Multi-objective design optimization for mild moxibustion considering heat penetration and patient comfort

Abstract Mild moxibustion is a type of moxibustion treatment approach. The burning end of the moxa stick is kept at a fixed distance from the moxibustion skin, aiming to make the patient feel warm without burning pain. The key to enhancing the efficacy of mild moxibustion is to balance heat penetration and patient comfort by controlling the temperature field distribution. This paper proposes a multi-objective design optimization approach based on global sensitivity for mild moxibustion considering heat penetration and patient comfort. The purpose is to provide optimal mild moxibustion parameters ensuring heat penetration and patient comfort. Firstly, a parametric model of mild moxibustion temperature field was developed through the numerical simulation software of COMSOL. Secondly, global sensitivity was analyzed on the moxibustion parameters, such as the moxa stick burning temperature, moxa stick size, stick-to-skin distance, and ambient temperature. A functional decomposition algorithm was employed to calculate the global sensitivity indexes to accurately evaluate the sensitivity of moxibustion parameters. Thirdly, a multi-objective design optimization problem for the mild moxibustion was modelled. The non-dominated sorting genetic algorithm-II (NSGA-II) was adopted to solve the model to obtain the Pareto-optimal solutions. The proposed approach can solve the moxibustion parameter optimization problem balancing the heat penetration and patient comfort to achieve the optimal mild moxibustion efficacy. Consequently, the theoretical and clinical development of mild moxibustion is promoted.