P M V Subbarao Professor Mechanical Engineering Department

Slides:



Advertisements
Similar presentations
Design of Experiments Lecture I
Advertisements

CSUN Engineering Management Six Sigma Quality Engineering Week 11 Improve Phase.
Copyright © 2006 The McGraw-Hill Companies, Inc. Permission required for reproduction or display. 1 ~ Curve Fitting ~ Least Squares Regression Chapter.
Some terminology When the relation between variables are expressed in this manner, we call the relevant equation(s) mathematical models The intercept and.
Fractional Factorial Designs of Experiments
Experimental Design, Response Surface Analysis, and Optimization
6-1 Introduction To Empirical Models 6-1 Introduction To Empirical Models.
P M V Subbarao Professor Mechanical Engineering Department
11.1 Introduction to Response Surface Methodology
Chapter 3 Producing Data 1. During most of this semester we go about statistics as if we already have data to work with. This is okay, but a little misleading.
Using process knowledge to identify uncontrolled variables and control variables as inputs for Process Improvement 1.
EXPERIMENTAL ERRORS AND DATA ANALYSIS
MAE 552 Heuristic Optimization Instructor: John Eddy Lecture #19 3/8/02 Taguchi’s Orthogonal Arrays.
Statistics CSE 807.
Curve-Fitting Regression
L Berkley Davis Copyright 2009 MER301: Engineering Reliability Lecture 14 1 MER301: Engineering Reliability LECTURE 14: Chapter 7: Design of Engineering.
MAE 552 Heuristic Optimization
Evaluating Hypotheses
2-1 Sample Spaces and Events Conducting an experiment, in day-to-day repetitions of the measurement the results can differ slightly because of small.
MAE 552 Heuristic Optimization Instructor: John Eddy Lecture #16 3/1/02 Taguchi’s Orthogonal Arrays.
MAE 552 Heuristic Optimization Instructor: John Eddy Lecture #18 3/6/02 Taguchi’s Orthogonal Arrays.
MAE 552 Heuristic Optimization Instructor: John Eddy Lecture #20 3/10/02 Taguchi’s Orthogonal Arrays.
Analysis of Variance Chapter 3Design & Analysis of Experiments 7E 2009 Montgomery 1.
SIMPLE LINEAR REGRESSION
1 Seventh Lecture Error Analysis Instrumentation and Product Testing.
Improving Stable Processes Professor Tom Kuczek Purdue University
Determining the Size of
Go to Table of ContentTable of Content Analysis of Variance: Randomized Blocks Farrokh Alemi Ph.D. Kashif Haqqi M.D.
University of Florida Mechanical and Aerospace Engineering 1 Useful Tips for Presenting Data and Measurement Uncertainty Analysis Ben Smarslok.
Calibration & Curve Fitting
1 14 Design of Experiments with Several Factors 14-1 Introduction 14-2 Factorial Experiments 14-3 Two-Factor Factorial Experiments Statistical analysis.
SIMPLE LINEAR REGRESSION
Physics 114: Lecture 15 Probability Tests & Linear Fitting Dale E. Gary NJIT Physics Department.
Determining Sample Size
1 Chapter 1: Introduction to Design of Experiments 1.1 Review of Basic Statistical Concepts (Optional) 1.2 Introduction to Experimental Design 1.3 Completely.
Introduction to Design of Experiments
Design of Experiments Chapter 21.
DOE – An Effective Tool for Experimental Research
MultiSimplex and experimental design as chemometric tools to optimize a SPE-HPLC-UV method for the determination of eprosartan in human plasma samples.
