In the nuclear system thermal hydraulic analysis domain, mass, momentum and energy conservation equations for multiple phases are often solved with a set of selected empirical constitutive equations to close the problem. However, it is well known that the uniqueness of the solution is not yet proven thoroughly and not guaranteed. In order to provide a proof for the uniqueness of the solution for general situation can be quite challenging mathematically and cumbersome, but the authors think that it is still possible to study this problem under the physical domain of interest and at least provide a partial evidence that the obtained solution is unique or not within the confined region of interest. As a part of this effort, the authors propose to solve an inverse problem and track the initial condition of the solutions to check the uniqueness. In this paper, the authors used a large break LOCA example to illustrate the usefulness and implication of solving inverse problems to the evaluation of the nuclear system thermal hydraulic analyses, and how solving an inverse problem can provide a partial evidence to the uniqueness of the obtained solution.