Presenting a Smart Manufacturing Model Using Decision Tree (Case Study: Mineral Processing Industry)

This study was conducted with the aim of presenting and evaluating a smart manufacturing model using a decision tree (case study: the mineral processing industry). Given the necessity of responding to environmental and competitive pressures, the integration of modern technologies—including the Industrial Internet of Things (IIoT), automation, and data mining—within the framework of smart manufacturing was considered. The required data were collected through questionnaires and extraction of operational information from mineral processing units. After initial data cleansing, seven key variables—including energy consumption, pollution level, technological status of production lines, presence of an Information Technology (IT) system, green planning, technical readiness, and production efficiency—were entered into the model. Modeling was performed using the Classification and Regression Tree (CART) method, and the factors affecting unit productivity were analyzed. The results showed that the technological status of production lines and the technical readiness of employees had the greatest weight in determining production efficiency, while green planning and the use of IT systems played complementary and influential roles in enhancing production and reducing pollution. It was also found that implementing green policies and improving technical knowledge levels can enhance process efficiency even in units with outdated technology. Comparing the decision tree model with other machine learning methods demonstrated its superior interpretability and suitable accuracy for industrial applications. Overall, the findings indicated that the use of data-driven smart models based on decision trees can provide an effective decision-support tool for improving productivity, reducing energy consumption, and achieving sustainability goals in the mineral processing industry. It is recommended that strategies for technology upgrading, continuous technical training of employees, and expanding green management be prioritized in future policymaking. This study was conducted with the aim of presenting and evaluating a smart manufacturing model using a decision tree (case study: the mineral processing industry). Given the necessity of responding to environmental and competitive pressures, the integration of modern technologies—including the Industrial Internet of Things (IIoT), automation, and data mining—within the framework of smart manufacturing was considered. The required data were collected through questionnaires and extraction of operational information from mineral processing units. After initial data cleansing, seven key variables—including energy consumption, pollution level, technological status of production lines, presence of an Information Technology (IT) system, green planning, technical readiness, and production efficiency—were entered into the model. Modeling was performed using the Classification and Regression Tree (CART) method, and the factors affecting unit productivity were analyzed. The results showed that the technological status of production lines and the technical readiness of employees had the greatest weight in determining production efficiency, while green planning and the use of IT systems played complementary and influential roles in enhancing production and reducing pollution. It was also found that implementing green policies and improving technical knowledge levels can enhance process efficiency even in units with outdated technology. Comparing the decision tree model with other machine learning methods demonstrated its superior interpretability and suitable accuracy for industrial applications. Overall, the findings indicated that the use of data-driven smart models based on decision trees can provide an effective decision-support tool for improving productivity, reducing energy consumption, and achieving sustainability goals in the mineral processing industry. It is recommended that strategies for technology upgrading, continuous technical training of employees, and expanding green management be prioritized in future policymaking.

کلیدواژه ها:

Smart manufacturing ، Decision tree ، Machine Learning ، Mineral processing industry