.png){kind=icon}
.png){kind=icon}
Return
A hybrid approach to supplier selection and order allocation using Z-number-based data envelopment analysis and machine learning
.png){kind=icon}
PRE
AI
W
S
M
D
DOI：10.1080/17509653.2026.2659340.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In today’s competitive business, accurate evaluation of suppliers and relevant optimal order allocation to them is quite important and challenging. To overcome its underlying difficulties, this research proposes a hybrid approach for evaluating suppliers, forecasting demands, and allocating orders. In its implementation process, it initially applies Data Envelopment Analysis method based on Z-numbers to obtain more accurate, transparent and reliable data on suppliers. It then uses numerous machine learning algorithms to forecast future demands for various relevant items. Finally, it employs a multi-objective optimization model to minimize total supply costs, while maximizing order allocation to efficient suppliers and minimizing the number of suppliers to whom orders are placed subject to constraints of authorized delay in order delivery, capacity and demand satisfaction. The good accuracy of its generated results is related to its innovative part of using different machine-learning algorithms and integrating supply chain operation for more accurate evaluation of suppliers and relevant optimal order allocation to them. Its multi-purpose approach integrates supply chain decision-making regarding procurement costs while focusing on efficient and collaborative suppliers with strong, sustainable network connections. Its fuzzy programming approach has also played an effective role in the optimization process of its nonlinear multi-objective model.
Keywords:
Supplier evaluation and selection
order allocation
data envelopment analysis
machine learning
Z-numbers
multi objective modeling
Journal
IF:
2.6
Papers:
237
Citations:
739
Organization
Cited Papers