個案資料

On June 1, 2016 a student was preparing for a summer job with Agarwal Automobiles, a vehicle fuel station owned by his father. The student had taken courses that covered supply chain management, including inventory management and forecasting. His father had suggested that, as a summer project, the student examine the fuel station’s retail inventory management practices with the intention of replacing the current simple rules with a more rational and complex planning model. The student needed some ideas to use as a guide toward a better ordering policy. The available data for the previous six months suggested that the company held an average ending inventory of ?2.1 million worth of fuel products to maintain average daily sales of ?0.52 million. The challenge was to reduce the inventory levels, while maintaining a high customer service level in fuel sales.

This case is suitable for a postgraduate level production and operations management course that discusses item level forecasting as well as the formulation of inventory management policies and models in a joint-product-ordering scenario. It can also be used in a graduate level quantitative technique or management science course to demonstrate the use of the mixed integer linear programming model for complex inventory planning problems, using a spreadsheet-based optimization tool like OpenSolver. The case provides opportunities to explore analytical techniques in a real business scenario. After completing the case, students should be able to do the following:<ul><li>Perform quantitative business forecasting, including model selection and decomposition methods used in seasonality and trend analysis.</li><li>Manage the inventory of multiple products in a retail scenario, including economic order quantity and periodic review policy for joint ordering, under maximum order size and separation constraints.</li><li>Apply the mixed integer linear programming technique, using a spreadsheet-based solver to formulate and solve an inventory management problem dynamically by using the forecast data, under capacity and separation constraints.</li></ul>

• inventory management, forecasting, retail