Understanding Taguchi Approach
Design of Experiments (DOE) and the Taguchi Approach
You can learn about different topics in the technique by reading the brief descriptions in this page.
Taguchi Approach – Do not be confused by the association of the name Taguchi with DOE. Dr. Genechi Taguchi has offered a standardized and relatively simple method of applying the DOE technique. This is why his name is associated with DOE. Bear in mind that for simpler applications of the technique, you need not worry about the subtle differences. Just concentrate on learning what the technique is all about and how to apply it in your projects, assuming that learning how to apply is your intention. For your purposes view them as the same.
DOE/Taguchi- It is a technique to lay out experimental (investigation, studies, survey, tests, etc) plans in the most logical, economical, and statistical way. You can potentially benefit from it when you want to determine the: most desirable design of your product, the best parameters combination for your process, the most robust recipe for your formulation, the permanent solution for some of your production problems, the most critical validation/durability test condition, most effective survey/data collection plan, etc.
What it does for you- The DOE is about option selection. It works best when you already have a working design (product, process, system, plan, etc.) and you wish to put the finishing touches. If you are after developing a new product or process, it is not the right time for DOE. You need to look for other means to determine the working parameters. It is only after you reach a workable condition that satisfies your objectives, and know your process/system well, you would benefit from DOE. You will apply DOE to determine the best among many good conditions. In other words, it is something that will help you hone in the process to perfection or help you select something that will consistently produce what you want, all the time.
Why do DOE?
“I use DOE to help my clients optimize processes for value-added products while minimizing production costs. In the manufacturing of wood products like value-added oriented strand board panels and specialty plywood panels, several parameters affect the process. DOE is the tool to deal with processes with so many variables.”
– Dr. David Barrett, Professor, Department of Wood Science, Vancouver, BC, Canada.
“Designed experiments can help untangle the nature of complex and otherwise confusing relationships faster than many of the alternatives. ‘Thinking DOE’ helps one think more systematically, regardless of the application.”
– Paul Selden
“Imagine the feeling of finding something you want when it is on sale at a deeply discounted price. A well-thought-out experiment allows you to find out so much for relatively little time and effort; you just can’t beat it for economy, efficiency, and effectiveness. And it is so beautiful, watching knowledge unfold like a flower.”
– Larry Smith, Manager, “Champion of Quality”, Ford Motor Company, Dearborn, Michigan.
“When I need to adjust one thing to improve performance, or when the single source of the problem is known, often I can arrive at the solution intuitively. But when I’m dealing with more than one factor, or looking for unknown sources of problem, DOE comes to help.”
“I use DOE to identify the process parameters for enhancing the ceramic tensile strength. It saves me a lot of time by avoiding testing all the process probabilities. DOE/Taguchi method is an effective tool for me to study my process by experimental means.”
“Like all other quality tools, DOE is an important technique. But, the benefit is in the way one uses it. You have to learn how to apply first.”
“When comes to deciding what’s best for my product and process designs, opinions and judgments slow me down. When I make decisions based on DOE results, everyone agrees.”
“I consider DOE to be the tool to give a finishing touch before settling on designs. I believe we gain a lot when we use DOE to fine-tune product designs before release and optimize processes before production begins.”
“I use DOE to solve production-related problems when basic disciplines (like 8D) do not offer the technical solution.”
– Unknown Practitioners
“I use DOE in HAZOP (Hazard & Operability Studies) and QRA (Quantitative Risk Analysis) of offshore structure/process platforms, Oil rigs, and On-land oil installations like Group gathering stations, etc. I am quite well versed in Six Sigma techniques, and also in Dr Taguchi’s method of OA (Orthogonal Arrays) as a tool in the analysis phase of Six Sigma as well as Dr. Taguchi’s concept of loss function for a robust design. “
– S.R. IYER, March 28, 2003
“Simulation models of manufacturing systems involve many design or operation parameters. The optimal settings for these must be determined by running the model many times. DOE provides an efficient and effective way to conduct experiments with the model of the system after the model has been verified and validated. It allows the KPOV (key process output variables) to be modeled in terms of the KPIV (key process output variables). In general, DOE leads to a better understanding of the system and interactions among the design variables or operational variables.”
