A Hybrid F-G-D Approach for Reliability Risk Assessment of Surgical Robots

This paper investigates a hybrid F-G-D approach based on the fuzzy comprehensive evaluation, grey relational analysis, and data envelopment analysis to assess the reliability risk of surgical robots. The method improves the analysis of all levels of indicators step by step, addresses the shortcomings of the previous evaluation method that separate the analysis and calculation of all levels of indicators. Firstly, by adopting the fuzzy comprehensive evaluation method, multi-level index levels are listed. The weight of the first level index was determined by using the grey relational analysis, which preliminarily shows influencing factor has a higher correlation with the risk degree. Subsequently, the evaluation results of the DEA are normalized as second-level indicators. Finally, the fuzzy comprehensive evaluation results are calculated. The F-G-D method was applied to assess the reliability risk of each component of the surgical robots. The comprehensive scores of the ten key components of the surgical robot (i.e., instrument box, mechanical arm, suspended top plate, trolley base, electric coagulation forceps, steel wire rope, mobile caster, image processing host, two-dimensional display, and base) were 1.1081, 1.6525, 0.9217, 0.9539, 1.7090, 1.5686, 0.9705, 1.2810, 1.2495, and 0.8739, respectively. Experiments demonstrated that the hybrid method can better predict the fault level of each component in the surgical robot with respect to improving the reliability and stability of surgical robots.