This investigation is concerned with the availability characteristics of a multi-component repairable system with coverage factor and reboot delay. The time to repair the operating units is considered to be generally distributed whereas the life time, recovery time and reboot time of the units are assumed to be exponentially distributed. By introducing the supplementary variable corresponding to remaining repair time and using recursive approach, we evaluate steady state probabilities of the system states. The availability and failure frequency for different system configurations are obtained explicitly for some special distributions of repair time such as exponential, gamma and uniform distributions. The numerical experiments are performed to illustrate the computational tractability of the derived analytical results. The sensitivity analysis is carried out to examine the effect of system parameters on the reliability indices. Neuro-fuzzy results are also depicted graphically to explore the possibility of soft computing approach for the performance prediction of complex redundant systems.