시장데이타와 실험데이타를 이용한 제품 수명 예측

An ALT(accelerated life testing) method is to test a product in over-stressed conditions, and then the test result is extrapolated to an usual (normal stressed) condition. It is the major disadvantage of ALT method that the more extrapolation to an usual condition applies the bigger error is indispensable. Therefore a reliability model combining field failure data and laboratory test data is required in pratice. We propose several methods of estimating the failure rate of a product life which is assumed to follow an exponential distribution. Structural similarity and technological advances are also cosidered. Finally, We derive the acceleratio factor which can be used to predict the failure rate for a new product.