Abstract The paper suggests a method to optimize the quantity of spare power transformer components. The presented sparing policy is conceived to provide minimum annual cost consisting of expected capital cost for spares, failures renewal and load curtailment costs. The method identifies minor and major failures. Power transformer is a complex system, consisting of six components (functional parts). It is assumed that each component has two independent, competing failure modes: wear-out failure mode, modeled by two-parameter Weibull distribution, and a chance failure mode, characterized by an exponential distribution. Duration of failure renewal is not a deterministic variable. It is assumed that failure state residence time of each power transformer component is Weibull distributed (renewal times of power transformer components may follow any probability distribution, Weibull distribution being only a special case). The application of the method suggested and the benefits it provides are demonstrated for one transformer station (TS) 110/ x kV/kV with 2 × 31.5 MV A transformers in case of radial supplying of customers and when outage of one power transformer does not affect customers supplying. In addition, the influence of power transformer refurbishment on expected total cost has also been analyzed.