Abstract: This paper focuses on the dynamic sorting and scheduling optimization problem in automated sorting systems for customized furniture panels, which is affected by the randomness of production links and leads to the uncertainty of panel arrival time. First, a mixed integer programming model is formulated based on the analysis of problem characteristics. Second, due to the difficulty in quickly solving the dynamic scheduling problem with solvers and the challenge of single heuristic algorithms adapting to dynamic environments, an adaptive sorting and scheduling algorithm based on Q-learning is presented by designing an action set, state space and a reward function. Finally, comparison experiments are conducted with designed test cases. It shows that the algorithm proposed in this paper achieves good performance in the optimization of order delivery efficiency and buffer congestion rate, providing decision support for sorting and scheduling personnel in planning sorting strategies.