Improving Machine Learning Predictive Capacity for Supply Chain Optimization through Domain Adversarial Neural Networks DOI Creative Commons
Javed Sayyad, Khush Attarde, Bülent Yılmaz

et al.

Big Data and Cognitive Computing, Journal Year: 2024, Volume and Issue: 8(8), P. 81 - 81

Published: July 28, 2024

In today’s dynamic business environment, the accurate prediction of sales orders plays a critical role in optimizing Supply Chain Management (SCM) and enhancing operational efficiency. rapidly changing, Fast-Moving Consumer Goods (FMCG) business, it is essential to analyze products accordingly plan supply. Due low data volume complexity, traditional forecasting methods struggle capture intricate patterns. Domain Adversarial Neural Networks (DANNs) offer promising solution by integrating transfer learning techniques improve accuracy across diverse datasets. This study presents new order framework that combines DANN-based feature extraction various machine models. The DANN method generalizes data, maintaining behavior’s originality. approach addresses challenges like limited availability high variability behavior. Using approach, model trained on training this pre-trained extracts relevant features from unknown products. contrast, Machine Learning (ML) algorithms are used build predictive models based it. hyperparameter tuning ensemble such as Decision Tree (DT) Random Forest (RF) also performed. Models DT RF Regressor perform better than Linear Regression Support Vector Regressor. Notably, even without tuning, Extreme Gradient Boost (XGBoost) outperforms all other comprehensive analysis highlights comparative benefits establishes superiority XGBoost predicting effectively.

Language: Английский

Leveraging deep learning for risk prediction and resilience in supply chains: insights from critical industries DOI Creative Commons
Waleed Abdu Zogaan, Nouran Ajabnoor, Abdullah Ali Salamai

et al.

Journal Of Big Data, Journal Year: 2025, Volume and Issue: 12(1)

Published: April 17, 2025

Language: Английский

Citations

0
Comparison of deep and conventional machine learning models for prediction of one supply chain management distribution cost DOI Creative Commons
Xiaomo Yu, Ling Tang, Long Long

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Oct. 15, 2024

Strategic supply chain management (SCM) is essential for organizations striving to optimize performance and attain their goals. Prediction of distribution cost (SCMDC) one branch SCM it's For this purpose, four machine learning algorithms, including random forest (RF), support vector (SVM), multilayer perceptron (MLP) decision tree (DT), along with deep using convolutional neural network (CNN), was used predict analyze SCMDC. A comprehensive dataset consisting 180,519 open-source data points make the structure each algorithm. Evaluation based on Root Mean Square Error (RMSE) Correlation coefficient (R2) show CNN model has high accuracy in SCMDC prediction than other models. The algorithm demonstrated exceptional test dataset, an RMSE 0.528 R2 value 0.953. Notable advantages CNNs include automatic hierarchical features, proficiency capturing spatial temporal patterns, computational efficiency, robustness variations, minimal preprocessing requirements, end-to-end training capability, scalability, widespread adoption supported by extensive research. These attributes position as preferred choice precise reliable predictions, especially scenarios requiring rapid responses limited resources.

Language: Английский

Citations

2
Improving Machine Learning Predictive Capacity for Supply Chain Optimization through Domain Adversarial Neural Networks DOI Creative Commons
Javed Sayyad, Khush Attarde, Bülent Yılmaz

et al.

Big Data and Cognitive Computing, Journal Year: 2024, Volume and Issue: 8(8), P. 81 - 81

Published: July 28, 2024

In today’s dynamic business environment, the accurate prediction of sales orders plays a critical role in optimizing Supply Chain Management (SCM) and enhancing operational efficiency. rapidly changing, Fast-Moving Consumer Goods (FMCG) business, it is essential to analyze products accordingly plan supply. Due low data volume complexity, traditional forecasting methods struggle capture intricate patterns. Domain Adversarial Neural Networks (DANNs) offer promising solution by integrating transfer learning techniques improve accuracy across diverse datasets. This study presents new order framework that combines DANN-based feature extraction various machine models. The DANN method generalizes data, maintaining behavior’s originality. approach addresses challenges like limited availability high variability behavior. Using approach, model trained on training this pre-trained extracts relevant features from unknown products. contrast, Machine Learning (ML) algorithms are used build predictive models based it. hyperparameter tuning ensemble such as Decision Tree (DT) Random Forest (RF) also performed. Models DT RF Regressor perform better than Linear Regression Support Vector Regressor. Notably, even without tuning, Extreme Gradient Boost (XGBoost) outperforms all other comprehensive analysis highlights comparative benefits establishes superiority XGBoost predicting effectively.

Language: Английский

Citations

1