procedures is reached by testing on a set of randomly generated problems.