Visual Anomaly Detection (VAD) plays a critical role in ensuring quality control and efficiency across multiple domains, particularly in modern industrial manufacturing systems, where it has become crucial for maintaining product quality, optimizing workflows, improving cost-effectiveness, and enabling the early identification of defects. Available literature does not currently cover the full range of challenges present in this research field. This survey offers a comprehensive exploration of recent advancements in VAD within industrial manufacturing systems by exhaustively identifying and grouping the key domain challenges. We also classify, highlight, and analyze the latest progress in VAD. In addition, we enumerate introduced public datasets and their different visual modalities. Finally, through an in-depth analysis of the VAD field, we outline potential future directions and opportunities for VAD development.