Demo Topic: Visual Data Discovery-Cyan Coupler
Purpose
The purpose of the demo is to provide an introduction on how to prepare, explore and visualize data to understand potential root causes of Y.
Analytical Workflow
Industry
- General Purpose
- Chemical
- Manufacturing
Questions Being Answered
- What are potential root causes to yield performance?
- What is the current state of the manufacturing process?
- What are the relationships and patterns in the data shown visually?
What is the Story/Scenario
The production of a key chemical is experiencing lower than expected yields in manufacturing. The yield has a requirement of greater than 80% to meet commercialization goals. ~15% are not meeting the target. Variable yield results in undesirable inventory. This case study explores the potential root causes of yield variability visually.
JMP features and platforms
- Excel Import Wizard
- Join Tables
- Recode
- Distribution
- Graph Builder
Discovery call, what to ask and possible pain points
- How do you visualize your data today?
- How do you identify potential root causes currently?
- Are you using Excel currently to graph your data?
- Do you find it time consuming to create graphs?
- Do you have difficulty joining datasets?
- Do you have difficulty in exploring data to get a baseline state of the data?
Included Material
- Journal
- Coupler Data.jmp (example of final dataset once joined, used in journal for script examples)
- Original Coupler Data.jmp
- Supplier Info-Coupler.xlsx
- Join.jmp (used in the journal for script examples)
Tips for delivering the demo
This dataset is another version of the tablet production dataset in the JMP sample data library originally by Lou Valente (retired). The same workflow can be used with tablet production.
Key variables that match up between the two datasets:
Yield=Dissolution
Vessel Size= Screen Size
Carbamate Amount= Mill Time
