How AI is Transforming Manufacturing Webinar
Agenda TIME TOPIC KEY ITEMS PRESENTER Rob Capozziello • 3 Introduction & Housekeeping About Zoom, Q&A, Agenda EVP Services, Mariner Mitch Landess • 5 Conexus Intro Conexus overview VP Innovation and Digital Transformation Conexus • Digital feedback loops David Breaugh Transforming Quality 10 • The next frontier of manufacturing efficiency Manufacturing Business Performance with Applied AI • High impact applications in manufacturing Lead, Microsoft Data Driven Decision Making • Creating real change in manufacturing Robbie Jones 20 • Case Study: Global Chemical & Textiles Company for the Factory Floor Enterprise Sales Manager • Case Study: Automotive Fabric Inspection w/Deep Deep Dive into Deep Learning Learning Stephen Welch 40 • Myths and Challenges in Deploying Deep Learning for Visual Inspection VP Data Science Mariner • Driving real business value and how to get started Moderator: Rob 15 Q&A Capozziello
Multiple catalytic innovations are enabling digital feedback loops Microsoft
Leveraging innovation to enable the next frontier of manufacturing efficiency CUSTOMER SERVICE ENG Next gen efficient frontier INTELLIGENT OPERATIONS PLATFORM Scale innovation across value chains MFG DIST ▪ Connectivity ASSEMBLY ▪ Flexible Automation Amplify with algorithmic ▪ Intelligence decision making and Law of diminishing returns automated execution Move from reactive to predictive with big data, Top challenges with current CI programs machine learning, IoT ▪ Making changes (and results) sustainable ▪ Deploying what works with speed and scale Leverage the cloud to ▪ Finding and unlocking new funding sources connect, automate, visualize end-to-end business view Microsoft
Artificial Intelligence - High impact applications in manufacturing Object Speech Machine Machine Knowledge BOT Machine Recognition Recognition Translation Learning Mining Services Teaching Microsoft
Data Driven Decision Making for the Factory Floor Robbie Jones, Mariner
Mariner – Manufacturing Analytics Analytics Teams-as-a-Service Right-Sized, Agile AI/ML, IoT, Analytics, Data Science Teams IP Solutions Project-based Services BI/DW Analytics AI/MACHINE LEARNING Reporting DATA SCIENCE MODERN DATA DATA VISUALIZATION WAREHOUSE/ESTATE CLOUD DATA PLATFORM IOT AND IOT EDGE INFORMATION LIFE CYCLE MANAGEMENT & GOVERNANCE
Spyglass Connected Factory Your Virtual Production Manager Be the Change Agent
The Problem with Industrial Process Improvement Manufacturers have made From Adam Smith’s “The Wealth of Nations” through the Toyota significant investments in Production System, manufacturers have historically sought ways to continuous improvement eliminate waste from industrial process systems. methods SPC, 6 Sigma, Lean have From Statistical Process Control to Six Sigma to Lean, these methods have delivered significant delivered measurable improvements leading to reducing waste. improvements Much of the value has For mature companies, the value from low hanging fruit has been captured. been realized. New To gain more value, manufacturers must leverage new techniques and approaches are required technologies. to get more.
