ABSTRACT
Recently the US Navy has invested in supercavitating torpedo research and development. The supercavitating torpedo is a new technology to the United States and many challenges are associated with its design. One such challenge is the uncertainties involved. Furthermore, there are no full data sets to create probability distributions for classical reliability based analysis. Most information is in the form of expert opinion from the designers. Thus, evidence theory is a natural choice to determine reliability of the system. In this paper a cavitator structure is presented and optimized for shape and structural thickness with respect to stress, buckling and weight. Finally, evidence theory is used to handle the limited data situation as an alternative to classical probability theory for reliability assessment.