SKF Coupling in Hofors manufactures screws and couplings to the energy sector and the marine industry. The products have tight tolerances to be more attractive at the market through the tolerances. With increased competition on the market, the quality requirements increase, which means that many products fail to meet the quality requirements in order to avoid customer complaints. The purpose of this work was to improve measurement accuracy. To get an overview of the current situation, interviews and observations were made. A measurement system analysis was conducted by tests where a certain number of products were measured by a certain number of operators, with a certain number of repetitions. The test results were analyzed by a two-way ANOVA and Gauge R&R, which are common methods in measurement system analysis. A cause-and-effect diagram was used to investigate the causes of quality problems. The results indicated that the existing measurement systems were not reliable enough. The shortcomings may depend on the measuring device and the operator. The measuring device may be defective or have calibration defects. Operators can have different methods when measuring which contributes to variations in the measurements. The work led to suggestions for improvements that could imply higher accuracy in the measurement systems. For example, a tighter calibration interval of the measuring device would lead to reduced impact from the measuring device. A standardized education of measurement method, measurement environment and the importance of deviation management can also lead to improved measurement accuracy. / SKF Coupling i Hofors tillverkar skruvar och kopplingar till energisektorn och den marina industrin. Produkterna har snäva toleranser för att genom toleranserna vara mer attraktiva på marknaden. I och med ökad konkurrens ökar kvalitetskraven. Därför kasseras många produkter, som inte uppnår kvalitetskraven, för att undvika returer från kunderna. Syftet med arbetet var att förbättra mätnoggrannheten. För att få en överblick av nuläget gjordes intervjuer och observationer. En mätsystemanalys gjordes genom tester där ett utvalt antal produkter mättes av ett utvalt antal operatörer, med ett visst antal upprepningar. Testresultaten analyserades med en tvåvägs-ANOVA och Gauge R&R som är vanliga metoder inom mätsystemanalys. Ett orsak-verkan-diagram användes för att undersöka orsakerna till kvalitetsproblemen. Resultatet visade att det finns brister i det nuvarande mätsystemet. Bristerna kan bland annat bero på mätdonet och operatören. Mätdonet kan exempelvis vara defekt eller ha brister i kalibreringen. Operatörerna kan ha olika metoder som bidrar till variationer i mätningarna. Arbetet ledde till förbättringsförslag, som kan innebära högre noggrannhet i mätsystemen. Exempelvis skulle ett tätare kalibreringsintervall hos mätdonen leda till minskad påverkan från mätdonen. En standardiserad utbildning av mätmetod, mätmiljö och vikten av att avvikelsehanteringen följs kan också leda till en förbättrad mätnoggrannhet.

abstract: The Partition of Variance (POV) method is a simplistic way to identify large sources of variation in manufacturing systems. This method identifies the variance by estimating the variance of the means (between variance) and the means of the variance (within variance). The project shows that the method correctly identifies the variance source when compared to the ANOVA method. Although the variance estimators deteriorate when varying degrees of non-normality is introduced through simulation; however, the POV method is shown to be a more stable measure of variance in the aggregate. The POV method also provides non-negative, stable estimates for interaction when compared to the ANOVA method. The POV method is shown to be more stable, particularly in low sample size situations. Based on these findings, it is suggested that the POV is not a replacement for more complex analysis methods, but rather, a supplement to them. POV is ideal for preliminary analysis due to the ease of implementation, the simplicity of interpretation, and the lack of dependency on statistical analysis packages or statistical knowledge. / Dissertation/Thesis / Masters Thesis Industrial Engineering 2015