With the complexity of industrial systems and the increasing degree of intelligence, the normal operation of the system is seriously restried by the reliability and safety of the system. Long-term work operation will increase the risk of system failure and reduce its safety and stability. In order to reduce the impact of system failure on product quality and production cost, the optimal maintenance decision problem of the system has gradually become a research hot topic. Analyzing the degraded state of the system facilitates can help make correct maintenance decisions for the system, extending system uptime and reducing the economic losses. For system consisting of multiple components that are identical and degraded independently, a multi-state joint degradation space partition model for multi-component systems under discrete state modeling is established. According to the division of the joint degradation state space, the combination of all maintenance requirements and their probability calculation formula are given, a stationary probability model of the system state is established using Markov process theory. The correctness and effectiveness of the model are verified by numerical experiments.