![Erik Vanhatalo](/img/default-banner.jpg)
- 36
- 74 078
Erik Vanhatalo
Sweden
Registrace 9. 10. 2012
This channel is primarily used for publishing recorded lectures and teaching material from my teaching at Luleå University of Technology, Sweden.
Robust design | Example session | Minitab Tutorial
This video is the second part of a guest lecture in the "Quality Engineering and Design" course given at Linköping University. The guest lecture was given through a recording made in August 2021. The video provides an example session and a Minitab tutorial using an example of a robust design experiment with ONE NOISE FACTOR and THREE CONTROL FACTORS. Example data are from Example 6.2 and 12.1 in Montgomery, D.C. (2017). "Design and analysis of experiments", Wiley. It may be best to follow along in Minitab and pause the video when needed. Best wishes, Erik
The analysis is based on the so-called "Combined array design" using a factorial desgn and "the response model" which can be used to produce one model for the expected value and one model for the variance.
Chapters:
0:00 Introduction
1:40 Example, data, and design. Example 6.2 and 12.1 in Montgomery (2017).
5:21 Building the experimental design (2^4 factorial) in Minitab.
7:12 Naming and classifying the factors in Minitab.
9:31 Some discussion of Minitab options when creating the design.
10:40 Created experimental design in Minitab.
11:33 Adding the response variable in Minitab.
13:30 Complete design in Minitab (with the response variable)
13:48 The main steps in the analysis (repetition).
14:25 Analyzing the experimental design in Minitab.
15:08 Choosing the model terms in Minitab
15:59 Some discussion on other analysis options in Minitab.
18:50 First analysis step - full model with all terms.
23:16 Second analysis step - Reduced model (only significant terms).
27:59 Residual analysis of the entertained model.
31:39 Producing the "response model."
33:43 Expected value (mean) and variance given the response model.
35:55 Using the response model to find "good" values of the control factors.
36:33 Studying interactions between control and noise factors in Minitab.
41:09 Using useful interactions between control and noise factors.
43:02 Minimizing the variance equation in MS Excel.
48:20 Setting of significant control factors to minimize variance.
The analysis is based on the so-called "Combined array design" using a factorial desgn and "the response model" which can be used to produce one model for the expected value and one model for the variance.
Chapters:
0:00 Introduction
1:40 Example, data, and design. Example 6.2 and 12.1 in Montgomery (2017).
5:21 Building the experimental design (2^4 factorial) in Minitab.
7:12 Naming and classifying the factors in Minitab.
9:31 Some discussion of Minitab options when creating the design.
10:40 Created experimental design in Minitab.
11:33 Adding the response variable in Minitab.
13:30 Complete design in Minitab (with the response variable)
13:48 The main steps in the analysis (repetition).
14:25 Analyzing the experimental design in Minitab.
15:08 Choosing the model terms in Minitab
15:59 Some discussion on other analysis options in Minitab.
18:50 First analysis step - full model with all terms.
23:16 Second analysis step - Reduced model (only significant terms).
27:59 Residual analysis of the entertained model.
31:39 Producing the "response model."
33:43 Expected value (mean) and variance given the response model.
35:55 Using the response model to find "good" values of the control factors.
36:33 Studying interactions between control and noise factors in Minitab.
41:09 Using useful interactions between control and noise factors.
43:02 Minimizing the variance equation in MS Excel.
48:20 Setting of significant control factors to minimize variance.
zhlédnutí: 443
Video
Create timestamps in Youtube videos | Create chapters in Youtube videos | Online teaching
zhlédnutí 87Před rokem
In this brief tutorial I wanted to share an easy way with my colleagues to create time stamps or chapters in already recorded CZcams videos. I simply re-watch my video lecture using a notepad document and note down the time stamps. Chapters: 0:00 Introduction 0:39 Locate your video in CZcams studio 1:05 Using a simple notepad to note down timestamps 1:24 Locating appropriate timestamps of "modu...
Paper Helicopter lab - final and winning team 2020
zhlédnutí 950Před 4 lety
A short video from deciding run of the 2020 final in the design and analysis of experiments laboratory work assignment. The video features the winning helicopter and the runner-up helicopter. The laboratory work is part of the course: Quality Management, introduction course.
Analysis problems and potential solutions (in the analysis of designed experiments)
zhlédnutí 323Před 5 lety
This video exemplifies a number of analysis problems that may be encountered during the analysis of a planned experiment. Potential solutions to overcome these problems or to reduce their impact are discussed. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning material...
Robust design - introduction
zhlédnutí 3,2KPřed 5 lety
This video provides a short introduction to robust design. Two main approaches: Crossed Array Design and Combined Array Design are briefly explained with some examples. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning material for undergraduate and graduate students ...
Common second order designs in Response surface methodology
zhlédnutí 1,6KPřed 5 lety
This video discusses the central composite design (CCD) and the Box-Behnken design. Blocking in RSM is briefly discussed. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning material for undergraduate and graduate students following a course in design and analysis of ex...
