A semi-supervised production scheduling method based on co-training deep neural network for smart shop floors

Traditional data-driven production scheduling methods rely on a large amount of labeled data, which is costly and difficult to acquire. However, smart shop floors possess abundant unlabeled data that is easily accessible and contains valuable information. Leveraging this unlabeled data can eliminate the need for expensive and laborintensive labeling processes while efficiently acquiring and optimizing scheduling knowledge. To address this, this paper proposes a semi-supervised production scheduling method based on co-training deep neural network. In the proposed method, a deep neural network is employed as the scheduling model to represent scheduling knowledge. And a co-training algorithm based on semi-supervised learning is designed to extract relevant information from unlabeled data, continuously enhancing the scheduling model during the training process. The effectiveness of the proposed method is validated in an aerospace component simulation production workshop. Experimental results demonstrate that the method effectively addresses the issue of limited labeled data in production scheduling, significantly reducing time costs without compromising scheduling requirements.