Volunteer Dispatch Method Based on the Q-learning Algorithm and Two-Sided Matching

With the rapid development of modern technology and the frequent occurrence of public crises, efficient volunteer dispatch has become a key issue in emergency rescue. This paper proposes a volunteer dispatch method based on the Q-learning algorithm and two-sided matching theory to achieve optimal matching between volunteers and rescue tasks. By combining multi-attribute similarity assessment with interview results, a Q-table and a reward table are constructed to overcome the shortcomings of relying on subjective weight settings in traditional dispatch methods and achieve dynamic adaptive optimisation of dispatch plans. The experimental results show that this method outperforms conventional methods in terms of matching accuracy, computational efficiency, and decision adaptability, particularly in terms of robust convergence and stability in large-scale data environments. This study provides a flexible and efficient solution for volunteer dispatch, which holds significant potential for application for volunteer management during public crises.