Introduction to response surface methodology (RSM)
zhlédnutí 12KPřed 5 lety
This video introduces response surface methodology. The general principles and the method of steepest ascent is in focus in this video. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning material for undergraduate and graduate students following a course in design and ...
More on transformations (of the response) when analyzing experiments
zhlédnutí 289Před 5 lety
This video discusses standard or typical transformations of the response variable useful when analysing experiments. The video also shows some help on this topic found in teh Design Expert software (version 11). This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning materi...
Fractional factorial designs and fold-over
zhlédnutí 4,8KPřed 5 lety
This video provides an introduction to fractional factorial two-level designs. Fold-over techniques are also discussed. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning material for undergraduate and graduate students following a course in design and analysis of expe...
Introduction to blocking in experimental design
zhlédnutí 7KPřed 5 lety
The video introduces the blocking strategy with a focus on blocking in two-level factorial designs. Randomization, blocking and replication are also discussed as three basic principles of experimental design. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning material ...
Regression terms and output in Design Expert (v. 11)
zhlédnutí 3,5KPřed 5 lety
This video focuses on explainig important model performance measures (fit statistics), residual plots, and plots for influential observations in the experiment available in Design Expert, version 11. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning material for under...
On orthogonal designs and regression
zhlédnutí 4,8KPřed 5 lety
This video gives an introduction to the concept of ortogonal designs. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning material for undergraduate and graduate students following a course in design and analysis of experiments at LTU. The course book used in these cour...
Regression analysis and Design and Analysis of experiments
zhlédnutí 1,8KPřed 5 lety
This video is focused on explaining the close connection between multiple linear regression and the analysis of designed experiments. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning material for undergraduate and graduate students following a course in design and an...
Battery Design Example in Design Expert (v. 11)
zhlédnutí 1,5KPřed 5 lety
This video gives provides some basics in how to design and analyze a general factorial experiment in Design of Experiments. The battery design example comes from Example 5.1 on page 192 in the 8th ed. of D.C. Montgomery, Design and Analysis of Experiments. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary p...
Residual analysis in factorial design
zhlédnutí 1,9KPřed 5 lety
This video provides a short introduction to residual analysis in the analysis of factorial designs. The plots shown in the slides come from the software: Design Expert version 10. This video was recorded by Dr. Erik Vanhatalo, Quality technology and logistics, Luleå University of Technology (LTU), Sweden. The primary purpose of this video is to be learning material for undergraduate and graduat...
Effect sparsity, hierarchy and heredity principles
zhlédnutí 596Před 5 lety
Effect sparsity, hierarchy and heredity principles
Multipel regressionsanalys i Statgraphics
zhlédnutí 356Před 5 lety
Multipel regressionsanalys i Statgraphics
Försöksplanering (del 10) Robust design och försöksplanering
zhlédnutí 888Před 7 lety
Försöksplanering (del 10) Robust design och försöksplanering
Försöksplanering (del 9) Styrande principer vid analys
zhlédnutí 1,1KPřed 7 lety
Försöksplanering (del 9) Styrande principer vid analys
Försöksplanering (del 8) Exempel på reducerade försöksdesigner och överlagringar
zhlédnutí 2,3KPřed 7 lety
Försöksplanering (del 8) Exempel på reducerade försöksdesigner och överlagringar
Försöksplanering (del 7) Reducerade tvånivåers faktorförsök
zhlédnutí 2,1KPřed 7 lety
Försöksplanering (del 7) Reducerade tvånivåers faktorförsök
Sir.?
Bra beskrivet Erik, uppskattar verkligen dina videos om försöksplanering, det har verkligen hjälpt mig att sammanfatta viktiga delar. Jag har dock försökt klura ut länge nu hur formeln för standard avvikelse används vid 5.10 och framåt i videon. Jag skriver exempelvis in Rotenur (( (20,4 - 21,25)^2) / 2-1)) Men får då istället svaret till 0,85, får även helt andra svar på de övriga delförsöken där denna formeln har nyttjats. Är svaret 0,85 eller är det jag som inte riktigt slår rätt i miniräknaren?
Good morning from here... Please, I need assistance getting the setup file for a full version of the Design Expert...
Thanks for the informative video! But what if I have one point that exceeded the limit for DFFITS and DFBETAS?
Ok, then the design run may be influencial in terms of the model or the individual coefficients. One should at least go over the experimental log book and double-check that everything went fine with the experiment. These plots are tools to uncover influential design runs. Best wishes, Erik
@@erikvanhatalo7 Ok, thanks. I will double check. Just a bit unsure about these plots. Does it mean that in order for a design to be accepted, all runs must lie within the limit of these two plots?
Keep up the good work
Thanks for doing this.
Glad you found it valuable. 👍
Are the highest order interactions chosen to be confounded because they provide the least useful information?
I've been thinking about it, and I think info from higher order effects is less likely to be statistically significant and more difficult to interpret. Is that the reason why they are selected?
