VERTICAL - INDUSTRIAL AI

SCOPE

Cyber-Physical System Security (Across Segments), Digital Twins, Prognostics and Health Management (PHM)

ABSTRACT

Today’s industrial machines are hyper-connected to the Industrial Internet of Things (IIoT) and have highly capable Edge computing. As a result, they are constantly generating a wealth of diverse data (aka, Industrial Big Data) that needs to be analyzed and consumer by other machines or human operators. This requires advanced descriptive, predictive, and prescriptive analytics capabilities, which typically combine physics-based and data-driven models. The vertical on Industrial AI is addressing these issues by exploring three topics:

Robust Cyber-Security

to protect deployed AI systems from being breached for blackmailing, sabotage and industrial espionage.
Read more...

Effective Use of Digital Twins

to provide reliable, actionable information.
Read more...

Comprehensive Prognostics & Health Management

for industrial assets in Aviation, Oil & Gas and Power Generation to reduce or eliminate unplanned maintenance events.
Read more...

COMMON TECHNOLOGY CHALLENGES

  • AI-enabled process improvement (Metrics for efficiency, reliability, quality, overall cost)
  • Productization of AI-enabled services and products
    -streamlined provisioning to handle scale
    -low configuration overhead
    -self-monitoring performance
    -guaranteed performance
    -safety
    -cybersecurity
  • AI technology development and deployment
    -from sandboxes to full scale
    -technology sustainability
    -Model maintenance
    -new technology upgrades
    -technologists career development
    -etc…

COMMON BUSINESS CHALLENGES

  • Business Models to justify technology investment and monetize AI-improved processes (cost reduction via productivity)
  • AI-enabled services or products (growth)

Examples: Contractual Service Agreements; Condition based maintenance leading to zero unscheduled maintenance events.

  • Adoption of disruptive AI technologies
  • Strategy for R&D investments
  • Managing AI technology transfer

SOLUTIONS ATTENDEES MIGHT WANT TO KNOW

  • The different phases of cyber-security; When does cyber-security need to be considered.
  • Digital Maintenance, Repair and Operations (MRO) tools for proactive scheduled maintenance
  • Health forecasting with mixed known and unknown failure modes

TAKE AWAY

  • Integrate Cyber-security into AI solutions as a matter of design principle
  • Unlike other sectors, risk in PHM for Aviation can be quantified very explicitly for the purposes of financial decision-making (for initial and growth investments)

ORGANIZING TEAM 

Paul Ardis (GE Research, Niskayuna, NY)
Eric Bechhoefer (NRG Systems)
Piero Bonissone (PPB Analytics)
Jose Celaya (Schlumberger)
Neil Eklund (PARC)
Kai Goebel (PARC)
Naresh Iyer (GE Research, Niskayuna, NY)
Chetan Kulkarni (NASA)
Jay Lee (Univ. of Cincinnati)
Nikunj Oza (NASA Ames)
Liang Tang (Pratt & Whitney)
Abinav Saxena (GE Research, San Ramon, CA)
Weizhong Yan (GE Research, Niskayuna, NY)
Huafeng  Yu (Boeing)