Product & Scope - Automotive Power Window Gear Housing. Study molding process to determine causes of voids.
Objective - Produce parts that are completely filled. Reduce rejects. QC = Smaller is Better.

Process Factors and Interactions - Four 2-level factors and One 3-level factor: A: Shot Size(90 mm, 95 mm), B: Material Temperature(88 Deg C, 102 Deg C), C: Mold Temperature(170 Deg C, 185 Deg C), D: Injection Pressure(40 kgf/cm, 130 kgf/cm), and E: Injection Speed(15 mm/sec, 20 mm/sc, 25 mm/sec). Noise Factors - Not included (random noise)

Design Strategy - Use an L-8 OA. Assign factors A, B, C, and D to columns 4, 5, 6, and 7 respectively. Modify column 1 into a 3-level column by first upgrading columns 1, 2, and 3 into a 4-level column (col. 1), then downgrading this column to a 3-level column. Assign the 3-level factor, E: Injection speed, to column 1.

Reduction of Maxi-Blast Cycle Time (Stripping of Nylon Coated Brake Lines)

Product & Scope - In the stripping process using Maxi-Blaster, nylon coated brake lines are inserted into the cleaning cabinet. Stripping time is recorded when the coating is completely off the end of the tube.
Objective - Reduce time of stripping. QC = Smaller is Better.

Process Factors and Interactions - Six 2-level factors and One Interaction: A:Pressure(40 psi, 100 psi), B: Nozzle Diameter(5/16 inch, 3/8 inch), C: Nozzle Angle(45 degrees, 90 Degrees), D:Maintenance(2 hours, 4 hours), E: Flow Rate (Open, Closed), F: Material Flow(2 seconds, 6 seconds), and interaction BxC Noise Factors - Not included (random noise)

Design Strategy - Use an L-8 OA. Assign factors B, C, D, E, F and A to columns 1, 2, 4, 5, 6, and 7 respectively. Reserve column 3 for interaction between factors B and C.

Increasing the Crush Strength of Quick Connector Housing

Product & Scope - The Quick Connector is assembled by crimping a length of P-TEC onto the stem. Generally about 5% of the production parts have been found to crack while crimping. This experiment was designed to optimize the injection molding process parameters for the Quick Connector housing which produce increased crush strength.
Objective -Increase the compressive force (under tensile testing machine) required to crush the stem. QC = Bigger is Better.

Process Factors and Interactions - Four 3-level factors were studied: A: Mold Temperature(180 Deg C, 190 Deg C, 200 Deg C), B: Hold Time(3 Sec, 9 Sec), C: Cooling Time(25 Sec, 45 Sec), and D: Packing Pressure(20% Maximum, 40 % Maximum).
Noise Factors - Not included (random noise)

Design Strategy - Use an L-9 OA. Assign the four 3-level factors A, B, C, and D to columns 1, 2, 3 and 4 of L-9 array respectively.

Magnet Adhesion and Placement Process Study

Product & Scope - Automotive power window motor.. Study assembly process parameters.
Objective - Increase impact strength measured in terms of drop height. QC = Bigger is Better.

Process Factors and Interactions - Eight 2-level factors and three interactions were studied: A: Cure Temperature(100 Deg C, 150 Deg C), B: Cure Time(1 minute, 3 minutes), C: Quantity of Glue(0.1 gm, 1.5 gm), D:Magnitism(3000Max, 4000 Max), E:Viscosity(40,000 cps, 50,000 cps), F: Surface Cleanliness (normal, cleaned), G: Tooling Pressure(2 kgf, 4.2 kgf), H: Index Motion(20 sfm, 50 sfm), interactions AxB, CxG, and DxH. Noise Factors - Not studied (kept random)

Design Strategy - Use an L-16 OA. Assign factors A, B, C, D, E, F, G and H to columns 1, 2, 4, 5, 6, 7, 8, and 10 respectively. Reserve column 3 for AxB, column 12 for CxG, and column 15 for DxH. Keep columns 9, 11, 13 and 14 as unused.

Optimizing Jet Ski performance Through Bolt-On Component Design

Product & Scope - When purchasing Jet Ski, customers usually order modification packages to increase the performance.
Objective -The objective of the study was to evaluate the performance in terms of increased acceleration for various add-on component alternatives. For evaluations, the Jet Ski with modifications was tested through a course of known distance with body-mounted speed traps at the start, middle and end, used in conjunction with a radar gun that determined the top speed and the acceleration. QC = Bigger is Better.

Process Factors and Interactions -Six 2-level factors and One interaction were studied: A: Carburetors (Mod Stock, 44mm Aft.Mkt.), B: Exhaust (Short, Long), C: Impeller (Skat Trak, Solas),D: Intake Gate (Open, Top Loader), E: Timing Plate (Stock, Aftermarket), F: Cylinder Heads (ADA, RPM), and Interaction AxB.
The carburetors and exhausts are similar to most two-stroke motors, but the other parts are not. The impeller is a propeller in the housing. The intake grate covers the area in which the water enters the pump to prevent debris from entering the pump. The timing plate advances the timing by rotating the coils beyond factory setting. Finally the cylinder has no valves and controls the compression based on volume inside the dome of the head. All these items are easy to remove and replace which makes it easy for the customer to alter the performance.
Noise Factors -Test location and Assembly method are considered noise factors (kept random)

Design Strategy - Use an L-8 OA. Assign factors A, B, C, D, E, and F to columns 1, 2, 4, 5, 6, and 7 respectively. Reserve column 3 for interaction between factors A and B (AxB).

Automobile Operating Cost Study

Product & Scope - A group of administrators responsible for maintenance of test vehicles for a large automobile manufacturers group undertook a study to determine the most economical operating condition for its fleet. Fuel economy(60% relative weighting) and maintenance cost(40% relative weighting) were considered as the prime criteria of evaluations.
Objective -The objective of the study was to determine the operating condition that gives the most gas mileage with least maintenance cost. An Overall Evaluation Criteria (OEC) was defined for analysis purposes.

Operational Factors and Interactions - Four 3-letter factors were selected for the study: A: Type of Tire (Economy, Mid-Price, and Expensive), B: Type of Gasoline(87 Octane, 89 Octane, and 92 Octane), C: Inflation Pressure(5psi below, Rated, 5psi above), and D: Engine Oil( Standard 10W30, Lower Grade, and Standard GT). Interactions between factors were assumed to be negligible.

Noise Factors -Influence of uncontrollable factors was considered absent..

Design Strategy - Use an L-9 array. Assign factors A, B, C, and D to the four columns of the array.

References

• Text: A primer on the Taguchi Method by R. Roy
• Training: 3-day seminar in Detroit area by Nutek, Inc.
• Software: Qualitek-4 for Design and Analysis of Taguchi Experiments
• Special Topics: Overview of DOE/Taguchi