An Inventory Management Decision Model for Managing Misplaced Spare Parts at Mycronic

Abstract: There are currently many spare parts in the Mycronic global supply chain which are wrongly stored in various local warehouses, even though the Mycronic stocking policy indicates they should be stored elsewhere. This results in increased inventory carrying costs and decreased inventory turnover ratio. According to employees working with spare parts distribution, there exists no standard procedure for deciding what to do with these spare parts. The purpose of this study is to present an inventory management decision model for managing misplaced spare parts, adapted with instructions to fit the Mycronic logistics organisation. The aim of this model shall be to increase inventory efficiency in the Mycronic global spare part supply chain. The first step to achieve the purpose was to find relevant theory and to construct the inventory management decision model. For this reason, the fields of inventory management and reverse logistics were studied and described in terms of methods and related costs, among other things. As a result, six different methods for handling misplaced spare parts that would theoretically increase inventory turnover ratio were identified: lateral transshipment, pull-back to the central warehouse, pull-back to cannibalise in the central warehouse, local cannibalisation, scrapping and remain in stock. An inventory management decision model was then created one step at a time, by using the identified methods as end nodes and combining them with theory-based decision nodes necessary to evaluate for each spare part. The result of interviews held with Mycronic personnel from various departments provided an insight into what the implementation of the decision model would look like, based on current practices for e.g. sourcing and distribution of spare parts, bookkeeping and inventory management. After analysing the interview results, conclusions were drawn with regards to how each decision node could be evaluated. To determine both the local and global need of a spare part, the conclusion was to use the past year’s demand for high demand spare parts and the average of the past two years of demand for low demand spare parts and let all stock exceeding that number be regarded as excess stock. To determine if a spare part is worth redistributing between local warehouses or returning to the central warehouse, the different activities involved in the different processes were identified and summarised in tables. Thereafter, methods for approximating the related costs were decided. For local cannibalisation, the conclusion was that it was not relevant to include as an option nor as and end node. This led to some modifications being made to the decision model, where certain decision nodes were removed so as to ensure the implement ability and relevance of the decision model. The final inventory management decision model suggests keeping any spare part that is needed locally in the local stock. Spare parts that are not needed locally but are needed elsewhere should be sent to where they are needed, provided that it is worth the redistribution costs. In case it is not worth redistributing, it is recommended to pull the spare parts back to the central warehouse if the profit from doing so outweighs the distribution costs. If a spare part is not needed anywhere, the decision model suggests pulling back the spare part to the central warehouse for cannibalisation as long as its subparts are needed, and it is worth the distribution costs. Lastly, if a spare part does not fulfil the requirements for any of the methods mentioned above, the spare part is suggested to be scrapped. Based on the conclusions of the study, instructions for evaluating the remaining decision nodes were created. The level of detail of the instructions varies for different decision nodes and also depending on the amount of information retrieved through the interviews. In some cases, the method for evaluating a certain decision node is clearly identified, whereas in other cases further investigation is required.