Rainbow SPC Process: Using Statistical Tools for Accelerated Product Development and Enhanced Reliability
By Thimmiah Gurunatha
Traditional approaches to statistical process control (SPC) charting are effective when it comes to monitoring process behaviors and providing useful data for continuous improvement efforts. However, many of the same tools typically reserved for inspection and basic quality control can have much broader applications for enhanced quality and reliability.
I created the Rainbow SPC approach in 1990 while working at Xerox and continued to develop it in different environments and problem-solving applications until I retired in 2008. The approach offers a road map for quality professionals to follow in applying tools that may already be familiar to them in other contexts. The key is to start thinking of those tools earlier than the inspection stage. When treated as part of product design and development processes, they can have an impact on the bottom line in terms of improved reliability, an accelerated design process, and reduced costs of poor quality.
Learn more about applying the Rainbow SPC approach. View these Webinars hosted by
How to Use the Road Map
The tools referenced in the road map represent a collection of classical statistical tools complimented by newer tools developed by experts around the world. It is not the purpose of the road map to provide instruction in the application of the tools. Rather, the intention is to show where they fit in the research and product/process development process.
Page 1 shows the overall flow of the product development process, from gathering customer requirements--including needs, wants, and delights--to the launch of the product.
Page 2 provides help with problem solving during product or process development. If you encounter problems, the tools on this page can help you identify and solve them with speed. This page also shows tools for sustaining the gains in manufacturing applications and for reducing or eliminating early failures.
The table on pages 3 and 4 presents a listing of tools typically associated with reliability and maps them to the phases of product development.
The table on pages 5 and 6 presents a listing of statistical techniques and tools used for problem-solving, providing details on objectives they can help achieve, where and when they can be applied, and sample sizes associated with them.
New Tools for Enhanced Reliability
Since the road map incorporates tools that are widely known to those in quality roles, it can be used immediately, applied to current urgent needs, with little training required. Just a couple of the tools referenced may be unfamiliar to readers.
The Rainbow Reliability Chart
This is an SPC chart for continuously monitoring reliability. In the figure below, the green zone represents expected performance. The yellow, orange, and red zones correspond to performance that is 1, 2, and 3 sigma worse than expected. The blue, indigo, and violet zones are 1, 2, and 3 sigma better than expected. If all of your data points fall in the orange zone, you can be 95% sure that reliability is worse than expected, and if they all fall in indigo or better, you can be 95% confident that reliability is better than expected.
Why “Rainbow”? Ultimately, what I wanted to achieve was a simple, visible, reliable, and responsive process.
Rainbow stands for…
- Responsible staff
- Activities focused
- Implementation of problem solving
- Natural analysis
- Better work environment
- Optimization of products and services
- Waste reduced, efficiency increased, work accomplishment maximized
Thunder and Lightning Accelerated Test (TALAT)
I developed this tool for understanding and optimizing product reliability in order to enable an accelerated product delivery schedule. It integrates the best practices of commonly known accelerated testing methods while adding several features to further increase the value of these tests:
- First, in order to tap information about the percentage contribution of stresses and enhance understanding of the operating rectangle portion of the test, TALAT simulates a full factorial matrix for varying stresses instead of randomly varying the stresses as in multiple environment overstress testing (MEOST) (1).
- Second, TALAT calls for selecting samples for the test at the extreme and nominal values of the specifications to understand the robustness of the design. Of twelve samples, one represents the positive extreme of the critical parameters, one represents the negative extreme, one is at the nominal value of the specification, and nine fall in between.
- Third, TALAT offers a process which can condense test time even further.
If you would like to adapt or customize the Rainbow approach for your own problem-solving and product development challenges, or if you have an application experience to share, contact the ASQ Knowledge Center.
1. Kreucher, John, “Multiple Environment Over Stress Testing (MEOST): A Primer with Examples from an Automotive Component Supplier,” ASQ Automotive Division, October 26, 2011. Recorded Webinar available at http://asq.org/auto/106944/web.html?shl=106944. Presentation slides (PDF) available for download at www.asq-auto.org/s/MEOST-Webinar.pdf.
About the Author
Thimmiah Gurunatha is author of Systems Engineering Standards—The State of the Art: Integrating DFR, DFSS and DFX in a Systems Engineering Environment. Before retiring in 2008, he held senior engineering roles for Xerox for 32 years. With a career focus on quality and reliability—including design reliability, highly accelerated life testing (HALT), highly accelerated stress screening (HASS), accelerated testing, and Shainin techniques—he has consulted in problem solving, testing, and problem prevention for corporations in countries around the world. He has taught courses for ASQ; Sheridan College Institute of Technology and Advanced Learning in Toronto; and the Taipai, Taiwan Technical Centre; as well as for Xerox staff, earning a Xerox coach of the year award. An ASQ Fellow, Gurunatha holds a bachelor’s degree in mechanical engineering and a master’s in industrial engineering and operations research, and he earned ASQ’s Reliability Engineer and Six Sigma Black Belt certifications.