Scribd
Upload
77 views

Engineering Experiments: FE 536 Design of Experiments and Process Optimization

This document provides an overview of the course FE 536 Design of Experiments and Process Optimization. It discusses key concepts in experimental design including factors, levels, and responses. It provides examples of two-factor experiments and the analysis of variance. It also outlines the four eras in the history of design of experiments and common areas where design of experiments is applied, such as screening experiments, optimization experiments, and hypothesis testing. Students are assigned homework to design a simple experiment.

Uploaded by

Oznur Baskan
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
77 views

Engineering Experiments: FE 536 Design of Experiments and Process Optimization

This document provides an overview of the course FE 536 Design of Experiments and Process Optimization. It discusses key concepts in experimental design including factors, levels, and responses. It provides examples of two-factor experiments and the analysis of variance. It also outlines the four eras in the history of design of experiments and common areas where design of experiments is applied, such as screening experiments, optimization experiments, and hypothesis testing. Students are assigned homework to design a simple experiment.

Uploaded by

Oznur Baskan
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

FE 536 Design of Experiments and Process Optimization

Text book: Design and Analysis of Engineering Experiments


by

Course Management System https://cms.iyte.edu.tr/ doe2013

Douglas C. Montgomery Wiley, 6th edition, 2005


1 2

Design of Engineering Experiments Chapter 1 Introduction to DOX


An experiment is a test or a series of tests (purposeful changes are made to the input variables and output responses are observed) Experiments are used widely in the engineering world Process characterization & optimization Evaluation of material alternatives Product design & development All experiments are designed experiments, some are poorly designed, some are well-designed D.C.M. Design of experiments: planning and conducting experiments, analyzing the response data so that valid and objective conclusions are obtained.

Engineering Experiments
A process is a combination of machines, people, and other resources that transforms some input into an output.

Reduce time to design/develop new products & processes Improve performance of existing processes Improve reliability and performance of products Achieve product & process robustness Evaluation of materials, design alternatives, setting component & system tolerances, etc.
4

What is Experimental Design?


Objectives of Experimental Design Determine which variables (xs) are most influential on the response, y Determine where to set the influential xs so that y is near the nominal requirement Determine where to set the influential xs so that variability is small

Four Eras in the History of DOX


The agricultural origins, 1918 1940s R. A. Fisher & his co-workers Profound impact on agricultural science Factorial designs, ANOVA The first industrial era, 1951 late 1970s Box & Wilson, response surfaces Applications in the chemical & process industries The second industrial era, late 1970s 1990 Quality improvement initiatives in many companies Genichi Taguchi and robust parameter design, process robustness The modern era, beginning 1990 Use of DOE in business, financial services, government operations along the industrial productions.
5 6

The Basic Principles of DOX


Randomization Running the trials in an experiment in random order Notion of balancing out effects of lurking variables Replication Sample size (improving precision of effect estimation, estimation of error or background noise) Replication versus repeat measurements? Blocking Dealing with nuisance factors
7

Strategy of Experimentation
Best-guess experiments Used a lot More successful than you might suspect, but there are disadvantages One-factor-at-a-time (OFAT) experiments Sometimes associated with the scientific or engineering method Devastated by interaction, also very inefficient Statistically designed experiments Based on Fishers factorial concept
8

Basic Terms in DOE Planning, Conducting & Analyzing an Experiment


1. 2. 3. 4. 5. 6. 7. Recognition of & statement of problem Choice of factors, levels, and ranges Selection of the response variable(s) Choice of design Conducting the experiment Statistical analysis (Anova, linear regression) Drawing conclusions, recommendations
Pre-experimental planning

Response the output of the system you are measuring (e. g. activity, cost, yield) Factor an input variable that may affect the response (e. g. temp., pH, conc., catalyst type) - qualitative or quantitative. Level a specific value a factor may take Trial a single instance of the setting of factors and the measurement of the response Replication repeated instances of the setting of factors and the measurement of the response (Note the difference from the repetitions) Effect what happens to the response when factor levels change? Interaction joint effects of multiple factors
10

Example to Two-Factor Experiments


Consider a chemical process in which the effect of the concentration of one reactant and the amount of catalyst on the conversion (yield) is investigated. Factor A: reactant concentration Factor B: Catalyst amount Response: Yield The experiments will be performed at two levels of each factor; high levels and low levels

Example to Two-Factor Experiments


Factors A B low level 15 % 1 lb high level 25 % 2 lb

Number of replicates (n) : 3 This means the experiment of each combination is run three times. Replicates and repeated measurements are different than each other. In replications, the basic experiment is repeated certain times, n.

