Grey Prediction Method used in Failure Prognostics for Electronics
CALCE Team:
Jie Gu and Michael Pecht
Objectives:
This study is to demonstrate the possibility of using grey prediction model in the failure prognostics for electronics.
Introduction:
Prognostics is a process of predicting the future reliability of the system by assessing the extent of deviation or degradation of a product from its expected normal operating conditions. Failure prognostics for electronics provides data that can be used to meet several critical goals, includes (1) giving advance warning of failures; (2) minimizing unscheduled maintenance, extending maintenance cycles, and maintaining effectiveness through timely repair actions; and (3) reducing the life-cycle cost of equipment by decreasing inspection costs, downtime, and inventory [1].
The grey prediction model, which is part on the grey system theory, can be one step in the failure prognostics approach to perform reliability prediction. Generally speaking, systems can be categorized according to the known knowledge. A black system is a system in which nothing is known about its internal structure, parameters and characteristics. On the contrary, a white system is one in which complete information is known [3]. However, in our daily lives, we often face situations involving incomplete information where some information about a system is knowable and some is not. In this situation, the grey system, which is between the white system and the black one, can be applied.
The grey system theory was developed by Deng [2] in 1982. The main function of it is the effective processing of the analysis, modeling, prediction, decision making and control with incomplete data. The grey prediction model (GM) has been applied in many areas, such as information [11], energy and power [4], industry and economics [9], accident and risk [12], engineering [8], and the environment [3]. In this study, grey prediction was used in failure prognostics of electronics for the first time.