Spare parts stock management under demand uncertainty

It is crucially important to minimize stok loads which is a major cost item for companies, to get advantage in competing conditions. Inventory is an important asset that helps business run smoothly but large amounts of inventory can lead to monetary and operational problems. In that case, inventory planning and stock management tasks become essential in today's highly competitive environment. Inventory planning models vary in different industries and using an inventory planning model designed in accordance with company's goals and structure, will considerably provide monetary gainings. By following appropriate inventory policies, firms can benefit by formulating a suitable an inventory replenishment plan, which achieves stated customer service levels consistent with lower levels of inventory. So, invetory plans should be enough to meet customer demand as well as it should be cost effective. Industries tackle with inventory management issues in different ways. Inventory control defines when an how much to order. Placing an order could be performed either continuous or periodic types. It is important for industries to implement type of inventory management system that combines costs and business type. A method that is best for one company to reduce inventory while increasing service level may not be ideal for another company in a different industry. Competition is in aviation industry is intense and maintenance is one of the largest operational cost categories and a major opportunity to reduce costs. In terms of maintenance functions of the airline, by setting appropriate stock control models for spare parts used in maintenance activities, flight operation costs can be kept under control, thus brings airlines competitive advantages. Because of high costs and long leadtimes of aircraft service parts, constructing an appropriate inventory policy is crucial. As the parts have intermittent demand structures, variability and occurence uncertainity of these parts exist, thus, accurate demand forecasting of these parts can lead to decreasing inventory cost and time gainings in operations. In aviation industry, when aircraft parts fail, under ideal situation, those parts should be in stock and should be replenished by further activities such as purchasing or repairing. If demands are satisfied immediately, and aircraft maintenance work can take place on schedule. Properly manged inventory also ensures manhour of maintenance personnel is utilized. However, if required spare parts are not available at that time, even purchase orders can be accepted by suppliers at once, purchasing them potentian as Aircraft On Ground (AOG) status will incur huge costs. Postponed maintenance due to spare parts shortage will can lead to flight delay or cancellation which will cause huge extra cost and customer dissatisfaction (Gu et al, 2015). A major challange in inventory management is handling uncertainty. An important consequence of uncertain demand is the risk of shortages and shortage cost. As shortage cost has an inverse relationship with carrying cost, this affects, what level of safety stock should be maintained (Tanthatemee et al, 2012). Safety stock is common way companies manage their inventory. It is used as a buffer against uncertainty (Hamlett, 2011). According to Huiskonen (2001), items classification is an essential part of the inventory management systems. Classification of parts is a relevant step as demand forecasting process may be driven by data collected from different classes. The traditional classification method is ABC Analysis, and it is widely use to determine service level of the parts. As classical ABC Analysis does not include all control requirements, multi-criteria ABC Analysis is used to classfy items in detail (Teixeira et al, 2017). In liteature, Zhang, Hoop and Supatgiat (2001) developed a model for the minimization of parts inventories subject to average service level and replenishment frequency constrains. The solution is obtained through a modified ABC classification. Syntetos, Keyes and Babai (2009) proposed an additional category of items D, in the ABC classification, composed of the critical parts, defined subjectively. Spare part inventories are different from other types of inventories. They are needed for maintenance and repair acitivities, require high invenstments and affect customer satisfaction (Rego ve diğ, 2011). Some items have high demand whereas other have intermittent demand. (Muckstadt, 2004). For intermittent demands, shortage may result in high costs, as in maintenance operations in aviation. In literature, Dekker et al. (1998) presented a storage decision rule for spare parts where at least one unit should be kept in inventory if its unit annual holding cost is greater than the expected annual shortage cost. Tavares and Almeida (1983) considered the case of demand having Poisson distribution and set inventory options of zero or one. The model evaluates these options through the comparison of their costs: regarding inventory of "one", it considers regular costs of holding and ordering; for "zero" inventory, it eliminates holding costs and increases the ordering cost because it comprises emergency purchases. The option for "zero" inventory occur when the average demand of the item is greater than a lower bound demand. Within this scope of the study, in inventory management under uncertainty, in a company facilitating in aviation industry, a sample stock model is proposed for spare parts stock management. In the frame of the study, first section focuses on relative strenghts and weaknesses of stock keeping. In the second section of the study, inventory notion is given and inventory management concepts are reviewed. In the third section, stock control models are examined. ABC Analysis, which is going to be used determining the stock criticality is introduced. In modellig decision concern, it is assumed that individual demand events occur independently, thus leads to the Poisson distribution, which is highlighted in this section. Fourth section includes literature review. Related works contributed to spare part stock management in literature is explained. Taxonomy of parts in used maintenance is mentioned. In the study, expendable invetory is considered as spare parts. Expendables can range from fasteners to filters and they are scrapped upon usage. They could be as expensive and more expendive than inventory assets in rotable or repairable parts (Wimer, 2015). Fifth section is application section. In this section, for an MRO (Maitenance Repair Organization) company, for the materials used in maintenance activities, among other stock models, lot-for-lot or (s-1, S) model is used to define minimum stock level as to maintain desired service level. To apply the proposed model, historical demand data for last two years' are collected and analyzed for the materials purchased from a sample vendor Y. At first, classical ABC analysis was done and purchases were compared to usages to see and catch dead stocks. It is seen that, parts seen as dead stock in the inventory has an important fiancial burden. After that, among demand forecasting methods, (s-1, S) stock management method is used for Poisson distributed demanded materials, an application and cost analysis is done, with this method, minimum stock levels are assigned to get minimum cost results in long term. In this way, besides cost cuts, a plan is proposed to be used a decision support model in inventory planning. Current stock level costs are compared with the stock levels in the proposed model. As it is needed to analyse parts in a detailed way, in the second step, multi-criteria classification is done, adding the criteria lead time, price and frequency of demand. After categorizing materials with multi criteria ABC analysis, parts are detailed with different categories, as very important, important, minorly important and not important items, and possible carrying costs are calculated under different scenario analysis. Parts are classified as slow moving and fast moving, according to their montly demand. Safety stocks are calculated for products having low demand rate and products having high demand rate. All possible costs are compared and differences among them are observed.