Fractional Factorial Experiments (Continued) The concept of design resolution is a useful way to categorize fractional factorial designs. The higher the.
CPE 619 2k-p Factorial Design
Development of An ERROR ESTIMATE P M V Subbarao Professor Mechanical Engineering Department A Tolerance to Error Generates New Information….
Investment Analysis and Portfolio management Lecture: 24 Course Code: MBF702.
Ratio Games and Designing Experiments Andy Wang CIS Computer Systems Performance Analysis.
© 1998, Geoff Kuenning General 2 k Factorial Designs Used to explain the effects of k factors, each with two alternatives or levels 2 2 factorial designs.
Fundamentals of Data Analysis Lecture 9 Management of data sets and improving the precision of measurement.
A Really Bad Graph. For Discussion Today Project Proposal 1.Statement of hypothesis 2.Workload decisions 3.Metrics to be used 4.Method.
Quality of Curve Fitting P M V Subbarao Professor Mechanical Engineering Department Suitability of A Model to a Data Set…..
Why/When is Taguchi Method Appropriate? Friday, 1 st June 2001 Tip #7 Taguchi Method : When to Select a ‘larger’ OA to perform “Factorial Experiments”
EQT373 STATISTIC FOR ENGINEERS Design of Experiment (DOE) Noorulnajwa Diyana Yaacob School of Bioprocess Engineering Universiti Malaysia Perlis 30 April.
Experimental Design If a process is in statistical control but has poor capability it will often be necessary to reduce variability. Experimental design.
Various topics Petter Mostad Overview Epidemiology Study types / data types Econometrics Time series data More about sampling –Estimation.
1 Chapter 1: Introduction to Design of Experiments 1.1 Review of Basic Statistical Concepts (Optional) 1.2 Introduction to Experimental Design 1.3 Completely.
MSE-415: B. Hawrylo Chapter 13 – Robust Design What is robust design/process/product?: A robust product (process) is one that performs as intended even.
Design Of Experiments With Several Factors
Unit 1c: Scientific Method & Inquiry. The Methods Biologists Use The common steps that biologists and other scientists use to gather information and answer.
Review of fundamental 1 Data mining in 1D: curve fitting by LLS Approximation-generalization tradeoff First homework assignment.
Trees Example More than one variable. The residual plot suggests that the linear model is satisfactory. The R squared value seems quite low though,
Optimising Manufacture of Pressure Cylinders via DoE Dave Stewardson, Shirley Coleman ISRU Vessela Stoimenova SU “St. Kliment Ohridski”
Sampling Design and Analysis MTH 494 Lecture-21 Ossam Chohan Assistant Professor CIIT Abbottabad.
 Will help you gain knowledge in: ◦ Improving performance characteristics ◦ Reducing costs ◦ Understand regression analysis ◦ Understand relationships.
Designs for Experiments with More Than One Factor When the experimenter is interested in the effect of multiple factors on a response a factorial design.
TAUCHI PHILOSOPHY SUBMITTED BY: RAKESH KUMAR ME
Why/When is Taguchi Method Appropriate?
APPROACHES TO QUANTITATIVE DATA ANALYSIS
IE-432 Design Of Industrial Experiments
ENM 310 Design of Experiments and Regression Analysis Chapter 3
DESIGN OF EXPERIMENTS by R. C. Baker
14 Design of Experiments with Several Factors CHAPTER OUTLINE
Presentation transcript:

