Sensitivity Analysis of Medical Waste Sterilization Shredder Using Surrogate Model

Abstract

Medical waste has been excessively generated in various medical facilities due to COVID-19, and its treatment has become an important concern. Previously, an optimized medical waste sterilization and shredding system was developed for hospital scale but due to increased demand, it is necessary to scale such a system for different facilities. Therefore, in this paper, a sensitivity analysis for the design variables of the shredding system has been conducted and a surrogate model is developed for stress estimation. The surrogate model was generated using LHS (Latin hypercube sampling), which can represent the overall information of the design domain with a limited number of samples. The surrogate model was then used to increase the number of samples for sensitivity analysis which helped in reducing the computational time for finite element analysis. The sensitive variables for the shredder system were then estimated using sensitivity analysis. Consequently, an efficient design framework for various capacities of medical waste shredder was suggested using sensitivity analysis and a data-driven surrogate model.