This　paper　proposes　the　symptom-based　SEMIR　compartmental　model　with　age　groups　and　fluctuations　from　the　perspective　of　the　managers　of　the　local　government.　The　basic　reproduction　number　of　the　total　population　is　given　by　the　next　generation　matrix　method,　the　group-based　basic　reproduction　number　are　then　followed.　The　unique　global　positive　solution　is　derived　by　using　the　Lyapunov　function.　Furthermore,　the　sufficient　conditions　of　the　stochastic　extinction　and　the　stochastic　persistence　of　the　SEMIR　model　are　studied.　When　Fuzhou　COVID-19　epidemic　is　chosen　from　the　surveillance　data　for　the　further　numerical　investigation,　the　data　fittings　of　the　SEMIR　model　reveal　that　age　groups　are　profitable　to　control　the　infection　scale　of　intensive　cases,　and　that　the　susceptibility　to　infection　of　age　groups　are　higher　than　those　within　the　total　population.　Alternatively,　the　scenario　investigations　show　that　the　prompt　treatments　are　conducive　to　the　reduction　of　infection　scale　for　intensive　cases.　©　The　Author(s)　under　exclusive　licence　to　Korean　Society　for　Informatics　and　Computational　Applied　Mathematics　2024.