To solve the problems of complex decision-making and the difficulty of global optimisation involving many factors, this paper establishes a scheduling model for maritime search and rescue forces and uses a second-generation non-dominant genetic algorithm based on a reduplication strategy (DW-NSGA-Ⅱ) to solve these problems. Firstly, the probability of a successful search and survival (POSAS) is elaborated upon, taking into account the survival probability of the target and the relevant mathematical model of the time it takes the search and rescue force to arrive at the search area. A search and rescue force dispatch model is then established with the following optimisation goals: the maximum search success rate, the shortest arrival time of search and rescue materials, and the minimum search and rescue cost. Secondly, to solve the problems of crowding, distance failure, and poor global optimisation caused by multiple repeated solutions when solving the search and rescue force scheduling model using the second-generation non-dominant genetic algorithm (NSGA-Ⅱ), a second-generation non-dominant genetic algorithm based on a reduplication strategy (DW-NSGA-Ⅱ) is proposed. Finally, an accident is used as an example to demonstrate the use of DW-NSGA-Ⅱ and NSGA-Ⅱ to solve the model, with the resulting optimisation outcomes being compared. The experimental results show that the proposed method can consider marine environmental information, accident rescue requirements and search and rescue forces to formulate a reasonable and effective search and rescue force scheduling scheme. The DW-NSGA-Ⅱ algorithm's optimisation effect is better than that of the NSGA-Ⅱ algorithm under the same initial conditions, which verifies the superiority of DW-NSGA-Ⅱ in search and rescue force dispatch.