An ANN-based ensemble model for change point estimation in control charts

Abstract

Signaling in the control charts is usually followed by a substantial amount of delay, in which precise identification of the time when a change has occurred in a process simplifies the removal of change causes. This problem is referred to as change point (CP) estimation in the literature. This paper proposes a novel ensemble model to estimate CP under different processes’ changes in the phase II applications, known as ANNCP. It uses an evolutionary artificial neural network (ANN) as an underlying reasoning scheme to combine the predictions of multiple techniques and make the final decision. Specifically, a hybrid model of genetic algorithm (GA) and simulated annealing (SAN) with a new loss function is used to optimize the weights of ANN. The experimental results indicate that ANNCP can produce promising results under different conditions as compared with other state-of-the-art methods reported in the literature.