A new analytical model for reliability analysis of a launch vehicle system with limited test data

Abstract

Reliability is one of the critical design parameters for the launch vehicles and its systems. When the systems are ready to fly the first time, only limited test data are available and accordingly reliability assessed will be very low. However, in most cases, the new systems are derived and developed using the knowledge and experience gained from the heritage systems to meet the fresh challenges. Hence, the reliability assessed with the minimum number of tests done on the new system does not truly reflect the inherent reliability of the system. In this paper, an approach and a new analytical model are developed for the reliability assessment of systems with limited test data, giving an accurate weighting for the tests and flight experiences with similar systems. The method gives a systematic procedure for arriving at the weighting factor for test data of the pedigree system, with due consideration of the similarities between the systems and various factors influencing system reliability. The method is illustrated with a case study of a newly developed liquid propellant rocket system. The model is validated using the available test and flight data of two propulsion systems with adequate flight experience. The analytical model is generic in nature and can be applied to reliability analysis of any system, which has considerable similarities with a pedigree system.