Module 1 - Introduction to Data Science Flashcards
What’s some key words for Industry 4.0?
Cyber-physical systems
Key enabling technologies in industrial digitalization?
- Big data & AI
- Connectivity & 5G
- Additive Manufacturing
- Collaborative Automation
What’s OEE short for?
Overall Equipment Effectiveness
Describe AI
Any technique to make a machine mimic a human decision
Describe ML
Part of AI, make a machine learn a pattern, decision based on historical data.
Describe Deep Learning
Mimic the actual way of a human brain, neural networks
Mention 3 challenges in the maintenance area which can be solved with data science
- Automated production systems must work
- Even more technology to maintain
- Upgrading old production equipment
List the four dimensions of smart maintenance.
- Data driven decisions (DDD)
- Human Capital Resource (HCR)
- Internal integration (INI)
- External integration (EXI)
Which dimension of smart maintenance should you invest in?
Investment in the wrong dimension (non bottleneck) is a waste of resources
What’s impact-frequency mapping used for?
To decide what decisions which needs to be data driven. High impact (6-9 on a scale of 1-9) and high frequency (on a daily basis)
Name four analytics techniques
- Classification
- Regression
- Clustering
- Association
Name the four levels of data analytics
- Descriptive
- Diagnostic
- Predicitve
- Prescriptive
What does descriptive analysis mean?
Description of what has already happened, WHAT happened
What does diagnostic analysis mean?
Exploration of meaningful information, WHY did it happen
What does predictive analysis mean?
Prediction of future outcomes based on historical data, WHAT WILL happen
What does prescriptive analysis mean?
Recommendations on predictive model output, WHAT SHOULD we do
Describe the six steps of CRISP-DM
- Business understanding
- Data understanding
- Data preparation
- Modeling
- Evaluation
- Deployment
Describe the Business understanding step in CRISP-DM briefly.
Determine business objectives and analysis goals.
Assess situation (resources, requirements, costs).
Produce a project plan.
Describe the Data understanding step in CRISP-DM briefly.
Collect, describe and explore the data.
Verify quality in a data quality report.
Describe the Data preparation step in CRISP-DM briefly.
Select, clean, construct, integrate and format data
Describe the Modeling step in CRISP-DM briefly.
Select modeling technique, generate test design, build model and then assess model
Describe the Evaluation step in CRISP-DM briefly.
Evaluate results (with regard to success criteria), review the process and determine next steps
Describe the Deployment step in CRISP-DM briefly.
Make deployment plan, plan monitoring and maintenance, and review the project
