Submitted by admin on Mon, 08/08/2022 - 00:48

Inability to move towards autonomous (i.e., reduced staff) plants

Sequence Number 3 Industry Energy Banner Inability to move towards autonomous (i.e., reduced staff) plants How 5G enabled

Obtaining real-time transfers of data from a broad range of sensors can enable AI real-time decision-making to drive autonomous operations

Data Flows
Title Devices Icon Devices Description
  • Broad spectrum of sensors collects a wide variance of data (images/flow/temperature/speed)
  • Sensors: Fixed/Drones/Robotics (driving /swimming)
Title Connectivity Icon Connectivity Description
  • Imagery, time series, event, etc. data transport
Title Edge Compute Icon Edge Compute Description
  • Storage and compute, in some cases, at the edge level for all time-critical (real-time) decision making
Title Cloud Compute & Storage Icon Cloud Compute & Storage Description
  • Non-critical data stored in Enterprise cloud storage
  • Critical data to be copied from Edge to Enterprise storage
Title Applications & Services Icon Applications & Services Description
  • MV and ML applied to assist in decision making, from simple to advanced tasks over time
Title Inform Decision Makers Icon Inform Decision Makers Description
  • Final decision made by AI process, which will evolve to handle more complex decision making over time
Title Support Decision Making Icon Support Decision Making Description
  • End of process
Application Logic
Description
  • Data (event /timeseries/images/etc.) collected from various sensors (e.g., temperature, pressure, etc.) – currently present or to be installed on machinery
  • Assume broad spectrum and large numbers
  • Cameras collect high resolution video and image data to be accounted for in analysis
  • It will take time to get the right sensors in the right locations
Description
  • 5G/edge compute will enable AI to process time-critical data to take proper actions in real-time, which will be necessary for various actions
  • ML and MV processed data beyond prediction, building rule-based business logic for automated decision making
  • Iterative process for developing these models and achieving required accuracy needs a mandated step wise approach
  • SME involvement working with data scientists is required
Description
  • Incremental approach – automated decision making will be applied for low-risk tasks and will increase in risk level until complete autonomous operation is achieved (will require time)
  • Real-time explainable and trusted decisions using complex business logic – traceable to understand the logic flow of AI and build trust, with an emphasis on ensuring model is evergreen
Expected benefits Key value created