In human-robot collaborative manufacturing, industrial robot is required to dynamically change its pre-programmed tasks and collaborate with human operators at the same workstation. However, traditional industrial robot is controlled by pre-programmed control codes, which cannot support the emerging needs of human-robot collaboration. In response to the request, this research explored a deep learning-based multimodal robot control interface for human-robot collaboration. Three methods were integrated into the multimodal interface, including voice recognition, hand motion recognition, and body posture recognition. Deep learning was adopted as the algorithm for classification and recognition. Human-robot collaboration specific datasets were collected to support the deep learning algorithm. The result presented at the end of the paper shows the potential to adopt deep learning in human-robot collaboration systems.

This study aims to investigate the implications of supply chain (SC) complexity (based on static and dynamic complexity drivers) on constituents of resilience capabilities to deal with disruptions. A systematic qualitative analysis based on critical incident technique has been applied on secondary data collected on disruption incidents. Findings indicate that most of the complexity drivers have positive influence on several resilience constit-uents; however, negative effects were observed as well. Static SC complexity drivers seem to have both positive and negative effects on resilience, while dynamic complexity drivers seem to reinforce SC resilience.