Document Type : Original Article
Authors
1
Assistant Professor, Department of Computer Engineering, Abadan Branch, Islamic Azad University, Abadan, Iran.
2
Assistant Professor, Department of Computer Engineering, Ilam Branch, Islamic Azad University, Ilam, Iran.
3
Assistant Professor, Department of Computer Engineering, Golpayegan Faculty of Engineering, Isfahan University of Technology, Isfahan, Iran.Golpayegan, Iran
4
Assistant Professor, Department of Business Management, Abadan Branch, Islamic Azad University, Abadan, Iran.
Abstract
Introduction: Determining supply chain priorities in smart industries with data-driven analysis and modeling methods is essential to achieve sufficient accuracy and identify key factors affecting supply chain efficiency. The use of this information improves the effectiveness of supply chain management. This article investigates and presents a quantitative approach for evaluating the priorities of the smart supply chain using data-driven prediction methods. The main objective of this paper is to provide a systematic and efficient method for determining priorities in the supply chain. In this approach, first, the key efficiency indicators in the supply chain are identified. Then, using data-driven prediction methods based on machine learning, the efficiency of each indicator is calculated for each element of the supply chain. The proposed approach has advantages such as systematicity, flexibility, practicality, and high accuracy. This method helps companies and organizations improve their management decisions by evaluating and determining supply chain priorities, optimizing performance, and enhancing processes.
Method: The innovation dimensions of this research include two main aspects. The first aspect focuses on two widely used industries equipped with Internet of Things (IoT) technology. The second aspect combines traditional supply chain analysis methods with machine learning algorithms. Initially, key performance indicators in the supply chain were identified. These indicators were extracted through a comprehensive search of articles in reputable scientific databases using keywords related to the smart supply chain. Then, using data-driven prediction methods, the efficiency of each indicator for each element of the supply chain was calculated. In this study, the DEMATEL matrix was used to analyze the interrelationships between indicators, and the prediction method using Support Vector Machines (SVM) was applied to assess the relationships between the criteria. Finally, the final weight of each indicator was determined by combining the results of DEMATEL and SVM, and the indicators in the supply chain were prioritized accordingly.
Findings: The results of this article show that flexibility is the most important criterion in the supply chain due to its ability to respond to changes and fluctuations in demand. The quality of the products and services provided ranks second, as higher quality increases customer satisfaction and trust in the brand. The total cost of the supply chain is third, and reducing costs improves profitability and competitiveness. Product delivery time is fourth, as fast and accurate delivery significantly impacts customer satisfaction. Finally, supply chain-related risks are ranked fifth, and effective risk management can mitigate potential issues. This prioritization helps organizations better allocate resources and improve supply chain performance.
Conclusion: Using systematic and precise approaches to prioritize supply chain criteria can serve as a practical guide for selecting and determining suppliers, implementing supply chain optimization strategies, and allocating resources. This research demonstrated that combining traditional supply chain analysis methods with machine learning algorithms such as SVM can improve the accuracy and efficiency of forecasting and decision-making. By enhancing the supply chain, organizations can improve their performance and optimize processes. Moreover, approaches such as Just-In-Time (JIT) strategy, Total Quality Management, and the use of new technologies can contribute to supply chain improvements. Building relationships with suppliers, analyzing data, and forecasting supply chain needs and challenges are also useful strategies.
Keywords
Main Subjects