11

12

Treatment combinations:
Factors A B Treatment Combinations Response (Yield) Measurements _____Replicates_____ I II III 28 36 18 31 25 32 19 30 27 32 23 29

ANOVATable
Factorial Fit: Yield versus Reactant (%), Enzyme (lb) Estimated Effects and Coefficients for Yield (coded units) Term Constant Reactant (%) Enzyme (lb) Reactant (%)*Enzyme (lb) Effect 8.333 -5.000 1.667 Coef 27.500 4.167 -2.500 0.833 P 0.000 0.000 0.002 0.183

+ +
+

A low, B low A high, B low A low, B high A high, B high

significant term significant term significant term insignificant term??

PRESS = 70.5 R-Sq = 90.30% R-Sq(pred) = 78.17% R-Sq(adj) = 86.66% Estimated Coefficients for Yield using data in uncoded units Term Constant Reactant (%) Enzyme (lb) Reactant (%)*Enzyme (lb)
13

After running the experiments, the data are analysed for the effects of factors, and the regression equation between the response and the factors Analysis of Variance (ANOVA) is used.

Coef 28.3333 0.333333 -11.6667 0.333333


14

Areas where DOE is used


Screening: used at the beginning of the experimental procedure:
(i) to explore many factors in order to reveal whether they have an influence on the responses (ii) to identify their appropriate ranges. With a screening design, the experimenter is able to extract a yes or no answer with regard to the influence of a particular factor. Information is also gained about how to modify the settings of the important factors, to possibly further enhance the result.Screening designs need few experiments in relation to the number of factors.

A Typical Investigation:
Treatment of Water Aim: improve the quality of drinking water Cleaning of water: by using an ion-exchange resin for the adsorption of sulfate A new resin, which is more effective and cheaper in sulfate removal, was developed by the chemists.

Optimization : used after screening;


(i) to predict the response values for all possible combinations of factors within the experimental region (ii) to identify an optimal experimental point. With an optimization design the experimenter is able to extract detailed information regarding how the factors combine to influence the responses. Optimization designs require many experiments in relation to the number of investigated factors (factors: 3, number of experiments: 17).

Problem 1: Is there evidence that new resin reduces sulfate levels in partially treated water? comparison of an average with a mean (hypothesis test) Problem 2: How does this new resin compare to the expensive high-purity resin? comparison of two averages (hypothesis test) Problem 3: How does the new resin compare with five commercially available resins? comparison of more than two treatments (one factor ANOVA)
15 16

A Typical Investigation:
Treatment of Water
Problem 4: Since the removal of sulfate is still inefficient, could the modifications in equipments lead to improvement in sulfate removal? Factorial design to investigate the effect of more than one factor at different levels Problem 5: How is sulfate adsorption process affected by pH, hardness, presence of chemicals A, B, C,.....,H? Fractional factorial design Result: the adsorption of resin is insensitive to changes in pH, hardness and presence of some chemicals Problem 6: How do the process variables affect the quality and manufacturing cost of new resin? Optimization of production by Response Surface Methodology Result: Feasible process for the full scale manufacture of the resin with satisfactory quality at reasonable cost

Population Sample

17

Hypothesis Testing and Design of Experiments can be considered as sampling strategies, which are used to make inferences about the population. 18

HOMEWORK TO THE CLASS


1. Read Chapter I 2. A hypothetical design Think about a simple experiment that you can easily conduct at home or in the university labs in a similar way given in the our example. Types of experiment can be: A single factor at different levels Two or more than two factors at two levels Two factors at three levels Three factors at two levels List the possible factors and the response variable(s).
19

You might also like