طراحي نمودار كنترلي فازي C در صنايع توليدي ﴿مطالعه موردي : در صنايع الكترو اپتيك صا ايران﴿صاپا-اصفهان﴾

چكيده انگليسي :

Designing a Fuzzy Control c chart in Manufacturing Industries Case Study SaIran Electro Optics Industries SAPA Isfahan Masood Amiri m amiri@in iut ac ir Date of Submission Department of Industrial and Systems Engineering Isfahan University of Technology Isfahan Iran Degree M Sc Language FarsiSupervisor Gholamali Raissi raissi@cc iut ac irAbstract A control chart is a tool that is commonly used to monitor and examine a process Control charts arethe simplest type of on line process control techniques Charts like EWMA CUSUM adaptive ControlCharts and Fuzzy Control Charts have been suggested because of both inexpressiveness of the classicalcharts and developing the control charts performance When the quality related characteristics such asappearance softness color etc cannot be represented in numerical values the control charts for attributesare used Product units are classified as either conforming or nonconforming depending upon whether ornot they meet specifications The number of nonconformities deviations from specifications can also becounted The binary classification into conforming and nonconforming used in the p chart and c chartmight not be appropriate in many situations where product quality does not change abruptly fromsatisfactory to worthless and there might be a number of intermediate levels Without fully utilizing suchintermediate information the use of the p chart and c chart usually results in poorer performance than thatof the x chart This is evidenced by weaker delectability of process shifts and other abnormal conditionssuch as unnatural patterns To supplement the binary classification data may be expressed by using fuzzynumbers In this thesis after a survey on designed classical control charts and fuzzy control c charts a newapproach for designing a dynamic fuzzy control c chart is presented The major contribution of fuzzy set theory lies in its capability of representing vague data Fuzzy logicoffers a systematic base to deal with situations which are ambiguous or not well defined In the literature there exist many papers on fuzzy control charts which use defuzziffication methods in the early steps oftheir algorithms The use of defuzziffication methods in the early steps of the algorithm makes it too similarto the classical analysis Linguistic data in those works are transformed into numeric values before controllimits are calculated Thus both control limits as well as sample values become numeric Whereas indesigning the suggested Fuzzy control chart a new description of Alpha cut using geometric terms andtrigonometric similarity which is a simple method without using deffusification has been developed forpreventing the loss of information included by the fuzzy samples Accordingly both the samples and thecontrol limits are presented by Fuzzy numbers in whole the charts designing process Meanwhile inpresented thesis the fuzzy process capability ratio was demonstrated for the first time In the meantime inother researches no method for determining the Alpha value in designing the fuzzy charts has presented Inthis thesis the value of Alpha cut is defined by applying a logical method From other remarkable topics inthis thesis presenting methods for identifying unnatural patterns and the efficiency of the suggested methodin calculating the type and error has been evaluated can be pointed out The suggested approach has beencoded applying Dephi and powerful software based on the mentioned innovations has been developed In conclusion the calculated results which gathered from SaIran Electro Optics Industries SAPA Isfahan have been presented for demonstrating the approach s efficiency Keywords Fuzzy quality control fuzzy control chart Alpha cut fuzzy process capability ratio fuzzynumber

اميري، مسعود

طراحي نمودار كنترلي فازي C در صنايع توليدي ﴿مطالعه موردي : در صنايع الكترو اپتيك صا ايران﴿صاپا-اصفهان﴾

كنترل كيفيت فازي , برش آلفا , شاخص فازي كارايي فرآيند , اعداد فازي