This thesis presents multiscale hybrid natural fiber-reinforced nanocomposite structures as viable alternatives to the traditional synthetic carbon and glass fiber composites commonly used in industries like automotive, aviation, and aerospace. These alternatives were created using stochastic optimization methods—Differential Evolution, Simulated Annealing, and Nelder-Mead algorithms—to optimize critical buckling load, fundamental frequency, and factor of safety, while reducing weight and cost. A broad range of design variables were employed, including fiber volume fraction, stacking sequences, and the volume content of Carbon Nanotubes (CNTs) or Graphene Platelets (GPLs) in each layer. The effective material properties of matrices reinforced with CNTs or GPLs were determined using the Modified Halpin-Tsai equations and the rule of mixtures, accounting for the agglomeration effects of the nanofillers. Vibration, buckling, and failure analyses of multiphase hybrid fiber-reinforced nanocomposite structures were performed using both analytical methods (Navier's solution with First-order Shear Deformation Theory (FSDT) and Classical Laminated Theory (CLT)) and the Finite Element Method (FEM). A multi-objective optimization problem was strategically executed using the Penalty Function approach to propose optimal eco-friendly, lightweight, and cost-effective alternatives to conventional composite materials, aiming for maximum mechanical response with minimal weight and cost. Additionally, optimal nanocomposite driveshaft designs were proposed for future automotive applications, featuring hybrid Carbon/Flax/CNT structures with non-uniform fiber and CNT distribution, accounting for agglomeration effects. The results indicated that optimizing natural fibers with GPLs or CNTs in engineering structures offers substantial benefits, enhancing both environmental sustainability and composite material performance in terms of weight, cost, frequency, and buckling properties.