Implementering av lean i ett gjuteri med låga volymer och hög variation

• Main Author: Sandberg, Patrik
• Format: Others
• Language: Swedish
• Published: Karlstads universitet, Fakulteten för hälsa, natur- och teknikvetenskap (from 2013) 2014
• Subjects: Värdeflödesanalys
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-33260

Valmets own foundry in Karlstad has during a long period of time had problems being profitable. The goal today is to turn this around. Today there are three different productionlines; engine blocks, Yankeecylinders and components for the wind turbine industry. This master thesis has focused on examining if value stream mapping is a useful tool to use in a foundry, and therefore outside the typical manufacturing industry. The target has also been to evaluate if there is a possibility to improve the flow in the foundry and also the results. During the 20 weeks this work has been done at the foundry data has been collected and analyzed. It has been possible to confirm irregularities in the production at several places. Due to limited time the focus of this theses has been at three different areas: The sand core manufacturing, the forming of concrete for Yankee cylinders and also the molds for the castings of engine blocks. Together with the measured times and values a new proposition of the sand core manufacturing has been developed. There has also been a request in examining the possibilities of changing from three shifts to two shifts at the sand core manufacturing. This has been examined and economical calculations has been made. Simulations have been done in a program called Flexsim. The Flexsim model was a simplification of the real situation which therefore affects the reliability of the result. To continue improving the model is something that is recommended for future thesises. As a complement to the simulations made in Flexim manually simulations was made too, to examine the possibilities of reducing the number of shifts from three to two. The result was clear, it was possible to go down to two shifts. It would though cost approximately 800.000 kr more per year. But with this investment it is possible to produce 5 motor blocks a week steadily for a long time which helps with the payoff of the investment. Every sold motor block results in a profit of approximately 150.000 kr. It was also found that it is problematic to make use of value stream mapping in a process that is large, contains a large number of sub-steps and also is very irregular. However, the value flow analysis could be used as a complement to the other simulations. For the production of Yankee cylinders, it was decided that the focus should be on the introduction of a new cement mixer that Valmet bought into. The problem has been how it will be able to reach all three pits, as these are large and extends over a large area. After discussion, it was decided to proceed with a concept that means that the entire cement mixer will be mobile. Valmet themselves then took the decision that the results shall be recorded and then transferred to Valmet since the introduction of the mixer extends too big for this thesis. In summary, it has been noted that there are a lot of things to do. The foundry is in great need of improvements in productivity, standardization, 5S and so on, and there is a huge potential in the future to improve profitability. Even outside the three areas that this thesis has focused on.