Yes, the fact that higher order interactions are less likely to accurr (a priori) makes them good candidates to "sacrifice" to blocks. Higher order interactions do occur but have been shown to occur much more seldom than main effects and low order interctions in real life. Best wishes! Erik
@@erikvanhatalo7 thanks
Tack Erik!
Good stuff Thank you sir
I am glad you liked it! :)
Thank you for this!!
Tack för videon! Mycket bra förklarat :)
What would be the value of K constant in response transformation?
The constant value k in the figures from the software is a constant that may be added to the ressponse value. For example, there may be instances if we have responses with value 0 (zero). Then the logarithm ln(0) is not defined and the conctant k can help in that problem.
It was extremely helpful sir :) Could you please help in analysis of combined array design in minitab or jmp
Hi! Thanks for the nice comment. I am not familiar enough with JMP or Minitab to know if they have some special functionality such as for the POE - propagation of error. But it should definitely be possible to do the analysis as a combined array approach and just keep track of which variables are "noise" variables and which are design variables. Sorry that I do not know the details. Best wishes, Erik
I am working on tuning NSGA-3 parameters (oil & gas industry) to optimiza one black box and must say your channel is a vault of great resources Sir.
Thanks! I am glad you find some useful videos here! It's a rather dry topic, but I like it :)
This was very useful, thanks!
I am glad it was of some value! 👍
Thank you so much Erik for wonderfull explanation
the pressure variables supposed to alternate 40 40 80 80 per the coded variable x2
Just spotted that too. Mistake in the book I think
Thanks Dennis, good of you to notice that! You are correct in that the pressure variable in the natural units does not match the coded pattern. This image is borrowed from the book companion slides so it is a typo in Montgomery's book. Nonetheless, the mistake was also mine in not noticing that. Best wishes, Erik
Thank you Sir Eric, I learned it in a very simple way. Thank you again :)
Thanks! Glad it was of value! 👍
Your videos are excellent Sir.
Very useful. Thanks a lot, sir.
Glad it was of value! 🙂
@@erikvanhatalo7 Dear sir, Very useful content for preparing a thesis.
Nice video
Thanks, cheers!
It was helpful, thanks
Thank you for the mountain analogy. You explained this very well
Thanks! Glad it was useful! :)
Nice video Erik
Hi Erik, Is there a way I can contact you e.g. by email? Thanks
Contact info: www.ltu.se/staff/e/erivan-1.81005?l=en
Thanks Erik
Great video Erik. How do you get the generators and the alias pattern for the full fold over?
Thanks! Do you mean the 2(7-4) design? after full fold over? Trying to remember from last year. I do not think it may be explained easily in this comment. But alias patters can be calculated from the complete design, by hand. In this case I think I took it from the Design Expert software.
@@erikvanhatalo7 Thanks for your reply Erik. I managed to figure out how to get the alias pattern for res III + full fold over. I’d like to ask two questions if I may 1. If you add a full fold over to a 2(3-1)III you get a 2^3 full factorial and you can analyse this using method of contrasts, Yates method or matrix methods. However, if we take the first 8 run res III (the 2(5-2)III) then add a fold over you don’t get a full factorial. What would you call this design? Does it even have a name? I can see how to calculate effects using the method of contrasts or matrix methods but I don’t think Yates method will work here. For 2 level full and fractional factorial designs Yates can be used to calculate effects but not for other designs (such as fractional factorial with full fold over added). Does what I’ve said make sense? Sorry for such a long rambling message.
i need pdf of this if u can .....
Hi! I will not provide pdf material of my teaching slides publically unfortunately. If you need to go in depth in this area I would recommend a textbook such as Montgomery's Design and analysis of experiments.
Kk sir
hello sir im doing project on this rsm method so i need ur help to understand all the basic concept of this process so i can give best presentation to the external
Please tell me about predicted equation in anova section ....where equation is written as A[2],B[2],.....and so on what is the meaning of 2
Hi! Since the factors are categorical these dummy variables A[1], A[2], etc are used to indicate effects for certain settings and combinations of the categorical factors. This is why it looks a bit strange.
Hello Sir, great video, thanks for this! A quick question. How would you regress the higher order term such as A^2, B^2 which are quite common in any response surface design? I researched a lot about this, but couldn't find eneough help.
Thanks! Glad you liked it. Now in order for you to fit second order terms, such as A^2, without confounding you typically need a second order design. You may want to chck out central composite designs or Box-Behnken designs. These are examples of common designs in response surface methodology. Cheers!
Great video. I wish there was a video that explained rotatability
Thanks! Glad you liked it. I never made a video just on rotatability alone. May come in the future. For the CCD you can make the design rotatable throuh the aloha value. Rotatable designs have equal prediction variance at all positions that have equal distance from the design center. Many times reasonable choice.
Thanks for your reply Erik. It’s the whole idea of why you’d pick a certain value and why a particular value of alpha gives equal prediction variance that I don’t understand. I don’t even understand what equal prediction variance means in the context of a certain response surface eg parabola, stationary ridge, rising ridge or saddle (col or minimax). Going to hit the books on this subject over Xmas