<p>Managing reconfiguration time is an important challenge in active fault-tolerant controller design. Reducing reconfiguration time makes the active fault-tolerant controller a more suitable candidate for applications. In this paper, an optimal active fault-tolerant attitude stabilizer is proposed that considers reconfiguration time management. The design process is as follows: the small perturbation technique is used to convert nonlinear differential equations into linear differential equations, and polynomial functions are then used to convert linear differential equations and control constraints into linear equality and inequality constraints. The reconfiguration mechanism consists of two loops. The inner (nested) loop receives polynomial order and final time and finds the appropriate polynomial coefficients to satisfy problem constraints. The outer loop uses multi-objective genetic optimization to find a Pareto optimal set that represents the optimal values of final and reconfiguration times. The point on the Pareto optimal set which has the smallest norm is then selected as the final solution or design point. Finally, a feedforward neural network is trained and used to predict the appropriate values of polynomial order and final time for the arbitrary values of the initial condition and fault. The advantages of the proposed optimal active fault-tolerant controller are its ability to tolerate a broad range of faults that may not be tolerable using an infinite-horizon linear quadratic regulator and considering final time and reconfiguration time optimization in the spacecraft attitude stabilization problem.</p>