We have noticed that you are visiting from North American areas. Would you like to browse the US site (US.DFI.com) for a better experience?
The design of conventional factories where production units are highly single-purposed and rely on additional treatments has been gradually obsoleted by automation. Only by optimizing computing cores for the added services can the yield rate live up to the level of the AIoT future, which does not necessarily lead to a total makeover for the factories. The one major goal of DFI's AIoT solution is to cut the corners by adopting the latest computing platforms and software technologies seen in Smart Factories, all the while keeping your legacy production lines intact.
Factories dread an upgrade not because the machines are worn, but because the renewed IPCs, OS environments, and software integrations lack backward compatibility. It's a valid concern, but it's not always true. DFI's IPC refresh plan combined with Intel hypervisors makes sure all the legacy configurations and data processing flows are migrated effortlessly. With legacies on board, the AIoT facets then come into play － now that new-gen computing resource is embedded, automated processes and intelligence can be implemented on top to further optimize manufacturing efficacy.
Defect rates of multiple machines on the product line may multiply due to their physical separation, and these machines based on separate OS environments may each require its own IPC, abusing synchronization and maintenance efforts. With DFI's computing excellence, these platforms can all be consolidated onto one single physical IPC using Intel ACRN™ Hypervisor VMware technology. By virtually dividing resources into multiple areas, different OS platforms are able to be simulated on one IPC to tackle multiple operations, for example, Robot ARM on RTOS, HMI on Windows, and AOI on Linux, breaking the boundaries between various purposes.
Facial recognition, object detection, and quality verification that base on machine vision rely not solely on hardware computing capacity such as CPU and GPU, but also on the integration with learning mechanisms. Conventional architectures of IPCs were not designed to tackle the hunger for machine learning in an AI future, not until the engines became market available, such as Intel's OpenVINO™ deep learning optimizer. The toolkit uses trained models to optimize hardware execution for syncing with specific frameworks used by different applications, pushing the curve of computer learning onto an exponential growth.
In alignment with Intel's vision, DFI has been zoning in on designing comprehensive solutions that foresee AIoT and AI ubiquity. Since smartization of high-end factories is an inevitable future, IPC refresh and the incorporation of Intel's latest roll-outs will be one of the pillars in the success of Smart Factories.