It’s Elementary

Following lines of evidence to reveal a root cause

by Matthew Barsalou

A root cause investigation doesn’t always result in a clear path to the root cause. For example, there may be insufficient information available or confounding data, such as irrelevant factors that were inadvertently included.

To overcome this, you can follow separate lines of evidence to the point that they join together. This point of convergence is called a consilience, and the concept is often used in scientific research.1

There are many examples of consilience in science, such as a study of chromosome variation that used evidence from archeology, languages and ancient texts to support a conclusion pertaining to the prehistoric spread of chromosomes.2 Continental drift was hypothesized based on multiple lines of evidence, and it was later supported by empirical evidence in the 1950s.3 NASA used a convergence of three lines of evidence to conclude there were signs of water on the planet Mercury.4

While commonly used in science, this approach also can be applied to an industrial setting. It’s especially useful when there are multiple or contradictory lines of evidence, which could be the case in the failure of a complex product.

Some evidence may lead you to the root cause, and other evidence can be unrelated to it. Misleading evidence is like a red herring in a detective story—a narrative device the author uses to distract the detective from the real evidence. For example, a quality engineer who investigates a failure may find several measurements out of specification, but these deviations may be unrelated to the cause of the problem under investigation. Lines of evidence should—but do not always—converge at the smoking gun.

Following lines of evidence is a method that supports the use of quality tools, such as the classic seven (cause-and-effect diagram, check sheet, control chart, histogram, Pareto chart, scatter diagram and stratification). Statistical analysis also should aid in the root cause investigation.

Each resulting line of evidence should be followed individually and as a part of the whole to determine which lines support or contradict one another. A special emphasis should be placed on determining the reasons for contradictory evidence.

Ideally, multiple lines of evidence should converge at one or more interrelated root causes. This consilience, however, should not be viewed as a guarantee that the convergence point is the true root cause. Confirmation testing of the final conclusion must be performed, particularly when multiple lines of evidence point to a root cause with limited direct empirical evidence.

A root cause can be verified by removing the source of the problem and ensuring the problem is gone. The root cause condition then should be recreated to determine whether the problem recurs to ensure the root cause did indeed result in the problem. While NASA scientists have followed lines of evidence on the surface of Mercury, it’s just as effective on the production floor back here on Earth.


  1. William Whewell, The Philosophy of the Inductive Sciences: Founded Upon Their History, Vol. 2, Forgotten Books, 2012.
  2. S.O.Y. Keita, "History in the Interpretation of the Pattern of p49a,f TaqI RFLP Y-Chromosome Variation in Egypt: A Consideration of Multiple Lines of Evidence," American Journal of Human Biology, Vol. 17, No. 5, 2005, pp. 559-567.
  3. Dale E. Ingmanson and William J. Wallace, Oceanography: An Introduction, fourth edition, Wadsworth Publishing Co. Inc., 1988.
  4. Tricia Talbert, ed., "Messenger Finds New Evidence for Water Ice at Mercury’s Poles," NASA.gov, Nov. 29, 2012, http://tinyurl.com/mercurywater.

Matthew Barsalou is a statistical problem resolution Master Black Belt in the problem resolution and statistical methods department at BorgWarner Turbo Systems Engineering GmbH in Kirchheimbolanden, Germany. He has a master’s degree in business administration and engineering from Wilhem Büchner Hoschschule in Darmstadt, Germany, and a master’s degree in liberal studies from Fort Hays State University in Hays, KS. An ASQ senior member, Barsalou is an ASQ-certified quality technician and engineer, and Six Sigma Black Belt. He is also a technical reviewer for QP, editor of the Statistics Division’s Statistics Digest and the ASQ country counselor for Germany.

good article.
insufficient knowledge or insufficient information / obvious inaccurate data / obvious invalid data / irrelevant factors / missing environmental factors / even missing on-line verification equipment, etc might be various obstacles against to correctly identifying the root cause of problems and of solving it.

Aylin N. M.
--Aylin N. M., 01-18-2017

Average Rating


Out of 0 Ratings
Rate this article

Add Comments

View comments
Comments FAQ

Featured advertisers