Use of Generalized distributions in Estimating Process Capability for Non-Normal data

ABSTRACT Process capability of a process is defined as inherent variability of a process which is running under chance cause of variation only. Process capability index is measuring the ability of a process to meet the product specification limit. Generally in traditional approach process capability is measured by 6 assuming that the product characteristic follows Normal distribution. In many practical situations the product characteristics do not follow normal distribution. Under this non-normal situation the traditional approach does not work. Many researchers pointed out some alternatives to handle this situations. In this work we will show the use of generalized quantile based distributions to estimate process capability indices.