Piecewise-constant optimal control strategies for controlling the outbreak of COVID-19 in the Irish population

• Published in: Mathematical biosciences Vol. 330; p. 108496
• Main Authors: Náraigh, Lennon Ó, Byrne, Áine
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.12.2020; Elsevier Science Ltd; The Author(s). Published by Elsevier Inc
• Subjects: Basic Reproduction Number - statistics & numerical data; Betacoronavirus; Biostatistics; Communicable Disease Control - economics; Communicable Disease Control - methods; Computer Simulation; Coronavirus Infections - economics; Coronavirus Infections - epidemiology; Coronavirus Infections - prevention & control; Coronaviruses; COVID-19; Disease control; Disease Outbreaks - economics; Disease Outbreaks - prevention & control; Disease Outbreaks - statistics & numerical data; Economic models; Epidemiology; Humans; Ireland - epidemiology; Models, Economic; Models, Statistical; Optimal control; Original; Outbreaks; Pandemics - economics; Pandemics - prevention & control; Pneumonia, Viral - economics; Pneumonia, Viral - epidemiology; Pneumonia, Viral - prevention & control; SARS-CoV-2; SEIR model; Simulated annealing; Viral diseases; Ireland; SEIR model; Optimal control; Simulated annealing
• ISSN: 0025-5564; 1879-3134; 1879-3134
• DOI: 10.1016/j.mbs.2020.108496

We introduce a deterministic SEIR model and fit it to epidemiological data for the COVID-19 outbreak in Ireland. We couple the model to economic considerations — we formulate an optimal control problem in which the cost to the economy of the various non-pharmaceutical interventions is minimized, subject to hospital admissions never exceeding a threshold value corresponding to health-service capacity. Within the framework of the model, the optimal strategy of disease control is revealed to be one of disease suppression, rather than disease mitigation. •Presentation of SEIR-type model for COVID-19, model fitted to data for Ireland.•Aim is to find best strategy for non-pharmaceutical interventions (NPIs).•Optimal-control theory is performed on the model; cost function: cost of the NPIs.•State constraint: Hospital admissions remain below a threshold.•Suppression is cheaper than mitigation, over a 1-year time horizon.