Optimizing inventory control through a data-driven and model-independent framework
Machine learning has shown great potential in various domains, but its appearance in inventory control optimization settings remains rather limited. We propose a novel inventory cost minimization framework that exploits advanced decision-tree based models to approximate inventory performance at an i...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-01-01
|
| Series: | EURO Journal on Transportation and Logistics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2192437622000280 |
| _version_ | 1797961049940426752 |
|---|---|
| author | Evangelos Theodorou Evangelos Spiliotis Vassilios Assimakopoulos |
| author_facet | Evangelos Theodorou Evangelos Spiliotis Vassilios Assimakopoulos |
| author_sort | Evangelos Theodorou |
| collection | DOAJ |
| description | Machine learning has shown great potential in various domains, but its appearance in inventory control optimization settings remains rather limited. We propose a novel inventory cost minimization framework that exploits advanced decision-tree based models to approximate inventory performance at an item level, considering demand patterns and key replenishment policy parameters as input. The suggested approach enables data-driven approximations that are faster to perform compared to standard inventory simulations, while being flexible in terms of the methods used for forecasting demand or estimating inventory level, lost sales, and number of orders, among others. Moreover, such approximations can be based on knowledge extracted from different sets of items than the ones being optimized, thus providing more accurate proposals in cases where historical data are scarce or highly affected by stock-outs. The framework was evaluated using part of the M5 competition’s data. Our results suggest that the proposed framework, and especially its transfer learning variant, can result in significant improvements, both in terms of total inventory cost and realized service level. |
| first_indexed | 2024-04-11T00:53:19Z |
| format | Article |
| id | doaj.art-6f0ce16b542749b6902757f8b1600f73 |
| institution | Directory Open Access Journal |
| issn | 2192-4384 |
| language | English |
| last_indexed | 2024-04-11T00:53:19Z |
| publishDate | 2023-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | EURO Journal on Transportation and Logistics |
| spelling | doaj.art-6f0ce16b542749b6902757f8b1600f732023-01-05T07:26:49ZengElsevierEURO Journal on Transportation and Logistics2192-43842023-01-0112100103Optimizing inventory control through a data-driven and model-independent frameworkEvangelos Theodorou0Evangelos Spiliotis1Vassilios Assimakopoulos2Corresponding author.; Forecasting and Strategy Unit, School of Electrical and Computer Engineering, National Technical University of Athens, GreeceForecasting and Strategy Unit, School of Electrical and Computer Engineering, National Technical University of Athens, GreeceForecasting and Strategy Unit, School of Electrical and Computer Engineering, National Technical University of Athens, GreeceMachine learning has shown great potential in various domains, but its appearance in inventory control optimization settings remains rather limited. We propose a novel inventory cost minimization framework that exploits advanced decision-tree based models to approximate inventory performance at an item level, considering demand patterns and key replenishment policy parameters as input. The suggested approach enables data-driven approximations that are faster to perform compared to standard inventory simulations, while being flexible in terms of the methods used for forecasting demand or estimating inventory level, lost sales, and number of orders, among others. Moreover, such approximations can be based on knowledge extracted from different sets of items than the ones being optimized, thus providing more accurate proposals in cases where historical data are scarce or highly affected by stock-outs. The framework was evaluated using part of the M5 competition’s data. Our results suggest that the proposed framework, and especially its transfer learning variant, can result in significant improvements, both in terms of total inventory cost and realized service level.http://www.sciencedirect.com/science/article/pii/S2192437622000280Inventory control managementMachine learningInventory costDemand patternsLarge scale optimization |
| spellingShingle | Evangelos Theodorou Evangelos Spiliotis Vassilios Assimakopoulos Optimizing inventory control through a data-driven and model-independent framework EURO Journal on Transportation and Logistics Inventory control management Machine learning Inventory cost Demand patterns Large scale optimization |
| title | Optimizing inventory control through a data-driven and model-independent framework |
| title_full | Optimizing inventory control through a data-driven and model-independent framework |
| title_fullStr | Optimizing inventory control through a data-driven and model-independent framework |
| title_full_unstemmed | Optimizing inventory control through a data-driven and model-independent framework |
| title_short | Optimizing inventory control through a data-driven and model-independent framework |
| title_sort | optimizing inventory control through a data driven and model independent framework |
| topic | Inventory control management Machine learning Inventory cost Demand patterns Large scale optimization |
| url | http://www.sciencedirect.com/science/article/pii/S2192437622000280 |
| work_keys_str_mv | AT evangelostheodorou optimizinginventorycontrolthroughadatadrivenandmodelindependentframework AT evangelosspiliotis optimizinginventorycontrolthroughadatadrivenandmodelindependentframework AT vassiliosassimakopoulos optimizinginventorycontrolthroughadatadrivenandmodelindependentframework |