Imagine+AI+in+Manufacturing presentation

abhinavanand4026 93 views 25 slides Aug 31, 2024
Slide 1
Slide 1 of 25
Slide 1
1
Slide 2
2
Slide 3
3
Slide 4
4
Slide 5
5
Slide 6
6
Slide 7
7
Slide 8
8
Slide 9
9
Slide 10
10
Slide 11
11
Slide 12
12
Slide 13
13
Slide 14
14
Slide 15
15
Slide 16
16
Slide 17
17
Slide 18
18
Slide 19
19
Slide 20
20
Slide 21
21
Slide 22
22
Slide 23
23
Slide 24
24
Slide 25
25

About This Presentation

AI


Slide Content

Artificial Intelligence (AI) in Manufacturing O’Connor & Trachilis

Moore’s Law 1960’s Apollo Integrated Circuit Today’s Desktop Computer CPU In 1965 Gordon Moore of Fairchild Electronics said that the number of transistors on a chip would double every two years. In 1968 he co-founded Intel which introduced the 4004 microchip in 1971. It transformed the world, enabling cell phones, calculators, pacemakers, ATMs, video games, PCs, etc.

Let’s Imagine Dick Tracy’s wrist TV from 1965 comic strip 2023 smart watch with Face-Time capability

Imagine Fresh Food William Durant dreamed of producing a home refrigerator that would be better than an ice box. In 1918 he introduced the first home refrigerator. It transformed the way we buy, store and eat food. It provides access to foods from long distances, or which are out of season or from different climates. Can you think of other innovations that have changed our world?

Imagine Running a Factory If: All the parts were good quality Production machines were always available The production system was never late Suppliers were on-time

What is AI?

All software has three basic building block algorithms: Linear, Conditional, and Loop AI software can “think” somewhat like a human. AI includes a range of technologies, including “machine learning”, or ML How does ML work? Traditional software has “explicit” instructions Some tasks need “ implicit” understanding ML uses data to decide which steps to follow It is repeatedly presented with training data and tasked to predicting the output. The program is corrected each time so that its next prediction is more accurate. Advanced ML programs have an ”artificial neural network” which is like an animal brain. Also referred to as “deep learning”. Software It starts with sample data or “training” data It learns through iteration (“training”)

Data (Manufacturing Example)

Data Collection Data from all sources must be collected Common protocol and/or conversion is required Data base is required Storage medium- best is easily accessible from multiple sites (cloud or server) Connecting devices together and collecting digital data from them is known as the “Internet of Things” or IOT Cybersecurity plan required

AI Policies

An AI system for breast cancer screening outperformed radiologists in a recent study in Nature. The technique spotted more cancers and raised fewer false alarms. It used a “deep-learning” AI software (Steeneva 2019) How does it work? It analyzes the data and recommends actions to improve outcomes

Big Data Enables AI AI needs digital data, more the better. All relevant information, even if it may not seem relevant Digital transactions are growing exponentially

Big Data Enables AI

AI is Transforming Manufacturing Key Areas Predictive Maintenance Quality Manufacturing Process Optimization Supply Chain Optimization

Predictive Maintenance On-condition vs scheduled Data collected: temperature, current, RPM, environmental parameters, vibration, time, hours, work order numbers. calibration, tool wear, etc. Reduced: downtime, production delays, inventory, maintenance cost, unproductive labor, late deliveries Sensors Identify machine parameter reaching threshold Determine cause and alert GE estimates it has reduced machine down time by 30% through using AI based predictive maintenance

Quality AI for non-destructive inspection (image processing): Visual, Liquid penetrant Radiographic Ultrasonic AI for testing products with long schedules (e.g. electronic assemblies) AI for Quality Management: “statistical process control on steroids” . Data collected: machine health quality records, environment Process parameters Reduced inspector fatigue Automated inspection Better defect detection, less false indications Faster, more repeatable Analyze effect of multiple variables, and their interaction Identify increased variability, shift in mean, trend toward spec limits Determines root cause and sends alerts Reduced risk by detecting defects early Improved yield, reduced scrap, less downtime, less inventory, lower cost, better customer satisfaction

Manufacturing Process Optimization AI for production management: Decides what steps to take vs MRP explicit instructions Uses any relevant historical and real-time data : lead times, run and set up times, batch size, part number, quality records, routings, maintenance data, efficiency, operator performance, environment parameters Advises when to launch orders based on forecast completion Improved forecasting Shorter lead times Better demand forecasts Reduced inventory Reduced part obsolescence Level load Improved resource utilization AI based production management can improve forecasting by 20%, reduce inventory by 5%, and reduce planner workload by 50% - A. McKinsey

Supply Chain Optimization AI for supply chain optimization: Like manufacturing process optimization Uses additional data: Transportation logistics Supplier delivery and quality performance Political and environmental factors Supplier internal data (where available) Better consumption forecasts Reduced inventory Reduced part obsolescence Fewer stock-outs Reduced shipping dwells Reduced lead time Improved on-time delivery

Manufacturing is Off to a Slow Start With AI Retail, travel, social media, health care and banking were early adopters- largely aided by extensive digital data Manufacturing is growing, led by GE, Samsung, Intel, Microsoft, and others Lack of digital data for manufacturing process is a disadvantage Investments in digital connectivity in manufacturing are forecasted to grow from $267 Billion in 2020 to $10.1 Trillion in 2025! - Kerr and Polano

Decide your requirements Get educated Find an expert Talk to other companies Develop a road map Do a pilot project Tell your people where you are headed How to Get Started

Summary www. TheAIengineers.com

What is AI?

AI is Transforming Manufacturing Predictive Maintenance, Quality, Manufacturing Process Optimization, Supply Chain Optimization Key Areas

Visit www.TheAIengineers.com to connect with us Education – Dave’s course www.umanitoba.ca/extended-education/programs-and-courses/courses/advanced-manufacturing Need Help?

Thank you! www. TheAIengineers.com