Parameter estimation in SIR epidemic model using dynamic selection preference with adaptive mutation factor enhanced differential evolution

Majeed, Bakhtawer; Soomro, Zuha; Ebrahim, Mansoor

doi:10.22124/jmm.2026.32034.2894

Parameter estimation in SIR epidemic model using dynamic selection preference with adaptive mutation factor enhanced differential evolution

Articles in Press

Document Type : Research Article

Authors

¹ Department of Computer Science, Faculty of Engineering, Science and Technology at IQRA University, Karachi, Pakistan

² Department of Computer Science, Faculty of Engineering, Science and Technology, IQRA University Main Campus, Karachi, Pakistan

³ Department of Computer Science, Faculty of Engineering, Science & Technology, IQRA University, Kaarachi, Pakistan

10.22124/jmm.2026.32034.2894

Abstract

To understand and manage the spread of infectious diseases in epidemiological models such as the Susceptible-Infected-Recovered (SIR) framework, it is vital to accurately estimate the transmission (β) and recovery (γ) parameters. This study proposes the dynamic selection preference with adaptive mutation factor differential evolution (DSP-AMF-DE) algorithm. The algorithm implements an adaptive mutation factor that dynamically regulates the balance between exploration and exploitation in the population over generations, and dynamic selection preference mechanisms that focus the selection of better candidate solutions and maintain diversity. Seven Pakistani regions covering several epidemic waves over a period of 671 days have been included in a multi-regional dataset. Robustness evaluation for multiple independent runs demonstrate the superiority of the proposed algorithm, which considerably outperforms six competing algorithms.

Keywords

Main Subjects