Abstract
When the quality characteristics data is non-normal, measuring process capability using conventional methods can lead to the erroneous results. The established process capability methods are based on the normality assumption but most of the real world problems do not exhibit normal behavior in nature. Different methods (e.g. Clements non-normal percentile method and data transformation method) have been proposed to deal with the non-normal situation. Although these methods are in practice in industry, yet, there is a lack of literature to assess the accuracy of these methods under mild and severe departures from normality. In this paper, we review Clements non-normal percentile method and the proposed modified Clements method using Burr distribution for severe non normality cases. A simulation study using Weibull distribution is conducted and the simulation results are then compared with the commonly used Clements' method Final v, a case study based on the proposed method is presented using the real world data.