For the countries or regions with insufficient medical resources and underdeveloped economies, it is an urgent issue to select hospitals and allocate patients to control the spread of COVID-19 (Corona Virus Disease, 2019). Considering the dynamic characteristics of the number and the severity of COVID-19 patients, this paper first presents a bi-objective dynamic location-allocation model for the designated hospitals with the limited medical resources, which simultaneously minimizes the total cost of designated hospitals and maximizes the treatment rate of patients. Then, this paper analyses the characteristics of the proposed model and designs a solution framework based on the Epsilon constraint approach to obtain Pareto optimal solutions. Finally, a series of numerical experiments are conducted to demonstrate the feasibility of the proposed model and effectiveness of the developed method on the basis of the epidemic data provided by Beijing Municipal Health Commission. The experimental results show that the bi-objective model can effectively trade-off the total cost of the designated hospitals and the treatment rate of patients, which can provide valuable guidance for the rational deployment of medical resources under the COVID-19 epidemic.