Article Preview
Top1. Introduction
Managing uncertainty while keeping a high service level has been an important issue in supply chains (SC). It can be found in many industries such as IT, defense, rail industry, education, health and safety, business, tourism, and social infrastructure service (Brownsword & Setchi, 2010; Suzuki & Kosaka, 2018; and Suzuki & Kosaka, 2018). In supply chain management, several studies used the SCOR model to analyze SC performance under uncertainties such as Akkawuttiwanich & Yenradee (2020) measuring SC agility and Fattah, Ezzine, Moussami, & Lachhab (2016) assessing inventory-management performance. Because business is driven by marketing and sales, uncertainties in sales and lead time variation are among the most common variabilities. Keeping a high level of inventory allows us to meet customer demand. However, carrying too much inventory results in higher holding and operating costs and requires more warehouse space and, thus, a longer time in order picking for shipping. On the other hand, maintaining a low level helps us minimize the costs. However, carrying too little inventory can cause stockout and lost sales resulting in low customer satisfaction. In fact, the inventory level should be optimized to reduce inventory operating costs and at the same time ensure probability of not being stockout and not losing sales. The higher the uncertainty on demand and lead time the more difficult and challenging is the holding of inventory in a SC and the calculation of safety stock to mitigate risk of stockouts should be adequately made (Nemtajela & Mbohwa, 2017). Inventory management is crucial for business since it enhances competitive ability by improving service level, reducing shortage cost, bringing more revenue from higher inventory turnover, and better utilizing warehouse space. To satisfy these benefits, the mix of stock and the different demands on that stock must be clearly understood.
To understand the mixed and different customer demands, a large number of SKUs should be categorized and prioritized. First developed by GE in the 1950s (Flores & Clay Whybark, 1986 and Guvenir & Erel, 1998), ABC inventory analysis is broadly used to categorize SKUs in three classes: A, B, and C in order of their importance. The method is like Pareto principle ranking SKUs in terms of how beneficial they are for achieving sales and revenue. It is often found that a small proportion of the SKUs lead to most of companies’ sales and revenue. The highest 20% of items are given the class A while 30% and 50% are categorized as B and C, respectively. To satisfy customer’s demand, the method assigns a high, medium, and low service level to A, B, and C classes as 96-98%, 91-95% and 85-90%, respectively. SKUs are categorized in three classes (not more or less) and their corresponding service levels are given in a range and it is not clear what value will be chosen for each class. This is difficult for making managerial decisions when there are no clear guidelines to optimally determine service levels. This problem is mentioned in Teunter, Babai & Syntetos (2010). Millstein, Yang & Li (2014) summarized three disadvantages of using classical ABC analysis: 1) Determining a service level for each group is unclear (as discuss before), 2) SKUs classification is made separately from and after service level decisions, so there is no guarantee that these ABC classes and the service levels are relatively optimized; and 3) ABC classes and service levels are determined prior to fitting them within an available budget so it cannot ensure that SKUs in the classes can fit under the inventory-operating budget. Though inventory managers reserve a huge budget, they cannot store SKUs over a warehouse capacity. For example, in a big city like Bangkok and suburbs where land is very expensive, the warehouse capacity is very limited. Therefore, warehouse capacity control is another issue in inventory management. To summarize, the popular and classical ABC analysis method categorizes inventory into three classes with associated service levels; however, it sometimes cannot be controlled within an available inventory-operating budget and warehouse space. Thus, classifying/prioritizing SKUs and controlling them under the budget and space must be done together to improve ABC inventory analysis.