Questions

Roundtable 3 Questions

Starting Premises from Workshop 1 See Important Definitions Below

We agree that industry-wide data sharing for AI data and modeled system solutions are needed. What do we mean by this in terms of what solutions are needed and what problems will be solved?

We agree that managing and having ways to exchange and combine curated data is key. What data need to be shared and what methods, tools, standards and research are needed for security,  protection and trust?

We agree there needs to be business models, agreements, and organization and workforce changes for manufacturer to participate in an adoption cycle that leads to the generation and sharing of data that are necessary to develop AI models for scaled use.

 

Driving Questions for Roundtable 3

  1. What manufacturing challenges or opportunities would see the greatest benefit from being able to share and combine data across companies for AI applications; what curated data are of greatest value to be positioned for scaled use?
    1. We agree on the AI opportunities for shared data – can we characterize what are the problem statements, what kind of modeling is to be used, what data are to be shared, what are the sources of data and how would the data need to be curated:
      1. data exchange for interoperability among manufacturers for product and productivity across a supply chaim – what is the problem statement, what kind of modeling is to b
      2. sharing and combining data for benchmarking, training and testing, algorithms, methods and tools, i.e. images for testing feasture extraction approaches
    2. We agree on the factory floor AI opportunities – can we characterize what are the problem statements, what kind of modeling is to be used, what data are to be shared and what are the sources of data and how would the data need to be curated:
      1. combining on data and modeled systems for providers of common machines and operations and providers of algorithms to improve performance and precision and scale across the industry
      2. improving quality assurance of product and processes while they are being made
    3. How are value propositions best explained/approached with small, medium and large manufacturers to compel investment in adoption journey?
  2. What are the current options, what needs to be developed for the scaled use of shared data and AI models in a secure and privacy-preserved manner while providing assurances of intellectual property protection to cultivate trust across the industry?
    1. What privacy-preserving techniques (e.g., differential privacy, various forms of encryption, de-identification) would be needed for these challenges? What techniques exist today or need to be developed to allow the industry to reap the benefits of AI, synthetic data (e.g. GAN generated), sharing model building instead of data sharing?
    2. Are current methods sufficient to securely share/pool data while ensuring integrity, security and privacy, and instilling trust across the industry?
    3. What lessons can be learned/applied from other industries approaches to secure and privacy-preserved data sharing?
    4. How are the trust and security propositions best explained/approached with small, medium and large companies to compel investment in adoption journey?
  3. What are the business models for data sharing adoption to start and accelerate scaled AI implementation in the SMM supply chain so all manufacturers can participate in the ecosystem of data and models?
    1. We agree that that supply chain and factory floor AI application opportunities heavily depend on SMMs so what is needed to involve and scale SMM engagement.
      1. What are the roles for
        • Large OEMs?
        • SMMs?
        • Providers
        • Public-private partnerships
        • Government
      2. What research, technology or tools are needed to better enable AI and data for SMMs?
    2. We agree on data shared for industry wide purposes, i.e. an ecosystem of scaled curated data
      1. Who needs to do what for this to happen?
      2. If data privacy, integrity, and security are addressed, are we talking about provider based AI and data services and an industry data supply chain and exchange?
    3. Are there governance models that would help accelerate manufacturing adoption of AI and sharing of data and models?
      1. What is the role of public-private partnerships?
      2. What is the role of government?

 

Working Definitions

Artificial Intelligence (AI) in manufacturing refers to software systems that can recognize, simulate, predict, and optimize situations, operating conditions, and material properties for human and machine action.

Machine Learning (generally seen as a subset of AI) refers to algorithms that use prior data to accurately identify current state and predict future state, with the goal of improving productivity, precision, and performance.

Networking creates digital connections among devices, machines, equipment, databases, computer programs, and users, to provide the connectedness needed to exchange information, make decisions, and take actions.

Predictive Modeling is the use of data, AI, machine learning, simulation, and digital twins to assess, predict, and anticipate process, product, and operational behaviors for control, design, optimization, health, and failure prevention and mitigation.

Network Effects produce increased benefits for network users as the number of connected user nodes increases by expanding the availability of information and knowledge accessible to all.

A Resilient Supply Chain recovers quickly from an unexpected event[1]