Resource Allocation Problem Based on Robust Optimization in Cloud Diagnosis Background

In the post-epidemic era of COVID-19, while medical resources are running under heavy load, the crowded phenomenon of large hospitals is more serious, and the graded diagnosis and treatment mode of remote cooperation of medical associations has been further developed. In this system, how to make the scheduling of expert resources not only meet the needs of known deterministic inpatients, but also adapt to the uncertain arrival of random dynamic outpatients has become a huge challenge. In this paper, the basic model without considering the arrival uncertainty of patients and the robust optimization model considering the arrival uncertainty of dynamic special patients are established, and the interval partition in the robust optimization model is carried out by using reinforcement learning method, and the influence of different interval partition on the solution of robust problems is studied, so as to realize the integrity of expert resource scheduling research. The results show that compared with the deterministic model, the proposed robust optimization model can not only significantly reduce medical costs, but also provide satisfactory services for patients.