Evolutionary Fuzzy Control of Pandemics with Vaccination and Isolation Constraints

Vaccination and isolation are crucial strategies for controlling pandemics, but healthcare infrastructure constraints, such as limited medical resources and workforce shortages, often limit their effectiveness. This research explores the impact of these constraints on vaccination and isolation policies through the development of a Vaccination and Isolation Constraint-based Evolutionary Type-۲ Fuzzy Controller (VIETFC) intended to assist policymakers. A genetic algorithm optimizes VIETFC's parameters to reduce the number of infected individuals and provide adequate medical care. As a realistic case study, we consider the COVID-۱۹ pandemic and use a stochastic SEIAR (S۲EIAR) model to include the pandemic's inherent uncertainties and establish four scenarios with progressively increasing constraints, with the fourth scenario reflecting the most realistic conditions. Results show that VIETFC adapts effectively to resource limitations, maintaining a ۱۵% standard deviation despite a ۵۰% parameter variation. VIETFC surpasses Proportional Integral Derivative (PID) and type-۱ fuzzy controllers by offering interpretable solutions through a more practical resource allocation strategy. Future research should focus on enhancing the flexibility of this method for application to additional pandemic models.

کلیدواژه ها:

Pandemics ، Type-۲ fuzzy control ، Vaccination rate ، Isolation rate ، Genetic algorithms