P M V Subbarao Professor Mechanical Engineering Department Design of Experiments P M V Subbarao Professor Mechanical Engineering Department Selection of Significant Parameters for Experimentation…..

Introductory Remarks Reduced Temperature TR = T/Tc Reduced Pressure pR = p/pc Reduced Temperature TR = T/Tc Introductory Remarks Many of the thermal experiments involve multi-variable functions. The goal of any experimental activity is to get the maximum realistic information about a system. Large number of variables demand large number of measurements to get maximum realistic information. Modern theory of experiments prove that it is not always true that higher number of measurements will give maximum realistic information. Larger the number of measurements, huge will be the total error that enters into the measurement equation. Larger number of measurements lead to more costly experimentation.

This Selection Process is known as It is important to obtain maximum realistic information with the minimum number of well designed experiments. An experimental program recognizes the major “factors” that affect the outcome of the experiment. The factors may be identified by looking at all the quantities that may affect the outcome of the experiment. The most important among these may be identified using: a few exploratory experiments or From past experience or based on some underlying theory or hypothesis. This Selection Process is known as Design of Experiments.

Special Terminology : Design of Experiments Response variable Measured output value Factors Input variables that can be changed Levels Specific values of factors (inputs) Continuous or discrete Replication Completely re-run experiment with same input levels Used to determine impact of measurement error Interaction Effect of one input factor depends on level of another input factor

Study of Real Diesel Engine Cycle

Experimental Analysis of Diesel Cycle

Measurement of Pressure Vs Crank Angle in A Diesel Engine

Repeated Measurement of Diesel Cycle

Design of Experiments (DOE) A statistics-based approach to design experiments A methodology to achieve a predictive knowledge of a complex, multi-variable process with the fewest acceptable trials. An optimization of the experimental process itself

Major Approaches to DOE Factorial Design Taguchi Method Response Surface Design

Factorial Design : Full factorial design A full factorial design of experiments consists of the following: Vary one factor at a time Perform experiments for all levels of all factors Hence perform a large number of experiments that are needed! All interactions are captured. Consider a simple design for the following case: Let the number of factors = k Let the number of levels for the ith factor = ni The total number of experiments (N) that need to be performed is

2k factorial design Used as a Preliminary Experimentation !!! Each of the k factors is assigned only two levels. The levels are usually High = 1 and Low = -1. Scheme is useful as a preliminary experimental program before a more ambitious study is undertaken. The outcome of the 2k factorial experiment will help identify the relative importance of factors and also will offer some knowledge about the interaction effects.

DOE - Factorial Designs - 23 Trial A B C 1 Lo 2 Hi 3 4 5 6 7 8

DOE - Factorial Designs - 23 Trial A B C 1 -1 2 +1 3 4 5 6 7 8

Output Matrix Let us represent the outcome of each experiment to be a quantity y. Thus y1 will represent the outcome of experiment number 1 with all three factors having their “LOW” values, y2 will represent the outcome of the experiment number 2 with the factors A & B having the “Low” values and the factor C having the “High” value and so on. The outcome of the experiments may be represented as the following matrix:

Outcome Matrix Trial xA xB xC y1 -1 y2 +1 y3 y4 y5 y6 y7 y8 How to find the degree of contribution of xA,xB & xC on y?

A simple regression model that may be used can have up to eight parameters. Thus we may represent the regression equation as The p’s are the parameters that are determined by using the “outcome” matrix by the simultaneous solution of the following eight equations:

-1,+1,+1 Spread of C +1,+1,+1 +1,+1,-1 -1,+1,-1 +1,-1,+1 xmean xA xB xC -1,-1,+1 Spread of B -1,-1,-1 +1,-1,-1 Spread of A

It is easily seen that the parameter p0 is simply the mean value of y. This is obtained by putting xA= xA= xC=0 corresponding to the mean values for the factors. It is thus seen that the values of y- p0 at the corners of the square indicate the deviations from the mean value. The mean of the square of these deviations is the variance of the sample data collected in the experiment. The influence of the factors may then be gauged by the contribution of each term to the variance.

The sample variance is thus given by The deviation with respect to the mean is obviously given by It may be verified that the total sum of squares (SST) of the deviations is given by The sample variance is thus given by

Contributions to the sample variance are given by 8 times the square of the respective parameter (p) and hence we also have Here SSA means the sum of squares due to variation in level of xA and so on. The relative contributions to the sample variance are represented as percentage contributions in the following table:

Variance contribution % SST 100 SSA 8pA2 SSB 8pB2 SSC 8pC2 SSAB 8pAB2 SSA/SST X 100 SSB 8pB2 SSB/SST X 100 SSC 8pC2 SSC/SST X 100 SSAB 8pAB2 SSAB/SST X 100 SSBC 8pBC2 SSBC/SST X 100 SSCA 8pCA2 SSCA/SST X 100 SSABC 8pABC2 SSABC/SST X 100 Thus the dominant factor is the factor which has the highest value of percentage of contribution.

DOE - Fractional Factorial Designs In a multivariable experiments, with k number of variables and l number of levels per variable demands lk number of measurements for complete understanding of the process or calibration. In statistics, fractional factorial designs are experimental designs consisting of a carefully chosen subset (fraction) of the experimental runs of a full factorial design. The subset is chosen so as to exploit the sparsity-of-effects principle using a fraction of the effort of a full factorial design in terms of experimental runs and resources. Fractional designs are expressed using the notation lk − p, where l is the number of levels of each factor investigated, k is the number of factors investigated, and p describes the size of the fraction of the full factorial to be eliminated. A design with p such generators is a 1/(lp) fraction of the full factorial design.

