Study On The Quality Control And Evaluation Of Non-normal Process Parameters In Electronic Components Manufacturing

With the rapid improvement of the requirement for the electronic produce quality, the Statistical Process Control(SPC) and Process Capability Index(Cpk) technology are widely used for the control and evaluation of the quality in the electronic components manufacturing.Because of the process feature, some characteristic parameters of the special process in the modern electronic components manufacturing do not follow the normal distributions, for example the shearing force of chip package,the leak rate of the packaging process etc. Because the distrubition of the charateristic parameters violates the basic hypothesizes of SPC, Cpk, the conventional control chart and process evaluation method is inapplicable for the speical process.So a new algorithm should be studied for the porcess whose characteristic parameters violate the basic premise of process control and evaluation.In the thesis, firstly based on the characteristic parameters non-normal distrubution of the special process, the application of the conditional control chart and process compability index on the non-normal distrubution parameters would cause the wrong judgment and results through the analysis of the simuation datas.The solution method that use the data transformation technology for the non-normal parameters control and evaluation is studied. The theory of the Box-Cox transformation and Johnson transformation are introuduced,and the SPC and Cpk models for the non-normal distrubution parameters are built respectively based on the the Box-Cox transformation and Johnson transformation.The analysis result on the simulation datas shows that the method introuduced in the thesis is correct and and effective using Minitab sofeware.Lastly the method discussed in the thesis is applied to the chip package process of a electronic component manufacturing company.The chip package process is one of thetytical,key process of electornic compent manufacturing.And the two transformation method are compared.The result shows that the method introduced in the thesis is effective and feasible.