– Dr. S. Balachandran, Professor of Industrial Engineering, UW-Platteville,
“DOE helps to reduce product/process development time and hence costs associated with product/process development process. It improves process yield, reliability, and process capability. It can be used to reduce product performance sensitivity to various sources of noise (such as environmental variations, manufacturing imperfections, product-to-product variations, machine performance deterioration, etc.) “
– Dr. J Antony, Intl. Mfg, Centre, University of Warwick, Coventry, England, UK.
Your learning strategy – For comprehensive knowledge of the technique, you would want to know about (1) theory and math, (2) application methods, and (3) Philosophy and working disciplines (planning). Do not spend too much in the theory and statistical calculation. You need to focus on what they mean rather than how it is done. Try to muster the application methods and standard experiment design techniques. Understand the philosophy and follow the discipline well. This is what gives you the most benefits. The theory and application methods are routine and same for all projects, the experiment planning is what will be unique to your project. Unfortunately, it is something you will not learn well by reading. To know it well, learn from expert practitioners or learn as you go on applying.
How to acquire application skills –
* Review and download DOE Topic Overview in PDF format from the link on this site (Free)
* Search for other literature on the web (Free)
* Visit your local library and borrow a book or two on the subject (Free)
* Buy books if you can afford them (costs vary between $50 – 150 USD, 2004 price)
* Download DEMO software (Free for L-8 experiments)
* Design small (L-4, L-8, or L-9) experiments and hand-calculate numbers (All above are must for students and researchers)
* If you are not comfortable, consider attending our public seminar. This will sort-circuit your learning time and help build the skills for immediate applications.
* If your interest is in companywide applications, consider hosting our 4-day seminar with an application workshop. This will make all attendees ready for immediate application. You should consider onsite seminars when your projects involve people from many areas within your organization. The purpose is not to make everyone an expert but to have all understand the benefits and secure support for the project. Most optimistically, a few among the attendees in a session (10 – 20 people) will develop and maintain application skills.
* Should your interest be in an immediate project application, seek help with the application. Often, the cost of outside consultation will be minimal compared to the potential benefits of the project. The area you would most benefit from an experienced facilitator is in the experiment planning (brainstorming). After you have acquired the knowledge about the technique, it is the discipline you need to follow in planning the experiments that will take longer time to develop.
* If you are interested more about project applications, but uncertain about when and where the needs will develop, consider retaining our application assistance on demand.
Subject Overview (The Taguchi Approach):
Design Of Experiments (DOE) is a powerful statistical technique introduced by R. A. Fisher in England in the 1920s to study the effect of multiple variables simultaneously. In his early applications, Fisher wanted to find out how much rain, water, fertilizer, sunshine, etc. are needed to produce the best crop. Since that time, much development of the technique has taken place in the academic environment but did help generate many applications on the production floor.
As a researcher in the Electronic Control Laboratory in Japan, Dr. Genechi Taguchi carried out significant research with DOE techniques in the late 1940’s. He spent considerable effort to make this experimental technique more user-friendly (easy to apply) and applied it to improve the quality of manufactured products. Dr. Taguchi’s standardized version of DOE, popularly known as the Taguchi method or Taguchi approach, was introduced in the USA in the early 1980s. Today it is one of the most effective quality building tools used by engineers in all types of manufacturing activities.
The DOE using the Taguchi approach can economically satisfy the needs of problem-solving and product/process design optimization projects. By learning and applying this technique, engineers, scientists, and researchers can significantly reduce the time required for experimental investigations. DOE can be highly effective when you wish to:
– Optimize product and process designs, study the effects of multiple factors (i.e.- variables, parameters, ingredients, etc.) on the performance, and solve production problems by objectively laying out the investigative experiments. (Overall application goals)
– Study the Influence of individual factors on performance and determine which factor has more influence, and which ones have less. You can also find out which factor should have tighter tolerance and which tolerance should be relaxed. The information from the experiment will tell you how to allocate quality assurance resources based on the objective data. It will indicate whether a supplier’s part causes problems or not (ANOVA data), and how to combine different factors in their proper settings to get the best results (Specific Objectives).