DOE – Factorial Designs (Fractional: 7 factor, 2 level; 128  8) Trial A B C D E F G 1 Lo 2 Hi 3 4 5 6 7 8

One half factorial design For a system with k factors and 2 levels the number of experiments in a full factorial design will be 2k. For example, if k=3, this number works out to be 23=8. The eight values of the levels would correspond to the corners of a cube as represented by Figure. A half factorial design would use 2k-1 experiments. With k=3 this works out to be 22=4. The half factorial design would cut the number of experiments by half. In the half factorial design we would have to choose half the number of experiments and they should correspond to four of the eight corners of the cube. We may choose any set as given below:

Half Factorial Matrix Point xA xB xC y1 + - y2 y3 y4 The three column vectors are Orthogonal…..

Half Factorial Matrix Point xA xB xC y1 - + y2 y3 y4 The three column vectors are Orthogonal…..

More on full factorial design We like to generalize the ideas described above in what follows. Extension to larger number of factors as well as larger number of levels would then be straight forward. Let the High and Low levels be represented by + an –respectively. In the case of 22 factorial experiment design the following will hold:

We note that the product of any two columns is zero. Also the column sums are zero. Hence the three columns may be considered as vectors that form an orthogonal set. In fact while calculating the sample variance earlier these properties were used without being spelt out. Most of the time it is not possible to conduct that many experiments! The question that is asked is: “Can we reduce the number of experiments and yet get an adequate representation of the relationship between the outcome of the experiment and the variation of the factors?” The answer is in general “yes”. Replace the full factorial design with a fractional factorial design. In the fractional factorial design only certain combinations of the levels of the factors are used to conduct the experiments. This ploy helps to reduce the number of experiments. The price to be paid is that all interactions will not be resolved.

In this simple case of two or three factors the economy of reducing the number of experiments by one may not be all that important. However it is very useful to go in for a fractional factorial design when the number of factors is large and When it is expected some factors or interactions between some factors to be unimportant. The fractional factorial experiment design is useful when main effects dominate with interaction effects being of lower order.

DOE - Taguchi Method Dr. Taguchi of Nippon Telephones and Telegraph Company, Japan has developed a method based on " ORTHOGONAL ARRAY " experiments. This gives much reduced " variance " for the experiment with " optimum settings " of control parameters. "Orthogonal Arrays" (OA) provide a set of well balanced (minimum) experiments serve as objective functions for optimization.

Taguchi Method : When to Select a ‘larger’ OA to perform “Factorial Experiments” We always ‘think’ about ‘reducing’ the number of experiments (to minimize the ‘resources’ – equipment, materials, manpower and time) However, doing ALL / Factorial experiments is a good idea if Conducting experiments is ‘cheap/quick’ but measurements are ‘expensive/take too long’ The experimental facility will NOT be available later to conduct the ‘verification’ experiment We do NOT wish to conduct separate experiments for studying interactions between Factors

Taguchi Method Design of Experiments The general steps involved in the Taguchi Method are as follows: 1. Define the process objective, or more specifically, a target value for a performance measure of the process. 2. Determine the design parameters affecting the process. The number of levels that the parameters should be varied at must be specified. 3. Create orthogonal arrays for the parameter design indicating the number of and conditions for each experiment. The selection of orthogonal arrays is based on the number of parameters and the levels of variation for each parameter, and will be expounded below. 4. Conduct the experiments indicated in the completed array to collect data on the effect on the performance measure. 5. Complete data analysis to determine the effect of the different parameters on the performance measure.

Determining Parameter Design Orthogonal Array The effect of many different factors on the performance characteristic in a condensed set of experiments can be examined by using the orthogonal array experimental design proposed by Taguchi. The main factors affecting a process that can be controlled (control Factors) should be determined. The levels at which these parameters should be varied must be determined. Determining what levels of a variable to test requires an in-depth understanding of the process, including the minimum, maximum, and current value of the parameter. If the difference between the minimum and maximum value of a parameter is large, the values being tested can be further apart or more values can be tested. If the range of a parameter is small, then less values can be tested or the values tested can be closer together. Typically, the number of levels for all parameters in the experimental design is chosen to be the same to aid in the selection of the proper orthogonal array.

Orthogonal Array Selector Number of Levels Number of Factors

Control Factors and Levels Factorial Combinations Taguchi Method : How to Select a ‘larger’ OA to perform “Factorial Experiments” Control Factors and Levels Factorial Combinations Suitable OA 2 CF / 2-levels 4 L4 3 CF / 2-levels 8 L8 4 CF / 2-levels 16 L16 5 CF / 2-levels 32 L32

L27 Array

L 50 Array