The Fit and Predict COVID-19 Using an Extended Compartmental Model in the Context of Indonesia Indrazno Siradjuddin1, a) Bella Cahya Ningrum1, b), Inta Nurkhaliza Agiska1, c), Arwin Datumaya Wahyudi Sumari1,2, d), Yan Watequlis Syaifudin3, e) and Nobuo Funabiki4, f)
1Electrical Engineering Dept., State Polytechnic of Malang, Jl. Soekarno Hatta no. 9, Malang, 65141, Indonesia
2 Faculty of Industrial Technology, Adisutjipto Institute of Aerospace Technology, Jl. Raya Janti, Sleman, Daerah Istimewa Yogyakarta, Indonesia
3 Information Technology Dept., State Polytechnic of Malang, , Jl. Soekarno Hatta no. 9, Malang, 65141, Indonesia
e) Corresponding author: qulis[at]polinema.ac.id
4 Electrical and Communication Engineering Dept., Okayama University
1 Chome-1-1 Tsushimanaka, Kita Ward, Okayama, 700-8530, Japan.
Abstract
. In a pandemic outbreak such as covid 19, a dynamic model is needed for quick and precise handling decisions. The presentation of a dynamical system can use the basic SEIR compartment model, in which critical (C) and death (D) states were considered, addressing more possibilities of mutual transitions between compartment states. SEIRCD estimates and forecasts the COVID-19 spread under uncertainties and constraints. This paper focuses on fitting the SEIRCD model with time-dependent basic reproduction number and health resource-dependent death rates to real data of COVID-19 in Indonesia. Data have been collected for a specific period of time, the best-fitting parameters were obtained, and the changes in the basic reproduction number over time were estimated. The presented model has included the transition probability of the infected state to the critical state and the transition probability of the critical state to the death state, based on different age groups, so that the number of hospital beds has also been cooperated for computing the compartment states. Using the best-fitting parameters of the SEIRCD model, the development of the compartment states in the near future was estimated. The simulation result shows that the obtained parameters of the SEIRCD model were reasonably satisfying where the simulated states from the model could fit the real data with a small degree of errors.
