سیاست سهام پایه تطبیقی ​​برای کنترل موجودی موردی اصلاح پذیر

This paper presents an adaptive base stock policy for a repairable item inventory control problem. Base stock level of a repairable item is updated based on the work-in-process (WIP) inventory level in repair facility with update frequency modeled as a separate tactical control parameter together with a standard base stock level. Stock-out situations are handled by emergency shipments, and priority shipments are used when updating the base stock level. A single-item single-location problem is considered, with repair facility having a limited capacity, and the problem is modeled by a two-dimensional continuous-time Markov chain, which is then solved explicitly by using matrix geometric methods. Numerical results are provided that show, for a given downtime target, the new policy achieves substantial reduction in inventory on-hand and significant saving in total cost.

A repairable item (spare part) is usually a technically complex part of some more complex capital intensive equipment, which is capable of being restored to acceptable operating condition after a failure. Repair service of an item is usually provided either by the original equipment manufacturer (OEM) that manufactures the equipment or outsourced to a supplier of OEM that manufactures this specific item. Customers of OEMs are companies operating usually in capital intensive industries (i.e., chemical industry, construction, semi-conductor manufacturing, aviation, and telecommunications). These companies incur high costs of downtime (due to lost sales, loss of productivity, or dissatisfaction of their customers) when one or more of their equipments are not operating. The cost of downtime is usually very high. As a consequence, customers are demanding high levels of spare part availability in stock. This causes much pressure on an OEM's cost budget for the after-sales maintenance support. Keeping spare parts stock at close locations to customers becomes expensive due to unpredictable, sporadic nature of equipment failures. Backordering is usually not desirable from a customer service perspective, and thus stock-out situations are usually handled via emergency shipments, in which case, a spare part is directly shipped from an external source to the customer. In order to avoid logistical defects, shipments are done via specialized couriers causing high shipment costs for OEMs. Stake holders put the pressure to keep operating costs down, and in contrast, customers demand high service levels. It is the duty of an inventory manager to continuously strive to control and to balance the tradeoff between operating costs and service level commitments. Cutting the operating costs down by meeting the agreed service levels has become important. We can state the typical characteristics of a repairable item inventory control system as follows: high service requirements, sporadic demand, high inventory costs, high shipment costs, and limited and expensive capacity to perform repair actions. These characteristics set the stage for the tactical design of repairable item inventory control systems. In this paper, we describe a practically intuitive and effective inventory control policy based on these special characteristics of repairable items. Our motivation is based on the existence of priority shipments in repairable item logistics operations of an OEM in semi-conductor manufacturing industry. Planners tend to respond to short-term supply shortages by priority shipments in order to avoid possible future emergencies. Independent of normal replenishment orders in the repair pipeline, a spare part is shipped directly to a warehouse from an external source other than the repair facility, because waiting for repair is considered as risky. This application is equivalent to a situation that, in presence of a stock-out risk, the planner temporarily increases the inventory position by using the priority shipments until the time that the risk is considered to clear out. On the contrary, when the planners foresee excess stock they then tend to decrease the inventory position. In this paper, we model this state-sensitive behavior by a mechanism that updates base stock levels based on the amount of WIP in the repair facility. If the WIP of an item is getting larger (thus the inventory on-hand becomes less), then the base stock level of that item is increased, and vice versa. We formulate the update procedure based on two tactical control parameters: the minimum base stock level and a parameter that refers to the frequency with which the base stock level is updated. Due to the recent evolutionary advancements in information technologies, monitoring the statuses of proprietary or outsourced facilities has become possible. Thus, revising and updating current plans or tactical control parameters based on the observed operational changes has become a popular practice. Usually, decision making with regard to updating depends on planners’ individual assessments, and it is not done in a systematic approach. In this paper, we aim to provide a formal model for this art of business, and shed light for practitioners by presenting quantitative insights about the benefits of such approach. Our model is also new for repairable items inventory management literature. By using analytical techniques, we showed how limited updating activities can be put into design of an inventory control policy for repairable items. The primary purpose of this study is to quantitatively evaluate the effectiveness of updating the base stock level of a repairable item based on the WIP level at the repair facility. We will refer to this situation as the dynamic case. The alternative case with no updating is called the static case. This paper is organized as follows: in the following, relevant literature is reviewed, and then a detailed formal description of the problem context is provided. In Section 2, a single-item single-location problem is modeled as a continuous-time Markov chain. Then, its stability condition and steady-state solution are provided. Numerical analysis of relevant performance metrics is given in Section 3. In Section 4, the paper is concluded, and ideas for possible extensions for future research are presented.

We have proposed a new inventory control procedure for repairable items. Our procedure was based on the idea that status information about WIP level in the repair facility can be used to update base stock levels to achieve cost efficiency and higher service level at the same time. We have inspired from the priority shipments done in practice. By using matrix analytic techniques, explicit solutions were provided, and our numerical results showed that the adaptive base stock level promises high potential for decreasing costs. The usage of real-time status information gathered from company databases to revise past decisions has been an important discussion topic for the practicality of the models that have been developed based on theory. There has been a lack of academic papers about this issue in repairable item inventory management literature. In this paper, we have discussed how real-time status information can be used to update control parameters (the base stock level in this paper) by explicitly incorporating the concept of update frequency. We think that the update frequency as a design variable has been ignored in previous similar studies, and this paper is the first to present quantitative insights about it. We have provided a formal mathematical model of this practically interesting problem, and showed potential benefits of it. The main insights delivered through this paper are: •(S0,r)(S0,r) policy generates significant reductions in the average inventory on-hand for a given service level or for the same level of average inventory on-hand it will improve upon the service level. •(S0,r)(S0,r) policy generates significant cost savings. The saving is larger for higher utilization levels or higher unit inventory holding cost. • A higher update frequency is preferable for the cases with higher service level targets, higher unit inventory holding cost or higher utilization levels. Further investigation of dynamically updating the base stock levels in a multi-item and multi-echelon setting remains as an interesting topic for future research. In a multi-echelon setting, insights can be sought for relative impacts of updating the base stock levels in a central or a local warehouse.