The Seru Production System (SPS) is a Japanese cellular production system, that originated with the need for improvements in assembly processes of the electronic industry. Distinct from conventional production systems, SPS has lots of autonomous production units called seru. In these serus, there is the opportunity for producing different kinds of products, which helps to increase the flexibility of the production system. The workers who are working in these seru units, are cross-trained, so they are capable to complete the manufacturing or assembly process by themselves. In this study, we present a seru loading model in multi-skilled worker environment. The objective of the model is to minimize cycle time, while decisions are assignments of workers and product types to different serus with workers having different skill levels for different tasks and including product changeover setup times. The total number of seru, the number of workers and the number of product types and their demand are predetermined. The proposed math model has been implemented in open-source solver and results were reported in the paper.