Spyglass Connected Factory Your Virtual Production Manager Alerting/Monitoring OEE Analytics Predictive Maintenance When detecting emergent Benchmark your progress. Reduce uplanned downtime by conditions sooner rather than Measuring plant productivity predicting the probability of later will save time/money is the first step towards failures that impact operations improving it
Spyglass Connected Factory Case Study – Milliken & Company Global Chemical & Textiles Company
Business Challenges • No condition monitoring on valves & motors • Root cause analysis on failed equipment was difficult or nonexistent • Engineers spend hours per week producing spreadsheets and analytics
Results • Prioritized list of equipment that can be serviced during an unplanned outage • AI on telemetry statistics to identify root-cause of failures on critical equipment • Engineers spend less time reporting and more time solving problems
Industrial Equipment Manufacturer Results • Collect telemetry on presses for lot traceability and Kaizen • Ultimately use production rate information to change shift patterns • Resulting in increased throughput and improved overall production quality metrics
Industrial Equipment Manufacturer Results • Monitor equipment for improper usage and failure to conduct preventative maintenance process. • Deliver alerts and reminders to control room operators while also feeding the information back into their management systems for full visibility into compliance with SOP (Remember – Your Virtualized Production Manager)
Automotive Manufacturer Results • Apply basic IoT enabled devices to log important process variables (temperature, speed, pressure, etc.) and automatically match them to lots and SKUs and real-time Statistical Process Control (SPC) Out-Of-Control (OOC) alerts • Use lot numbers to match images with production telemetry and actual routing to alert department supervisors to the upstream process responsible for visible defects and provide a situation report for each incident
Spyglass Connected Factory and AI AI to monitor the accuracy of current MCBF numbers (mean cycles between failures) for your equipment • and compare to the planned outage schedule AI forecasts the cycles by the time the outage is due, with a confidence limit, and suggests if this machine • should have a PM (planned maintenance) work order generated for the next planned outage, or if it can wait until the next outage AI can create personalized MCBF. Some of the products you make are more destructive to your tooling and • equipment than other products, in effect altering the MCBF Using a personalized MCBF can improve your forecasts and ultimately improve uptime. •
Spyglass Connected Factory and AI Lots of part numbers, lots of machines, and lots of operators doing hundreds of setups and stoppages • within thousands of hours of scheduled work Knowing where to find opportunities in your OEE can be problematic • Humans suffer from ego and exhaustion and often shed analytic workloads by focusing on outliers because • they are easy to spot AI can help see the recurring patterns and clusters of behavior that are the systemic contributors to an • impaired OEE score
Why Spyglass Connected Factory? What would be valuable to know from your data? Why? • If your data was processed and summarized in a better way could you take specific actions to improve • Quality, Production, Maintenance activities? Do you know why you have Downtime? • Is your production process telling you something, but you can't understand because of all the noise within • your data? Do you want to have a better understanding of how your machines are performing or how your process is • functioning?
The Path to Better Performance Our Guaranteed Approach to Your Personal Virtual Production Manager Define Success Be a Change Agent Install & Train SCF The Mariner team works Instead of fighting fires, Connect Spyglass with you to define your you can mentor your Connected Factory to your success criteria teams to ensure you production lines and teach remain competitive. You it to recognize recurring are the change agent. issues.
For more information please visit https://mariner-usa.com/ Please send detailed questions to robbie.jones@mariner-usa.com
Deploying Deep Learning For Quality Inspection Stephen Welch, Mariner
Which Images Show Defects? 2 1 3 4
Our data comes from a tricky fabric manufacturing problem
Traditional machine vision systems use a two step process to make decisions Human Machine Interface MAX_CONTRAST > THRESHOLD 1 AND DEFECT_SIZE > THRESHOLD 2 ? Diversion Gates Feature Extraction Decisioning IMAGE CAPTURE COMPUTER VISION ALGORITHM SOFTWARE Pick-and-Place Robots … COMPUTE (Integrated or Discrete)
Which Images Show Defects? 2 1 3 4
Which Images Show Defects? DEFECTIVE GOOD DEFECTIVE GOOD
TRADITIONAL MACHINE VISION IMAGE CAPTURE IMAGE CAPTURE FEATURE EXTRACTION These algorithms are typically designed once by vision system manufacturer, and “baked in” to production software. MODEL TRAINED ON YOUR DATA Deep learning model trained using labeled examples from your experts, and updated as conditions change. MAX_CONTRAST > DECISIONING THRESHOLD 1 DEEP LEARNING MODEL May consist of many tunable AND parameters, often difficult to find DEFECT_SIZE > optimal configuration, even for experts. THRESHOLD 2 ? PREDICTIONS/RESULTS PREDICTIONS/RESULTS
Alexnet Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. "Imagenet classification with deep convolutional neural networks." Advances in neural information processing systems . 2012. ResNet He, Kaiming, et al. "Deep residual learning for image recognition." Proceedings of the IEEE conference on computer vision and pattern recognition . 2016.
97.7% ResNet Accuracy Classification accuracy on held out test set 30X False Rejects Reduction 2-5X Improvement Over Manual Inspection
Recommend
More recommend
