Please wait a minute...
Chin. Phys. B, 2021, Vol. 30(12): 120203    DOI: 10.1088/1674-1056/ac3229
Special Issue: SPECIAL TOPIC— Interdisciplinary physics: Complex network dynamics and emerging technologies
SPECIAL TOPIC—Interdisciplinary physics: Complex network dynamics and emerging technologies Prev   Next  

Optimal control strategy for COVID-19 concerning both life and economy based on deep reinforcement learning

Wei Deng(邓为)1, Guoyuan Qi(齐国元)1,†, and Xinchen Yu(蔚昕晨)2
1 Tianjin Key Laboratory of Advanced Technology in Electrical Engineering and Energy, School of Control Science and Engineering, Tiangong University, Tianjin 300387, China;
2 School of Mechanical Engineering, Tiangong University, Tianjin 300387, China
Abstract  At present, the global COVID-19 is still severe. More and more countries have experienced second or even third outbreaks. The epidemic is far from over until the vaccine is successfully developed and put on the market on a large scale. Inappropriate epidemic control strategies may bring catastrophic consequences. It is essential to maximize the epidemic restraining and to mitigate economic damage. However, the study on the optimal control strategy concerning both sides is rare, and no optimal model has been built. In this paper, the Susceptible-Infectious-Hospitalized-Recovered (SIHR) compartment model is expanded to simulate the epidemic's spread concerning isolation rate. An economic model affected by epidemic isolation measures is established. The effective reproduction number and the eigenvalues at the equilibrium point are introduced as the indicators of controllability and stability of the model and verified the effectiveness of the SIHR model. Based on the Deep Q Network (DQN), one of the deep reinforcement learning (RL) methods, the blocking policy is studied to maximize the economic output under the premise of controlling the number of infections in different stages. The epidemic control strategies given by deep RL under different learning strategies are compared for different reward coefficients. The study demonstrates that optimal policies may differ in various countries depending on disease spread and anti-economic risk ability. The results show that the more economical strategy, the less economic loss in the short term, which can save economically fragile countries from economic crises. In the second or third outbreak stage, the earlier the government adopts the control strategy, the smaller the economic loss. We recommend the method of deep RL to specify a policy which can control the epidemic while making quarantine economically viable.
Keywords:  COVID-19      SIHR model      deep reinforcement learning      DQN      secondary outbreak      economy  
Received:  30 August 2021      Revised:  14 October 2021      Accepted manuscript online:  22 October 2021
PACS:  02.70.-c (Computational techniques; simulations)  
  05.45.-a (Nonlinear dynamics and chaos)  
Fund: Project supported by the National Natural Science Foundation of China (Grant No. 61873186) and the Tianjin Natural Science Foundation, China (Grant No. 17JCZDJC38300).
Corresponding Authors:  Guoyuan Qi     E-mail:

Cite this article: 

Wei Deng(邓为), Guoyuan Qi(齐国元), and Xinchen Yu(蔚昕晨) Optimal control strategy for COVID-19 concerning both life and economy based on deep reinforcement learning 2021 Chin. Phys. B 30 120203

