Change search
ReferencesLink to record
Permanent link

Direct link
En arbetsmodell för off-line-analys av variation i processer baserad på statistisk processtyrning: En fallstudie vid Siemens Industrial Turbomachinery AB
2012 (Swedish)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Organizations need to continuously improve to stay competitive. One effective method when working with continuous improvements is statistical process control (SPC). The method is used to find assignable causes of variation in, for example, a manufacturing process. The assignable cause should then be eliminated through improved work routines or with some other process improvement.Today many companies with modern manufacturing processes perform continuous quality inspections but often collected data are not used in further analyses. Using SPC these data can be used for process monitoring and to identify process improvements by follow-up analyses. However, due to the lack of knowledge of how to perform such follow-up analyses the possibility to identify potential process improvements is lost.Siemens Industrial Turbomachinery AB (Siemens) in Finspång, Sweden, is at the moment in the situation described above. Their quality inspections are only used to declare if a certain quality characteristic of the product lies within the stated tolerance limits or not. The information from the quality inspections is not used to analyze how the processes perform over time, which makes it difficult for Siemens to improve their processes.The purpose of this master thesis has been to help Siemens as well as other organizations with existing quality inspections without any specific method for process follow-up and process improvement. In this thesis we develop a sequential work model based on existing literature on SPC for companies to use when improving their processes. Although SPC usually is described as a method used on-line for daily monitoring and analyses of the process, the intention with the sequential work model is that process follow-up also can be done off-line when there is time. The result of using the sequential work model is a statistical analysis as basis for improvement decisions summarizing to what extent the process variation is affected by a set of predetermined factors. The sequential work model does not include the actual improvement work procedure. The sequential work model was tested in a case study at Siemens in Finspång at the department MPSH. With the sequential work model it was possible to confirm and reject what the production technician suspected to be assignable causes of variation in the manufacturing process at MPSH. The results in the case study also show in what way the different factors influenced the process variation and from that we recommend simple process adjustments to reduce variability.Since the sequential work model has a broad layout, in terms of what should be performed in each step of the model, it should be applicable for many types of processes – not only manufacturing processes.

Place, publisher, year, edition, pages
Keyword [en]
Keyword [sv]
Teknik, Statistisk processtyrning, SPS, Statistical process control, SPC, variation, duglighetsstudie, autokorrelation, arbetsmodell, sekventiell arbetsmodell
URN: urn:nbn:se:ltu:diva-56555Local ID: d4f44b50-6c2c-4941-adca-9ddb6548bd85OAI: diva2:1029942
Subject / course
Student thesis, at least 30 credits
Educational program
Industrial and Management Engineering, master's level
Validerat; 20120628 (anonymous)Available from: 2016-10-04 Created: 2016-10-04Bibliographically approved

Open Access in DiVA

fulltext(2541 kB)0 downloads
File information
File name FULLTEXT02.pdfFile size 2541 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Johansson, JonasSöderström, Carl

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 1 hits
ReferencesLink to record
Permanent link

Direct link