OptiMal-DOE
Sample DOE's
OptiMal by the Taguchi Methodology is the most effective and cheapest way to optimize complex problems being influenced by many factors simultaneously.
The Method ensures that you avoid the primitive an ineffective one at a time approach.
The design of experiments can be used in almost any industrial
branch.
Some of examples are part of the demonstrations included with
the program.
1. Chemical, pilot plant examination. L8A
model
Factors:
- Catalyst
- Concentration
- Temperature
The response is measured as the yield of the process.
Strategy: The larger the better.
2. Car experiment. L4 model
Find the car reaching 0-400 yards before the other.
Factors:
Strategy: The smaller the better.
3. Heat seal assembly. L9
design.
Factors:
- Temperature
- Pressure
- Holding time.
Strategy: The bigger the better
4. Plastic foundry. L18A
design Xerox.
Factors:
- Inlet pressure
- Cycle time
- Temperature of the form
- Melting temperature
- Counter pressure
- Cooling time
- Settling time.
Strategy: Nominal is best.
5. Clutch performance in the
car industry. L9 design.
Factors:
- Compression
- Angle error
- Linear error
- Supplier
Strategy: The larger the better.
6. Speedometer cable. Quinlan.
L16A design.
Factors:
- Kappe ydre diameter
- Kappe pressestempel
- Kappe materiale
- Kappe fremføringshastighed
- Skærmtråd type
- Skærmtråd spænding
- Tråd diameter
- Kappe spænding
- Kappe temperatur
- Yderkappe materiale
- Yderkappe stempel type
- Smelte temperatur
- Skærm pakning
- Kølemetode
- Fremføringshastighed
Response: Shrinkage of the cable.
Strategy: The smaller the better.
7. Pizza production, recipe,
L18A design.
Factors:
- Temperature
- Pressure
- Baking time
- Mixing time
- Number of revolutions of the mix
Response: The best yield (taste is the measure),
Strategy: The bigger the better.
8. Nuclear power plant, material refinement, Design:
L16A
Factors:
- Water/Cement ratio
- PFA/OPC ratio
- Water temperature (C)
- Powder feed rate (Kg/min)
- Powder addition Sequence
- Mixing time
- Paddle speed (rpm)
- Grout recirculation rate (L/min)
Response: Grout fluidity, Strategy: The bigger the better.
9. Automobile driven plate and
"Torsional Hysteresis". An engineering problem.
- Dimensional characteristics
- Flatness
- Squareness
- Surface condition
- Frictional characteristics
- Environmental characteristics
Response: Reduction of variability, Strategy: Nominal is
best.
A reduction of 50% of the variability was achieved at Automotive
Products PLC.
L16A design.
Factors or characteristics:
10. Automotive brakes, foundry
optimization at Ford Leamington, Design: L12.
Factors:
- Catalyst age
- Pins/Bushings
- Blow head
- Cope temperature
- Cure time
- Mixing method
- Blow pressure
- Blow chamber level
- Core box cleanliness
- Spray mount
Response: Scabbing. The scrap factor was drastically reduced below
0,5%. The brakes are of the types with internal ribs to force
cooled air to the plates.
11. Automotive seam welding
Ford. L16A design.
Factors:
- Weld time
- Cool time
- Weld current
- Cradle speed
- Wheel condition
- Air pressure
Response: Tensile strength; Strategy: The bigger the better.
12. Plastic moulding of
buttons for a telephone at Bang & Olufsen a/s. L16 A
design.
Factors:
- Cycle time
- Inlet temperature
- Cooling time
- Cooling temperature in the form, front part
- Rear part
- Counterpressure
Response: Two: Surface quality and nominal diameter; Strategy: The
smaller the better and Nominal is best.
13. Plastic cover for a
gramophones. At Bang & Olufsen. Design L8A.
Factors:
- Temperature
- Preheating
- Cooling time
Response: Nominal size; Strategy: Nominal is best.
14. Optimization of the
soldering process of printed circuit boards. L36
design.
Factors:
- Conveyor speed
- Preheating above
- Preheating under
- Solder wave angle
- Pump speed
- solder paste flow
- and many more
15. Optimization of an
aluminum foundry process at JAI.
16. Welding rivets to aluminum
process at Mekoprint.
17. Coating plastic parts at
Bang & Olufsen.
18. Printed wiring board
Optimization of the lamination process at Bull.
19. Tile manufacturing
process. Reduction of the noise from the spread in the temperature
from tiles placed in the middle and at the outer limits. L8
Design.
20. SMD assembly line process,
Kirk Electric. L12 design.
Factors:
- Sucking force
- pressure of pickup 1
- Pressure of pick-up 2
- counterpressure
- Glue pressure
Response: Placement accuracy; Strategy: the smaller the better.
21. Zink coating of nails at
NKT. L27A design.
Factors:
- Concentration of the zink
- Sodiumhydroxide
- Sodium cyanide
- Sodium carbonate
- revolutions (rpm)
- Current (Amps)
Response: coating size; Strategy: The bigger the better.
22. Thick film printing at
Hytek. L16 design.
Factors:
- Emulsion
- snap-off
- printing speed
- Rakler speed
- Rakler pressure
Bicycle Experiment
In experiment number 1 the student, Norman Miller, using a
factorial design with all points replicated, studied the effects of
three variables-seat height (26, 30 inches), light generator (on or
off), and tire pressure (40, 55 psi)-on two responses-time required
to ride his bicycle over a particular course and his pulse rate at
the finish of each run (pulse rate at the start was virtually
constant).
To him the most surprising result was how much he was slowed down
by having the generator on.
The average time for each run was approximately 50 seconds.
He discovered that raising the seat reduced the time by about 10
seconds, having the generator on increased it by about one-third
that amount and inflating the tires to 55 psi reduced the time by
about the same amount that the generator increased it. He planned
further experiments.