[1] World Health Organization:"Coronavirus disease (COVID-2019) situation reports," (accessed on April 13, 2021)
[2] Chan J F, Yuan S F, Kok K H, To K K, Chu H, Yang J, Xing F, Liu J, Yip C C, Poon R W, et al. 2020 Lancet 395 514
[3] Tong Z, Tang A, Li K, Li P, Wang H, Yi J, Zhang Y and Yan J 2020 Emerging Infectious Diseases 26 1052
[4] Centers for disease control and prevention:2019 novel coronavirus, 2021 (accessed on April 13, 2021)
[5] Chetty R, Stepner M, Abraham S, Lin S, Scuderi B, Turner N, Bergeron A and Cutler D 2016 JAMA 315 1750
[6] Cutler D M and Huang W "A. Lleras-Muney, Economic conditions and mortality:evidence from 200 Years of Data," NBER Working Papers
[7] Enserink M and Kupferschmidt K 2020 Science 367 1414
[8] Fang Y, Nie Y, and Penny M 2020 J. Med. Virol 92 645
[9] Mandal M, Jana S, Nandi S K, Khatua A, Adak S and Kar T K 2020 Chaos, Solitons & Fractals 136 109889
[10] Huang J and Qi G 2020 Nonlinear Dyn. 101 1889
[11] Yu X, Qi G and Hu J 2021 Nonlinear Dyn. 106 1149
[12] Wang Z, Xia C, Chen Z and Chen G 2020 IEEE Transactions on Cybernetics 51 1454
[13] Huang J, Wang J and Xia C 2019 Chaos, Solitons & Fractals 130 109425
[14] Rong X, Yang L, Chu H and Fan M 2020 Math. Biosci. Eng. 17 2725
[15] Cui Y, Ni S and Shen S 2021 Chin. Phys. B 30 048901
[16] Tong Y, King C and Hu Y 2021 Chin. Phys. B 30 098903
[17] Arvind V, Kim J S, Cho B H, Mehmood A, Geng E and Samuel K 2021 Journal of Critical Care 62 25
[18] Vaid S, Cakan C and Bhandari M 2020 JBJS 102 e70
[19] Rustam F, Reshi A A, Mehmood A, Ullah S, On B W, Aslam W and Choi G S 2020 IEEE Access 8 101489
[20] Goldsztejn U, Schwartzman D, and Nehorai A 2020 Plos One 15 e0244174
[21] Berger D W, Herkenhoff K and Mongey S 2020 University of Chicago, Becker Friedman Institute for Economics Working Paper No. 2020-25
[22] Atkeson A and Andrew G 2020 Federal Reserve Bank of Minneapolis 1 25
[23] Wu J, Xu X, Zhang P and Liu C 2011 Future Generation Computer Systems 27 430
[24] Jamil A, Ganguly K and Nower N 2021 IET Intelligent Transport Systems 14 2030
[25] Fotuhi F, Huynh N, Vidal J M, Jose M and Xie Y 2013 Research in Transportation Economics 42 3
[26] Mnil V, Kavukcuoglu K, Silver D, Rusu A A, Veness J, Bellemare M G, Graves A, Riedmiller M, Fidjeland A K, Ostrovski G, et al. 2015 Nature 518 529
[27] Sharma A, Anand S and Kaul S K 2020 Image and Vision Computing 103 104022
[28] Chen M and Chan C 2021 Proc. Inst. Mech. Eng. Part D-J. Automob 235 541
[29] Mousavi S S, Schukat M and Howley E 2016 In Proceedings of the SAI Intelligent Systems Conference, London, UK, 21-22 September 2016, p. 426
[30] Kermack W O and McKendrick A G 1927 Proc. R. Soc. Lond. A 115 700
[31] Hu J, Qi G, Yu X and Xu L 2021 Dyn. 106 1411
[32] Covid-19 vaccination in Italy on April 13, 2021)
[1] Using agent-based simulation to assess diseaseprevention measures during pandemics
Yunhe Tong(童蕴贺), Christopher King, and Yanghui Hu(胡杨慧). Chin. Phys. B, 2021, 30(9): 098903.
[2] Prediction of epidemics dynamics on networks with partial differential equations: A case study for COVID-19 in China
Ru-Qi Li(李汝琦), Yu-Rong Song(宋玉蓉), and Guo-Ping Jiang(蒋国平). Chin. Phys. B, 2021, 30(12): 120202.
[3] Electron beam irradiation on novel coronavirus (COVID-19): A Monte-Carlo simulation
Guobao Feng(封国宝), Lu Liu(刘璐), Wanzhao Cui(崔万照), Fang Wang(王芳). Chin. Phys. B, 2020, 29(4): 048703.
No Suggested Reading articles found!