A voting-based hybrid machine learning model for predicting backorders in the supply chain

Q: How can I download an article?

To download an article from SID, first log in to the site, search for the article title, and click on the 'Download Article' option.

Q: How can I download an ISI article?

To download an ISI article on SID, enter the keyword or article title in the search bar, view the relevant results, click on the desired article, and select the 'Download Article' option.

Q: How can I access the SID database?

To access the SID database, visit SID.ir, create an account, and log in to access scientific resources.

Q: Is downloading articles from SID free?

Some articles on SID are available for free, while others require payment. Details are specified on the article's page.

Rezasoltani Arman; Jafarnejad Ahmad; Khani Amir Mohammad

مرکز اطلاعات علمی Scientific Information Database (SID) - Trusted Source for Research and Academic Resources

Journal Paper

Paper Information

Journal: DECISIONS AND OPERATIONS RESEARCH Year:2025 | Volume:10 | Issue:1 Page(s): 194-213

Download Full-Text

Persian Verion

View:

Download:

Cites:

Information Journal Paper

Title

A voting-based hybrid machine learning model for predicting backorders in the supply chain

Author(s)

Rezasoltani Arman | Jafarnejad Ahmad | Khani Amir Mohammad | Issue Writer Certificate

Keywords

Machine Learning

backordering

hybrid voting

Supply chain management

Demand forecasting

Abstract

Purpose: Backorder prediction is one of the most basic challenges in supply chains which may have a direct impact on operational costs, inventory levels and customer satisfaction. The main objective of this study is to come up with a voting based hybrid Machine Learning (ML) model for prediction of backorder that will enhance the accuracy of prediction.Methodology: In this study, an attempt was made to use hard and soft voting models based on XGBoost, CatBoost, Random Forest and LightGBM with model weights optimized through Optuna. The dataset used includes orders, inventory levels, supplier performance and other related features. To reduce data imbalance, the ADASYN method was used and also optuna parameter tuning was used to find the optimal model settings. The RFECV method was also used to identify key features affecting backorders.Findings: The soft voting model with accuracy of 0.9524 yielded the best performance when predicting backorders over all other individual ML models considered in this study. Moreover, inventory level related variables, supplier performance and demand predictability were also identified as most important in causing the occurrence of backorders. The proposed model was compared with other traditional methods and it was found that by using robust models in a voting framework, the forecast accuracy can also be improved.Originality/Value: The results indicate that the use of voting-based hybrid ML model can be a good mechanism for enhancing backorder prediction. This model helps organizations to more accurately manage order flow and their associated costs. Future research is suggested to use more advanced feature selection and optimization techniques such as genetic algorithms and deep neural networks to achieve better model performance.

Multimedia

No record.

Cites

No record.

References

No record.

Cite

Related Journal Papers

No record.

Related Seminar Papers

No record.

Related Plans