فهرست نشریات علمی دسترسی آزاد ایران

Small and Medium-sized Enterprises (SMEs) are greatly hindered in selecting the most appropriate smart technologies as they strive to enhance productivity and competitiveness. This research suggests a machine learning-supported decision support system for smart technology choice in SMEs that systematically addresses implementation risks. The methodology adopts data mining approaches with reinforcement learning algorithms, identifying 24 applicable criteria in four categories technical, organizational, environmental, and risk using an expert Delphi panel. A prediction model was then developed using random forest algorithms and convolutional neural networks to analyze the prospects of successful implementation of smart technology. The model was validated with utmost rigor with data from 85 SMEs from various industries and was discovered to be 87.3% accurate. Results indicate that organizational culture, digital readiness, underlying implementation costs, and cybersecurity threats constitute the most important determinants shaping the successful implementation of smart technology in SMEs. The proposed decision support system possesses a dynamic interface through which SME managers can explore various scenarios and select the most suitable technology for their specific context. Through offering a new solution for managing uncertainty in decision-making, this research immensely adds to intelligent technology selection exercises for SMEs.