Example

The Flaw of One-Factor-At-A-Time (OFAT) Testing

A materials engineer wants to optimize the compressive strength of a new concrete mix by adjusting the Water/Cement (W/CW/C) ratio and the Curing Temperature. Using the traditional OFAT approach:
  1. They fix the temperature at 20C20^\circ\text{C} and test W/CW/C ratios of 0.40.4, 0.50.5, and 0.60.6. They find 0.40.4 gives the highest strength.
  2. They fix the W/CW/C ratio at 0.40.4 and test temperatures of 10C10^\circ\text{C}, 20C20^\circ\text{C}, and 30C30^\circ\text{C}. They find 20C20^\circ\text{C} gives the highest strength. They conclude the absolute optimal mix is W/C=0.4W/C = 0.4 cured at 20C20^\circ\text{C}. Why might this conclusion be completely wrong?

Step-by-Step Solution

0 of 3 Steps Completed
1

Example

Introduction to DoE: Fractional Factorial Design

An environmental engineer is designing a new wastewater filtration system. They identify 7 potential factors that might affect the filtration rate (e.g., sand type, flow rate, backwash frequency, temperature, pH, coagulant dosage, influent turbidity). Testing every possible combination at 2 levels (272^7) would require 128128 experiments. The project budget only allows for 20 tests. What DoE strategy should they employ?

Step-by-Step Solution

0 of 3 Steps Completed
1

Example

Setting Up a 222^2 Full Factorial Design Matrix

A geotechnical engineer wants to investigate the shear strength of a compacted clay soil. They select two factors: Factor A: Compaction Effort (Low = Standard Proctor, High = Modified Proctor) Factor B: Moisture Content (Low = 10%10\%, High = 15%15\%) Set up the standard design matrix for this 222^2 full factorial experiment using standard 1-1 and +1+1 notation.

Step-by-Step Solution

0 of 2 Steps Completed
1

Example

Calculating Main Effects in a 222^2 Design

Using the design matrix from the previous example, the engineer conducts the 44 tests and records the following shear strengths (yy, in kPa): Run 1 (1,1-1, -1): y1=50 kPay_1 = 50\text{ kPa} Run 2 (+1,1+1, -1): y2=80 kPay_2 = 80\text{ kPa} Run 3 (1,+1-1, +1): y3=40 kPay_3 = 40\text{ kPa} Run 4 (+1,+1+1, +1): y4=60 kPay_4 = 60\text{ kPa} Calculate the Main Effect of Factor A (Compaction Effort) and Factor B (Moisture Content).

Step-by-Step Solution

0 of 2 Steps Completed
1

Example

Calculating the Interaction Effect (ABAB) in a 222^2 Design

Continuing with the soil compaction data: Run 1 (1,1-1, -1): y1=50 kPay_1 = 50\text{ kPa} Run 2 (+1,1+1, -1): y2=80 kPay_2 = 80\text{ kPa} Run 3 (1,+1-1, +1): y3=40 kPay_3 = 40\text{ kPa} Run 4 (+1,+1+1, +1): y4=60 kPay_4 = 60\text{ kPa} Calculate the Interaction Effect between Compaction Effort and Moisture Content (EABE_{AB}) and interpret the result.

Step-by-Step Solution

0 of 3 Steps Completed
1

Example

Advanced Optimization: Response Surface Methodology (RSM)

A civil engineer has completed a 232^3 full factorial screening experiment on a new geopolymer concrete mix. They found that all three factors (Fly Ash content, Activator Molarity, and Curing Time) significantly affect compressive strength, and strong interaction effects are present. They now want to find the exact combination of these three variables that yields the absolute maximum strength. Why is RSM the logical next step, and how does it differ from the factorial design?

Step-by-Step Solution

0 of 3 Steps Completed
1

Example

Advanced Optimization: Taguchi Methods and Robust Design

An automotive manufacturer is designing a new suspension system for off-road vehicles. Traditional DoE might focus solely on finding the shock absorber settings that provide the smoothest ride on a test track. How would the application of Genichi Taguchi's philosophy change the goal of this experimental design?

Step-by-Step Solution

0 of 3 Steps Completed
1