In today's competitive manufacturing environment, companies are constantly looking for ways to improve. Because of this, many companies are striving to become "lean" by implementing lean manufacturing, which is a difficult process. To aid in the implementation of lean manufacturing, simulation was used to reduce the trial-and-error period of lean manufacturing and find to optimum approach to implement the lean manufacturing principle. In this research, a case study of implementing level loading of the production schedule for BullFrog International, L.C. will be examined. To make it possible to implement level loading, the thermo-former machine at the beginning of the operations was improved to allow quick changeovers. The changeover time was reduced by 60% and with a few additional changes changeovers could be completely external. In order to be able to conduct simulation experiments to find the optimum production schedule, cycle times were gathered for each operation and a simulation model was developed of BullFrog International, L.C. current manufacturing operations. Historical data was gathered of previous month's sales orders and orders were divided into three different groups. Group 1 the spa orders are roughly 50% single-pump and 50% double-pump, group 2 the spa orders are roughly 60% or more single-pump spas and group 3 the spa orders are roughly 60% or more double-pump spas. Using historical data, level loading production schedules were developed using lean manufacturing principles by reducing lot sizes to the smallest possible and still preserving the correct ratios. All of these suggested production schedules were tested with the simulation model and through various experiments, the optimum production schedule were determined. The optimum production schedules were implemented and the results were recorded. The results were an average throughput increase of 49.1% in group 1, an average throughput increase of 58.7% in group 2 and an average throughput increase of 58.7% in group 3. These results support the hypothesis that level loading will increase throughput in a complex manufacturing system where there is a high mix and low volume production schedule. The results also support the hypothesis that the trial-and-error period was reduced by the use of simulation.

En metod som används för lagerstyrning, och som fått ökat intresse inom forskningen de senaste åren, är klassificering med multipla kriterier (MCABC). Däremot har forskningen hittills tenderat att koncentrera sig på utveckling av klassificeringsmodeller snarare än på vägledning för tillämpning av modellerna. Studien påvisade med en litteraturstudie att forskningsområdet klassificering med multipla kriterier hade en begränsad mängd vägledning för tillämpning relativt mängden klassificeringsmodeller som utformats. Därför undersökte studien en modifierad lagerstyrningsstrategi. Syftet med studien var att finna produkter som lämpar sig för lagring av produktionskapacitet och att undersöka effekterna på färdigvarulagret om dessa produkter används för lageruppbyggnad inför säsongen. För att finna sådana produkter utformades en klassificeringsmodell med multipla kriterier, GTS-modellen (Good to Store). Modellen testades genom att ett antal scenarier simulerades på ett urval av produkter av det egentillverkade sortimentet i en fabrik. Detta gjordes i en fallstudie av ett företag inom processindustrin vars sortiment har säsongsvariationer. Resultatet av simuleringarna visade att lageruppbyggnad med GTS-modellen inför säsongen innebar att det genomsnittliga lagervärdet och antalet omställningar för det egentillverkade sortimentet minskade med ungefär 14 % vardera, jämfört med nuläget. Samtidigt kunde servicenivån bibehållas på i princip identisk nivå. Görs motsvarande jämförelse enbart på den simulerade delen av sortimentet blir minskningarna ungefär 35 % vardera. / Multi criteria classification (MCABC) is a method used for inventory management which has gained interest in research in recent years. However, research has tended to focus on the development of classification models rather than guidance on the application of the models. This study demonstrated, using a literature review, that the multi criteria classification field had a limited amount of guidance for application compared to the amount of classification models that were developed. Therefore, this study examined a modified inventory management strategy. The purpose of the study was to find products suitable for storing production capacity and to examine the effects on the finished goods inventory if these products are used for stock build-up prior to the season. A multi criteria classification model, the GTS (Good to Store) model, was developed to find such products. The model was tested by simulating several scenarios on a selection of products from the factory-made range. This was conducted in a case study regarding a company in the process industry whose range has seasonal variations. The simulations showed that stock build-up with the GTS model prior to the season resulted in an approximately 14 % decrease in average inventory value and number of setups for the factory-made range, compared to the current situation. At the same time, the service level was almost completely maintained. If the comparison is made exclusively on the simulated part of the assortment, the decrease will be approximately 35 %.