Lean Six Sigma is a methodology that combines the waste-reduction principles of Lean manufacturing with the process improvement strategies of Six Sigma. One of the advanced techniques used in this approach is Fractional Factorial Experiments, a subset of design of experiments (DOE) that allows project teams to study the effect of multiple variables on a process with a reduced number of experiments. This article delves into real-world applications of Lean Six Sigma projects, particularly focusing on how Fractional Factorial Experiments play a pivotal role in achieving significant improvements.

Efficiency in Manufacturing

A common application of Lean Six Sigma in the manufacturing sector involves optimizing production processes to reduce waste and improve quality. In one case, a manufacturing plant facing high defect rates and significant variability in its production process employed Fractional Factorial Experiments to identify the critical factors affecting product quality. By conducting a reduced set of experiments, the team was able to quickly pinpoint a combination of temperature settings, machine speeds, and material feeds that consistently produced high-quality products. Implementing these settings across their production lines led to a significant reduction in defects and an increase in overall efficiency.

Healthcare Process Improvement

Lean Six Sigma is not limited to manufacturing and finds extensive application in healthcare, aiming to improve patient care and operational efficiency. In a hospital setting, Fractional Factorial Experiments were used to address the challenge of long patient wait times in the emergency department (ED). The project team selected several factors for investigation, including staffing levels, triage processes, and patient flow. Through a series of carefully designed experiments, they identified the most impactful changes that could reduce wait times without compromising patient care. As a result, the hospital was able to implement a more efficient triage system and adjust staffing schedules, leading to improved patient satisfaction and reduced congestion in the ED.

Service Industry Optimization

In the service industry, Lean Six Sigma projects often focus on improving customer satisfaction and operational efficiency. A retail banking branch used Fractional Factorial Experiments to enhance its customer service process. The bank identified several factors that could potentially affect customer waiting time and service quality, including teller training, queue management systems, and transaction processing time. By applying Fractional Factorial Experiments, the bank discovered the optimal combination of these factors that minimized waiting times and improved customer service ratings. Implementing these findings helped the branch achieve higher customer satisfaction scores and increased efficiency in transaction processing.

Product Development and Innovation

Lean Six Sigma also supports product development and innovation processes. A technology company looking to develop a new consumer electronics product utilized Fractional Factorial Experiments to streamline its design and testing phases. The company aimed to identify which features and design elements were most important to consumers and how these could be optimized for performance and cost. Through a reduced number of experiments, the team was able to focus their development efforts on the features that offered the greatest value to customers, leading to the successful launch of a highly competitive product.

Conclusion

Fractional Factorial Experiments within Lean Six Sigma projects provide a powerful tool for identifying the most influential factors in complex processes across a variety of industries. By enabling teams to conduct a reduced number of experiments while still gaining valuable insights, this approach significantly accelerates the improvement cycle. Real-world applications in manufacturing, healthcare, service industries, and product development highlight the versatility and effectiveness of Lean Six Sigma methodologies in driving substantial improvements in efficiency, quality, customer satisfaction, and innovation. As organizations continue to face the pressures of competition and the demand for higher quality at lower costs, the principles of Lean Six Sigma, coupled with the strategic use of Fractional Factorial Experiments, will remain essential to achieving operational excellence and sustainable growth.

Full real-life based scenario

Let's consider a real-life scenario from a coffee shop chain looking to improve customer satisfaction and operational efficiency. The management decides to focus on reducing the time it takes to serve a coffee while maintaining or improving the quality of the coffee served. They suspect several factors could influence both the service time and coffee quality:

1. Coffee Bean Type (A): Arabica vs. Robusta

2. Grind Size (B): Fine vs. Coarse

3. Water Temperature (C): High vs. Low

4. Tamping Pressure (D): Firm vs. Light

   
   

Conducting a full factorial experiment would require testing all possible combinations of these factors, resulting in 2^4 = 16 experiments. However, to save time and resources, they opt for a fractional factorial design, specifically a 1/2 fraction, reducing the number of experiments to 8.

They decide to focus on the main effects and interaction effects between the coffee bean type and grind size, assuming these have the most significant impact on both quality and service time. The reduced set of experiments is planned as follows:

1. Arabica, Fine, High, Firm

2. Arabica, Coarse, Low, Light

3. Robusta, Fine, Low, Light

4. Robusta, Coarse, High, Firm

5. Arabica, Fine, Low, Firm

6. Arabica, Coarse, High, Light

7. Robusta, Fine, High, Light

8. Robusta, Coarse, Low, Firm

   
   

For each experiment, they measure two outcomes:

• Service Time (Y1): Time to serve a coffee (in minutes)

• Coffee Quality (Y2): Rated on a scale of 1-10 by a panel of customers

Let's simulate the outcomes for these experiments and plot the results to identify patterns or trends that could inform the coffee shop chain's decision-making.

We'll create a simulation of these results in Python, and then we'll plot the main effects and any notable interaction effects on the service time and coffee quality.

The charts above illustrate the simulated impact of coffee bean type and grind size on service time and coffee quality in the fractional factorial experiment.

• Impact on Service Time: The average service time is slightly less for coarse grinds compared to fine grinds, suggesting that a coarse grind might facilitate faster service. The bean type does not show a significant difference in service time, indicating that the choice between Arabica and Robusta beans can be made based on taste preference or cost without greatly affecting service speed.

  
  

• Impact on Coffee Quality: Coffee quality ratings are higher for Robusta beans in this simulation, which might be due to their stronger flavor profile. Additionally, the fine grind size appears to slightly improve the coffee quality rating, possibly because of a better extraction process.

  
  

These findings could lead the coffee shop chain to experiment with Robusta beans and fine grinds to enhance coffee quality while also exploring ways to optimize the service process for coarse grinds to reduce service time. This example demonstrates how fractional factorial experiments can provide actionable insights with fewer trials, saving resources while still guiding significant improvements in both product quality and operational efficiency.