A Vision for the Self-Aware Machine
Sparked by IT megatrends, manufacturers are currently undergoing an operational transformation with increased agility and efficiency as well as fewer operators per machine. Key technologies influencing this change include digital manufacturing, cloud computing, mobile application, and big data.
We believe that at the intersection of these technologies there is an opportunity to create a self-aware machine. In this article we are excited to introduce and share our vision and approach to building it.
So what exactly is a self-aware machine? In terms of manufacturing, it will provide the ability to capture, characterize, and predict the condition of the equipment, workpieces, and the machining process in real time and without human interference or action. On top of it all, it will actually notify plant managers, operators, or maintenance personnel before a failure in the line even happens. For manufacturing, this is the ultimate tool and a complete paradigm shift.
With our research we are focusing on machines used in fabricating industrial components, such as automotive engine, medical device, or aerospace parts. Self-aware machines will positively impact production time, cost, and quality of any manufacturing plant by reducing unplanned downtimes, adapting for workpiece variability, and enabling specification of fault-tolerant process plans. This is achieved through seamless integration of the process plan, machining parameters, machine condition, metrology, and tool condition information, providing proactive actions for plant employees to take. And it will even have the potential to give feedback to designers too.
We foresee the largest disruption with maintenance practices. Machine self-awareness could shift the industry from a reliance on a preventative paradigm (checking performance and replacing parts on a set schedule, regardless of whether there is an immediate need for these activities), to a predictive paradigm (schedule maintenance before failure actually happens). The ability of a machine to not only report its future failures, but also adjust its own functions according to its health condition is integral.
In order to estimate when operational failures may happen in the future or make independent, intelligent corrections or adjustments in the meantime, self-aware machines will require data analytics and awareness of three key aspects:
- Operating Condition – provides information of operating history, which is used to correlate with the quantified machine health condition to build a ‘finger print’ relationship between the incremental degradation and each of the operations
- Equipment Condition – quantifies the machine degradation correlated with sensor signals, work piece condition, process condition, and labels converted from human inspections
- Workpiece Condition – provides an accurate quantization of the workpiece and process condition, which can be achieved by calculating the deviation between the designed specs and the monitored specs
New data streams from a variety of sources and sensors (see image below) provide the flood of information and potential of making a self-aware machine a reality. The challenge, of course, lies in how the machine obtains and processes all of this information and awareness. We believe our inroads and advancements are putting a truly viable self-aware machine solution within reach.
At PARC, we practice open innovation, and are looking forward to hearing from potential groups interested in working with us to help realize this vision. With the right partners and our core technologies in design and digital manufacturing in place, we believe this can be accomplished in the near future. To learn more, please email email@example.com.
You can find out more about our Automated System Fault Modeling technology by downloading our Information Sheet below.
Our work is centered around a series of Focus Areas that we believe are the future of science and technology.
We’re continually developing new technologies, many of which are available for Commercialization.