Assembly Production Systems (ASPS) are used to assemble complex components in manufacturing industries such as automobile production, aviation, mechanical equipment, electronic goods and household items. Importantly, ASPS may, in a single production process, merge two or more serial production lines. Contrastingly, disassembly systems (DisASPS), with split lines, are employed to disassemble/systematically remove constituent parts from the final product, without impairment. The buffer size of the bay in front of each machine tool should be optimized for an efficient ASPS or DisASPS. The characteristics of machine tools, including their uptimes and downtimes, have a significant impact on decisions concerning buffer size and subsequently on the production rate of an ASPS or DisASPS. Optimum buffer size can achieve a high production rate which enhances the performance and operation of the entire production system. This research presents an effective method for determining optimal and near optimal design for assembly and disassembly production systems. The proposed model optimizes the buffer size between each pair of machine tools in the ASPS and DisAPS for any given uptime and downtime parameters of the machine tools. A Genetic Algorithms (GA)-based optimization model is utilized to determine the optimal or near-optimal buffer size for serial production lines in the production systems. The numerical examples performed in the research validated the ability of the optimization model in determining the buffer size. Optimized selection can be carried out quickly as compared to previous selection techniques. Moreover, the proposed algorithm can support the manufacturing engineer’s decision-making during development stage of new ASPS or DisASPS.