Introduction to Design for Six Sigma
History of Six Sigma
Six Sigma is a business management strategy, developed by Motorola in 1986. Six Sigma provides methodology and tools to improve processes by removing defects and minimizing variability in the processes. When Motorola first deployed Six Sigma, the focus was on improving manufacturing processes. Motorola applied the methodology called “Six Steps to Six Sigma”. In 1995, Six Sigma became well known when GE made Six Sigma a central focus of its business strategy. Other companies began to apply Six Sigma to their transactional processes (service) also. In the 1990’s Six Sigma methodology evolved, called DMAIC (Define, Measure, Analyze, Improve, Control). Since then DMAIC has become the standard methodology to improve existing business processes and products.
Although, DMAIC provided billions of dollars of savings to companies, it was considered “fire fighting” since it was applied to processes already in place. A proactive methodology was needed that helped to achieve Six Sigma capability before products or processes were deployed. Some companies applied DMAIC to design new products and processes. However, it lacked tools to analyze customer requirements, to predict variances, statistical tolerances, and reliability.
What is Design for Six Sigma (DFSS)?
It is a methodology for designing new processes and products or redesigning existing processes and products. It implements Six Sigma early in a process or product life cycle. Similar to DMAIC, DFSS activities are carried out by the cross functional team. DFSS begins with focus on customer needs. All activities are carried out with the customer in mind.
Typically we make minor or major change(s) to a process or product in order to improve performance. Figure 1 compares the cost of implementing a change in early vs later, in a process or product life cycle.
Figure 1 indicates that the cost of implementing a change during production phase is 100 times more than during the concept/design phase. As shown in the graph, the focus of DFSS is during the early stages of process or product life cycle. DMAIC, on the other hand, focuses on the later stages of the life cycle-typically the production phase.
As mentioned before, DMAIC is applied to the existing process or product. Since the basic design features cannot be changed easily, the pace and level of improvement that we can attain in existing processes or products is limited. On the other hand, by applying DFSS to designing or redesigning process or product, we can fundamentally change the process or product and achieve a leap frog improvement. Figure 2 shows improvements that can be attained using DMAIC and DFSS.
Goal of DFSS
Create compelling processes or products that meet/exceed customer expectations in terms of quality, reliability, cost, and cycle time. DFSS provides tools and methodology to create robust processes and products. Companies can bring new processes and products into the market ahead of competition. It increases customer satisfaction and helps generate higher revenue and profits.
DFSS Methodologies (Roadmaps)
DMAIC methodology is universally adopted by companies that use Six Sigma business management strategy to improve existing processes or products. For DFSS, there are several methodologies that are used by different companies. Some companies have modified the original DFSS methodologies to provide focus on certain phases and/or deliverables. Although these methodologies have different acronyms, the objectives and goals are similar. Below are the most common DFSS methodologies used by companies:
1. DMADV - Design, Measure, Analyze, Design, Verify
2. DMEDI – Design, Measure, Explore, Develop, Implement
3. CDOV – Concept, Design, Optimize, Verify
4. IDOV – Identify, Design, Optimize, Verify (Validate)
Since these above methodologies are similar in nature, we will describe CDOV briefly.
Figure 3 shows the use of DMAIC and CDOV methodologies. Typically, when an existing process or product does not meet customer or business expectations, we improve it by applying DMAIC. After going through Analyze or Improve phases, if we realize we cannot meet the improvement goals or it is not cost effective to implement the solutions to meet these goals, we pursue CDOV (DFSS).
One of the most important characteristics of DFSS methodology is gate reviews. After completion of each phase (C, D, O, and V), the DFSS and leadership teams conduct the gate review. The objective of the gate review is to evaluate and determine if it is worth continuing with the project. Several factors are considered including market dynamics, customer expectations, and financial goals during the gate review.
Below is the list of activities performed, and tools used in each phase of CDOV. It is not intended to provide a complete list since the activities and use of tools vary by the project.
· Identify customers and gather VOC
· Analyze VOC and prioritize customer requirements
· Map customer requirements to functional requirements
· Identify and mitigate risks
Tools: Voice of the Customer Gathering and Analysis, Kano Analysis, KJ Analysis, House of Quality, Design FMEA
· Generate concepts and select superior concept
· Map functional parameters to design parameters, and design parameters to manufacturing parameters
· Capability flow-down (sub-system, assembly, parts, manufacturing requirements)
· Perform Measurement System Analysis (MSA)
· Develop transfer functions
Tools: Brainstorming, Mind Mapping, TRIZ, Pugh Matrix, Measurement System Analysis, Hypothesis Testing, ANOVA Regression, Design of Experiments
· Identify noise factors and perform robust design
· Estimate variance of Ys. Perform sensitivity analysis.
· , , Perform statistical tolerance analysis
· Optimize parameters at all levels
· Design for manufacturability, assembly, reliability
Tools: Robust Design, Response Surface Methodology, Multiple Response Optimization, Monte Carlo Simulation, Reliability Modeling
· Build Prototypes
· Verify capability of functional parameters
· Perform Accelerated life testing
· Implement Poke Yoke
· Map manufacturing parameters to process control parameters
· Implement SPC
· Develop Control Plans
Tools: Reliability Prediction, Capability Analysis, Statistical Process Control, Control Plans
We have provided an introduction to DFSS. It is a business management strategy and its success depends on leadership, commitment and involvement. Leaders set the stretch goals, provide support, and hold the DFSS team accountable for success of the project.