The technology of multi-stage fracturing in horizontal wells is an effective means for the economic development of shale gas. The method of multi-cluster, multi-perforation fracturing significantly improves the extraction efficiency between clusters; however, the shorter fractures generated by the multi-perforation mode do not contribute ideally to cluster development. Additionally, this fracturing mode limits well spacing and increases drilling and fracturing costs. This study, based on logging data from a specific gas well, established a semi-mechanistic model. Using UFM to simulate hydraulic fracturing, the study examined the impact of natural fracture parameters, comprehensive permeability, and injection procedures on the parameters related to the artificial fracture network. The study also conducted an optimization design for the multi-cluster fracturing scheme considering all influencing factors. The relationships between permeability in southern Sichuan, natural fracture spacing, hydraulic fracture length, and gas well production were established. Furthermore, the optimization results for cluster stages indicated that as the number of clusters decreases, production initially increases and then decreases.