Further, the experimental data will allow you to determine:
– How to substitute a less expensive part to get the same performance
– How much money you can save on the design improvement you propose
– How you can determine which factor is causing the most variations in the result
– How you can set up your process such that it is insensitive to the uncontrollable factors
– Which factors have more influence on the mean performance
– What you need to do to reduce performance variation around the target
– How you can adjust factors for a system whose response varies proportional to signal factor (Dynamic response)
– How to combine multiple criteria of evaluation into a single index
– How you can adjust factors for overall satisfaction of criteria of evaluations
– How the uncontrollable factors affect the performance
etc.,
Advantage of DOE Using Taguchi Approach –The application of DOE requires careful planning, prudent layout of the experiment, and expert analysis of results. Based on years of research and applications Dr. Genechi Taguchi has standardized the methods for each of these DOE application steps. Thus, DOE using the Taguchi approach has become a much more attractive tool to practicing engineers and scientists.
Experiment planning and problem formulation – Experiment planning guidelines are consistent with modern work disciplines of working as teams. Consensus decisions about experimental objectives and factors make the projects more successful.
Experiment layout -High emphasis is put on the cost and size of experiments… The size of the experiment for a given number of factors and levels is standardized… Approach and priority for column assignments are established… Clear guidelines are available to deal with factors and interactions (interaction tables)… Uncontrollable factors are formally treated to reduce variation… Discrete prescriptions for setting up test conditions under uncontrollable factors are described… Guidelines for carrying out the experiments and the number of samples to be tested are defined
Data analysis -Steps for analysis are standardized (main effect, NOVA, and Optimum)… Standard practice for determination of the optimum is recommended… Guidelines for the test of significance and pooling are defined…
Interpretation of results – Clear guidelines about the meaning of error term… Discrete indicator about confirmation of results (Confidence interval)… Ability to quantify improvements in terms of dollars (Loss function)
Overall advantage – DOE using the Taguchi approach attempts to improve quality which is defined as the consistency of performance. Consistency is achieved when variation is reduced. This can be done by moving the mean performance to the target as well as by reducing variations around the target. The prime motivation behind the Taguchi experiment design technique is to achieve reduced variation (also known as ROBUST DESIGN). This technique, therefore, is focused on attaining the desired quality objectives in all steps. The classical DOE does not specifically address quality.
“The primary problem addressed in classical statistical experiment design is to model the response of a product or process as a function of many factors called model factors. Factors, called nuisance factors, which are not included in the model, can also influence the response… The primary problem addressed in Robust Design is how to reduce the variance of a product’s function in the customer’s environment.”
-Madhav Phadke, Quality Engineering Using Robust Design
TAGUCHI METHOD REVIEW
APPLICATION STEPS
The Taguchi method is used to improve the quality of products and processes. Improved quality results when a higher level of performance is consistently obtained. The highest possible performance is obtained by determining the optimum combination of design factors. The consistency of performance is obtained by making the product/process insensitive to the influence of the uncontrollable factor. In Taguchi’s approach, optimum design is determined by using the design of experiment principles, and consistency of performance is achieved by carrying out the trial conditions under the influence of the noise factors.
1. BRAINSTORMING
This is a necessary first step in any application. The session should include individuals with first-hand knowledge of the project. All matters should be decided based on group consensus, (One person — One vote).
– Determine what you are after and how to evaluate it. When there is more than one criterion of evaluation, decide how each criterion is to be weighted and combined for the overall evaluation.
– Identify all influencing factors and those to be included in the study.
– Determine the factor levels.
– Determine the noise factor and the condition of repetitions.
2. DESIGNING EXPERIMENTS
Using the factors and levels determined in the brainstorming session, the experiments now can be designed and the method of carrying them out is established. To design the experiment, implement the following:
– Select the appropriate orthogonal array.
– Assign factor and interaction to columns.
– Describe each trial condition.
– Decide the order and repetitions of trial conditions.
3. RUNNING EXPERIMENT
Run experiments in random order when possible.
4. ANALYZING RESULTS
Before analysis, the raw experimental data might have to be combined into an overall evaluation criterion. This is particularly true when there are multiple criteria for evaluation.
Analysis is performed to determine the following:
– The optimum design.
– Influence of individual factors.
– Performance at the optimum condition & confidence interval (C. I.).
– Relative influence of individual factors. etc.
5. RUNNING CONFIRMATION EXPERIMENTS)
Running the experiments at the optimum condition is the necessary final step.