SLIDE 1
Process Moving our SMEs towards Industry 4.0 RPC Fredericton, - - PowerPoint PPT Presentation
Process Moving our SMEs towards Industry 4.0 RPC Fredericton, - - PowerPoint PPT Presentation
Connecting Your Process Moving our SMEs towards Industry 4.0 RPC Fredericton, Moncton & St. George, NB How does Atlantic Canada stack up? Demographic Bulge Peters, Paul A. (2017). New Brunswick Population Snapshot (Report No. 2017-01).
SLIDE 2
SLIDE 3
How does Atlantic Canada stack up?
SLIDE 4
Demographic Bulge
Peters, Paul A. (2017). New Brunswick Population Snapshot (Report No. 2017-01). Fredericton, NB: New Brunswick Institute for Research, Data and Training (NB-IRDT).
SLIDE 5
SLIDE 6
Hype vs Reality
- Greenfields vs brownfield
retrofits
- Knowing your process
- Find partners
- Local ecosystem growing
- Funding is available
SLIDE 7
Where do you start?
- Process Mapping
and Value Streams
- Pain Points and
Bottlenecks
- “Lean”ing
- Reduce paper and
manual entry (especially any doubling)
SLIDE 8
Data
- Data is king!
– But only when properly scrubbed and selected – Don’t drown in the flood
Data Insight Action
SLIDE 9
Automation
- Traditional Automation
– PLCs and machines
- Physical Robots
– “Arms”
- Digital Twins
- Robotic Process
Automation
SLIDE 10
Example # 1 – JDI Automated Somatic Embryo Processing
SLIDE 11
Example # 1 – JDI Automated Somatic Embryo Processing
- Fully automated workcell.
- Workcell automates previously manual laboratory steps
- Need to collect processing information from each step
- Information can be in the form of step process times,
temperatures, water levels, images, robot positions, machine throughput, final part count.
- How do we connect to this process ?
SLIDE 12
Example # 1 – JDI Automated Somatic Embryo Processing
SLIDE 13
Example # 1 – JDI Automated Somatic Embryo Processing
SLIDE 14
Example # 2 – SomaDetect Sensor Device
- Real time somatic cell milk quality sensor
- Real time sensor replaces once per month lab test and can
identify the health of each cow during every milking.
- Each sensor collects complex light scatter patterns generated
by somatic cells and fat, and uses machine learning and computer vision techniques to decipher these patterns
- Currently installed in 250 locations in Canada and the US and
generating data from 5,000 cows.
- How do we connect to this process ?
SLIDE 15
Example # 2 – SomaDetect Sensor Device
SLIDE 16
Example # 2 – SomaDetect Sensor Device
SLIDE 17
Example # 2 – SomaDetect Sensor Device
SLIDE 18
18 /
SLIDE 19
19 /
Des appareils connectés
SLIDE 20
SLIDE 21
21 /
Machine Learning: Key Ideas and Applications
- At its core, machine learning uses mathematical methods to find patterns in data.
- Can be applied to nearly all types of data.
- Despite being an old field, recent advances in computational technology has allowed
for massive growth.
- Some consider ML to be a key player in “The Next Industrial Revolution”.
SLIDE 22
SLIDE 23
23 /
SLIDE 24
24 /
SLIDE 25
25 /
Example: Computer Vision Using Convolutional Neural Networks
SLIDE 26
26 /
A List of Other Applications
SLIDE 27
27 /