Journal of Quality Engineering and Management

Abstract Purpose: Today, supply chain management and accurate demand forecasting are considered key factors in improving productivity, reducing operational costs, and enhancing flexibility across various industries. The printing and publishing industry, as one of the sectors highly influenced by demand fluctuations, requires efficient strategies for supply chain management and optimal resource allocation. The objective of this study is to examine the impact of supply chain management on demand forecasting in this industry and to analyze the differences between the current state and the desired conditions.
Methodology: This study adopts a quantitative approach and utilizes real sales and book demand data for analysis. Machine learning techniques (particularly feedforward artificial neural networks with the backpropagation learning algorithm) are used to model and forecast demand. The performance of these models is compared with that of traditional approaches, such as time-series models, to assess improvements in forecasting accuracy.
Findings:  The results reveal that machine learning models, especially feedforward neural networks, achieve higher accuracy in demand forecasting compared to traditional methods. Moreover, the application of these models reduces the bullwhip effect in the supply chain and enhances coordination among its members.
Originality/Value: By presenting a hybrid model integrating supply chain management and demand forecasting based on neural networks, this research introduces an innovative approach to optimizing decision-making in the publishing industry. The integration of machine learning techniques with supply chain analysis can serve as a foundation for developing intelligent solutions in inventory management and production planning across similar industries.

Abstract Purpose: In today's competitive environment, selecting suitable suppliers plays a pivotal role in enhancing the efficiency and sustainability of supply chains. The LARG supply chain model, as a comprehensive approach, integrates various dimensions in supplier management. However, despite its critical influence on interaction success and supplier selection, the cultural dimension has received limited attention. This study aims to evaluate suppliers within the LARG framework while incorporating the cultural dimension to improve supply chain performance and achieve sustainable competitive advantage.
Methodology: This research developed a data-driven and forward-looking decision-making model for supplier evaluation and selection. First, key criteria were identified through literature review and expert consultation, and weighted using the Fuzzy Best–Worst Method (FBWM). Then, suppliers' efficiency was assessed and ranked using Fuzzy Data Envelopment Analysis (FDEA). Subsequently, the Random Forest algorithm was employed to predict future supplier performance, yielding highly accurate results.
Findings: The initial results highlighted the significance of sub-criteria such as greenhouse gas emission reduction, risk management, quality, and delivery speed in supplier evaluation. In the second phase, supplier efficiency was analyzed under various α-cuts and classified into three performance groups: high, medium, and low. The Random Forest model demonstrated high accuracy in forecasting supplier performance. Moreover, the paired t-test results revealed that incorporating the cultural dimension significantly improves the supplier selection process.
Originality/Value: The proposed model contributes to strategic decision-making by identifying key performance factors, enabling predictive evaluation, and employing robust analytical tools. This approach not only reduces risks and costs but also serves as a practical model for improving supply chain performance in similar industries.

Abstract Couplings are widely used in the industry and this equipment are always subject to defects and failures due to continuous rotation. Vibration analysis is a suitable technique for failure analysis and failure detection of rotating equipment. The purpose of this research is to analyze the failures that occurred in a coupling, whose data was collected in normal state and three failure states with four sensors connected to the coupling. For this purpose, two different types of feature extraction have been used, and seven machine learning algorithms and one deep learning algorithm have been used to classify situations. In this research, the performance of each of the implemented algorithms and the importance of extracted features have been investigated, and the role of sensors and their importance to reduce the number of sensors have been investigated. From the results of this research, we can point out the high importance of the features of the frequency domain in the accuracy of the implemented models, as well as the high efficiency of two sensors for classification.