Unreasonable decisions for the lifetime performance index could have serious cost for business. It is important to use an efficient sampling scheme in order to achieve better statistical inference. Ranked set sampling (RSS) is one method to potentially increase precision and reduce costs obtaining a more representative sample. In this paper, RSS and median ranked set sampling are extended for hypothesis testing of the lifetime performance index when the lifetime data have the exponential distribution with mean θ. Critical values for this test are tabulated for introduced sampling schemes and comparisons between sampling schemes are considered by obtaining the power functions. The result indicates that using RSS and median ranked set sampling increases the power of the test.

Keywords