In modern Industry, I4.0, artificial intelligence technology like Machine Learning (ML) and Deep Learning (DL) are increasingly used to fully realize the digital transformation. And is no news that Sustainability and Sustainable Digitalization are key. To this end, automatic anomaly detection is a concrete area for improvement in production lines, focusing on processes. In this paper, we investigate how to build an optimal Intelligent Defect Detection (IDD) model for furniture manufacturing, by taking the case of kitchen cabinets. We study (ML) Support Vector Machine, K-Neighbour Network, and (DL) YOLO models on different datasets and by analyzing training time, accuracy, precision, recall, F1-score, and robustness to lighting conditions. We contribute with an optimal IDD and a critical discussion. Our conclusions are based on the experiments conducted on the real world industrial manufacturing of kitchen cabinets.