Optimizing Finished Goods Safety Stock Levels at a Swedish Food Company

Abstract

With rising interest rates increasing the cost of capital, businesses are focusing on reducing expenses. Tied-up capital, largely represented by inventory, is a critical cost driver. Therefore, optimizing inventory levels is crucial for controlling capital costs. Inventory levels consist of cycle stock and safety stock. Cycle stock covers average demand during replenishment, while safety stock mitigates the risk of stock-outs due to uncertainties. This thesis focuses on a Swedish food manufacturer that has implemented a new planning system, SAP IBP, featuring an Inventory Optimizer (IO) module designed to calculate safety stocks for its finished goods warehouses. Despite this, the company has been hesitant to rely on its outputs, leading to manual adjustments of safety stock levels based on staff experience. The primary purpose of this thesis is therefore to develop an inventory model that optimizes safety stock levels based on a target service level, enabling it to serve as a benchmark for assessing the IO's performance in the future.

To fulfill this purpose, the thesis utilizes the first four steps of the operations research framework: defining the problem and collecting relevant data, formulating a mathematical model, creating a computer-based solution based on that mathematical model, and validating the solution through practical testing. This resulted in an inventory model implemented in Microsoft Excel. The findings show that the new safety stocks could be reduced by 47% compared to the current levels at the case company while meeting the company's service level targets. Additionally, the model has been tested and the results indicate that it performs well enough to be utilized as a benchmark for the IO in the future.

Popular Abstract

Background
One aspect of inventory control is balancing low tied-up capital in inventory and high service levels towards customers. Our thesis addresses this aspect at a Swedish food producer referred to as the ''case company''. Traditionally, the company has relied on manual adjustments of safety stocks based on staff experience, often leading to excessive safety stock levels. This approach, while cautious, ties up significant amounts of capital unnecessarily. Therefore, the case company recently implemented a new planning system, SAP IBP, which includes an Inventory Optimizer (IO) module designed to automate the calculation of safety stocks. Despite this technology, a lack of trust in the system’s outputs has led to continued reliance on manual adjustments. Our thesis aimed to develop an inventory model that could optimize safety stock levels based on a target service level and serve as a benchmark for assessing the IO’s performance in the future.

Approach
To address the problem, an operations research framework was applied. The first step was defining the problem and gathering relevant data, mainly extracted from the company's database, including demand forecasts and lot size data. The second step was formulating a mathematical model to solve the problem. For the model to fulfill its purpose, theory was applied to practice by utilizing inventory control literature and connecting it to the case company processes and policies. The third step was transforming the mathematical model into a computer-based model, done by creating a Microsoft Excel file to calculate safety stocks. This model was then tested in the fourth step using historical data to assess its performance. These tests allowed us to understand whether the model in fact did achieve the service levels it intends to achieve.

Findings
The results of our model were promising as we found that the company’s safety stocks could be significantly reduced. More specifically, the case company has the potential of having 64,000 units less in stock on average (reduction from approximately 101,000 units to 37,000 units), should the changes be implemented. As mentioned, the model was tested by utilizing historical data and those tests showcased that the service level targets where achieved for each of the three tested periods. This showcases that the reduction in safety stocks is feasible and that the potential for cost savings for the company is high.

Further, mainly four factors were identified as influencing the safety stock levels. Firstly, The forecast error was identified as the main contributor to the output of the model. The purpose of safety stocks is to cover for future demand variability and since the forecast represents the plan for the future, the forecast errors represent the deviations from that plan and are therefore used to represent variability. It is therefore no surprise that they significantly affect the output of the model. Secondly, the target service level leads to higher safety stocks as it increases. The model calculates a safety stock based on a target service level, which makes it an important parameter to consider. Thirdly, the lead time is the period between placing an order and its arrival at the warehouse and affects the output of the model, but not in the same extent as the forecast errors. Lastly, the lot size decreases safety stocks as it increases. While this effect should be considered, its effect on the output is not significant.

Now that the model has been developed, the case company has been provided with insights into the theory behind safety stocks and which factors influence them. Additionally, the Excel file can be used by the company to evaluate the output of the IO when implementing its changes in the future.

• author: Jonasson, Kristinn and Blomborg, Emil
• supervisor: Danja Sonntag
• organization: Department of Mechanical Engineering Sciences; Production Management
• course: MIOM05 20241
• year: 2024
• type: H2 - Master's Degree (Two Years)
• subject: Technology and Engineering
• keywords: inventory control, forecast error, operations research
• other publication id: 24/5287
• language: English
• id: 9166736
• date added to LUP: 2024-06-25 12:23:39
• date last changed: 2024-06-25 12:23:39