EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
3464. Decentralized vs centralized optimal strategies for Covid-19 control with compliance issues
Invited abstract in session WC-15: COVID-19 (2), stream OR in Health Services (ORAHS).
Wednesday, 12:30-14:00Room: 18 (building: 116)
Authors (first author is the speaker)
| 1. | Laurent Alfandari
|
| ESSEC Business School | |
| 2. | Debajyoti Biswas
|
| Management Information Systems, University College Dublin |
Abstract
We design a Mixed-Integer Linear Programming model for minimizing Covid-19 infections and deaths over a given horizon of several weeks. We allow two different strategies: centralized decisions applying homogeneously to all citizens at the national level, or decentralized decisions that can vary across regions or cities having different initial situations and risks. The decisions to make are the level of Non-Pharmaceutical Interventions (NPIs) implemented each week, like lockdowns, travel bans, school closures, or combinations of them. Shortages in doctors and hospital beds are considered in the model. Beyond the epidemic model that estimates the number of infections each week, one key aspect of this study is, in the decentralized scenario, the degree of compliance of the population facing more severe restrictions. Our results on real data show from which level of compliance the decentralized strategy starts to outperform the centralized strategy, under the same severity budget over the whole horizon. Compliance has been shown in previous studies to be decreasing in time, hence several kinds of compliance functions are considered to get robust results. Our results are also analyzed with or without the presence of vaccines.
Keywords
- Health Care
- Scheduling
- Optimization Modeling
Status: accepted
Back to the list of papers