Predictive Analytics & Intelligent Control for Semiconductor Robotics

Project Lead: Apadix Corporation
Partners: Intel

Member % Cost Share: 50%
CESMII % Cost Share: 50%
Duration: 4 Months

Problem Statement

Establish a predictive maintenance, adaptive test, and visualization platform that supports Intel’s wafer Handling Robotics with a Quality Management focus of avoiding unplanned downtime and increase overall efficiency of Intel’s wafer processing capabilities.

Project Goal

The Adapdix solution will create SM Profiles and introduce a new AI-based software tool that will integrate full end-to-end, real-time AI/ML predictive analytics, maintenance, adaptive test, control and visibility to all equipment and processes in each individual process. This will include integration and real-time access of data from in scope equipment.

Back to projects

Leverage project outcomes for your manufacturing operations

Learn More

Technical Approach

The real-time, integration will occur across all equipment in scope. Adapdix’s EdgeOps™software ingests, accesses, and monitors machine and application log data to ensure consistent data connectivity and availability. The end-to-end system and integration will allow the EdgeOps™systems to collect, synchronize and react on any machine’s raw data and data from environmental sensors as well as use AI model outcomes to predict which tests can be skipped during functional test.

Key Tasks & Milestones

  • Performance Metrics & System Assessment – Adapdix provides a complete systems model with hypothesis and procedural steps to measure overall system function. Intel and Adapdix will discuss performance metric definitions and improvement opportunity areas/target values.

  • Data Collection and Observability – The EdgeOps™solution is optimized to work with time synchronized system, component and application log data that is clearly filtered by select variables. The EdgeOps™ suite of tools leverages the developer environment where data can be labeled and classified into segments of data for analysis and predictability.

  • Field Deployment (Deployment and Refinement) – At final delivery, the EdgeOps™ software will link back to the key components, applications, machine data and synchronized the data from these individuals to derive an end-to-end system view that’s used to report, visualize and monitor for all stakeholders.

  • Develop SM Profile – Design and develop SM Profile for use with CESMII Platform.

Potential Impact

  • Reduce Deployment Costs: EdgeOps software quickly and easily connects to any heterogeneous and complex manufacturing equipment in less than a day.
  • Adoption Costs: Installation costs are low and production capacity for the units being produced often increase by 10% to 15%.
  • Energy Efficiency and Productivity: Since our Platform enables the manufacturer and machine builder to use and monitor power usage in real-time, our deployments lower energy needed per unit manufactured.
  • Workforce: Our software does not require special training to install and/or deploy.

Benefits

  • Reusable SM Profile for other integrators, machine builders and/or manufacturers to use.
  • Demonstrated Use Cases showing measurable benefits of AI/ML at the Edge in a high precision low latency manufacturing environment.
  • Validation of ability to easily scale Edge AI/ML across multiple systems and broad geography.

Project Selection & Announcements

CESMII_RFP
CESMII_RFP