Companies that are involved in turbine repair service and maintenance often suffered from delayed work, which affected customer satisfaction and trust. Six-Sigma method is used to solve the delay issue. According to the research result, it is found that the capability of servicing process of a turbine in this company is 1.346s with DPMO of 560810.8. This condition is categorised as low, below the industry average in Indonesia of 2s with DPMO of 308.538. The main cause of the delay is with the unreliable work-hour prediciton, unreliability and unavailability of materials from suppliers, failure of the machines, and flawed scheduling. It is suggested to improve the service by work measurment to help predict the needs for repair schedule, rectifying the supply and maintenance system, as well as scheduling system according to the demands of repair service, repair time prediction, and the repair schedule of the machine itself