Gráfico de controle 3-D uma abordagem utilizando o software R

Abstract

The need to increase profits by increasing the quality of products produced, mean that companies worry about the process of product development. Some statistical methods are used to monitor the improvement of products and assess their flaws, among them are the charts of Statistical Process Control (GCEP) proposed by Walter A. Shewhart. The Shewhart model is, however, difficult and bureaucratic processes where the product comes from different production workflows, parallel processes, which may or may not be correlated, due to the need for the construction of various control charts. It can be observed that the construction of control charts in parallel processes there is the possibility of the existence of two forms of variation with different characteristics: longitudinal and transverse. The 3-D control charts are an alternative to solve this problem, which involves the construction of three graphs: one to monitor the average and two to monitor the variability (longitudinal and transverse). Obtaining all results was with the help of statistical software public domain R (version 3.0.1), which proved easy to use implementation of the 3-D graph. Thus, the aim of this work is an approach about Control Charts 3-D as an alternative to the Shewhart Control Chart using the public domain software R.