Instrumentation and Control System for AI-Based Skin Type Classification and Automated Skincare Sample Dispensing Using Computer Vision and Robotic Vending

The novelty of this study lies in integrating computer vision, CNN-based skin classification, AI recommendations, voice interaction, and robotic vending into a single retail platform for body skincare services. Unlike conventional retail systems, the proposed platform combines skin analysis, personalised recommendations, interactive communication, and automated product dispensing within one system. This study aims to develop and evaluate an AI-based skincare recommendation system that identifies customer skin types through image analysis and automatically dispenses product samples based on the recommendation results. Using a Design Science Research approach, a smart robotic vending prototype was developed by integrating image acquisition, AI processing, and robotic dispensing components. The prototype successfully demonstrated skin classification, recommendation generation, voice interaction, and automated sample dispensing. Functional testing confirmed the feasibility of real-time image acquisition, AI-based classification, recommendation delivery, and product dispensing. The results demonstrate the feasibility of supporting intelligent customer interaction through automated skin analysis, personalised skincare recommendations, and integrated sample dispensing.