During the COVID-19 period, randomly arrived patients flooded into the hospital, which caused staffing beds to be occupied. Then, elective surgeries could not be carried out timely. It not only affects the health of patients but also affects hospital income. The key to the above problem is how to deal with uncertainty, which is one of the most difficult problems faced in the field of optimization. Specifically, surgery duration, length of stay, the arrival time of emergency patients, and whether they are infected with the SARS-CoV-2 virus are uncertain. Therefore, we propose a bed configuration to ensure that elective patients are not affected by non-elective patients such as COVID-19 patients. More importantly, we propose a planning model based on robust optimization and fuzzy set theory, which for the first time consider different categories of uncertainty in the same healthcare system. Given that the problem is more complex than the classical surgical scheduling problem, which is NP-hard in most cases, we propose a hybrid algorithm (GA-VNS-H) based on genetic algorithm, variable neighborhood search, and heuristics for problem traits. Specifically, the heuristic for operating room allocation is used to improve the efficiency, the genetic algorithm and variable neighborhood can improve the global and local search capabilities, respectively, and the adaptive mechanism can reduce the algorithm solution time. Experiments show that the algorithm has better calculation efficiency and solution accuracy. In addition, the elective surgery planning model under the new bed configuration model can effectively cope with the uncertain environment of COVID-19.
Keywords: Bed configuration; COVID-19; Robust optimization; Surgery planning.
© 2022 Elsevier Ltd. All rights reserved.