Comparação de gráficos de controle para atributos baseado na distribuição Birnbaum-Saunders.

Abstract

Product quality has become one of the most important issues in the global market. Statistical methods are used by companies to improve and maintain the quality of the process. Control charts have been widely used in the industry for quality monitoring, as they instantly indicate process variation so that engineers can determine the causes of the variations. Using the control chart, it is possible to determine whether the production process is stable or whether it is affected by any attributable cause. A quick indication of the cause of the variation in time helps to minimize the nunber of non-compliant items, and consequently substantially reduce the cost of poor quality. Attribute control charts are useful when you want to monitor the number of non-compliant items. In the last years, many studies have been developed in the construction of new graphics of control of attributes and with applications in several areas. In this research, we make a comparative study of the graphs of attribute control using various types of sampling when the lifetime of a product follows the BirnbaumSaunders distribution. The efficiency of the control charts is assessed through the average sequence length. The control chart is considered more powerful if it provides a quicker indication of the changed process compared to other control charts. Through simulation studies we evaluate the efficiency of the proposed graphics.