Design of Experiment (DOE) for Heat Treatment Process

B

Bill Rudnicki

#1
Good morning;

We are going to be working on a DOE to help set the process parameters for a heat treat process. Currently, the heat treat results are taken 3 times; at both ends and in the middle of a piece of track. How should I handle the results for entry? Should I average the results and enter them or Should I run the analysis three times in each of the locations? Thanks for your help.

Bill
 
Elsmar Forum Sponsor
J

justncredible

#2
Each of the spots tested should be checked, also I would use the 3 sample size in a ol'fashoined chart. That way when you compare the diffrent settings you can see if it effected the entire process or one spot. But I am sure there are other ways.
 

Tim Folkerts

Trusted Information Resource
#3
To me, the first question should always be "what do I want to learn?"
The second question - close on its heels - is "how tough/expensive is it to do experiments".

Is it important to know just the average result of the heat treatment?
. . . Then just study the average.
Is it important to reach a certain minimum?
. . . Then you could study the minimum value of the set.
Is uniformity important?
. . . Then you could study the range or St Dev.

Since you will already have run the experiments and will have the results at each of the three spots, I would go ahead and run the analysis for each of the three spots separately, and perhaps for the range as well.

Some programs have a way to optimize several results (I know minitab does). So you could, for example, have the software look for values that would minimize the range while maximizing some result. Or find values that would maximize all three results. or ....


Tim F
 
B

Bill Rudnicki

#4
Thanks for your responses. I do have Minitab which includes the process optimizer which should make things easier.
 

Bev D

Heretical Statistician
Leader
Super Moderator
#5
justncredible has a good point that many tend to forget now that we have computers and statistical software like Minitab or JMP.

Leonard Seder got it right in middle of the last century and Dr. Deming and Dr. Ott also carried the flag: plot your data first and look at it. the visual display of your data will much more compelling than any 'blind' statistical output. it will also tell you which extra questions to ask.

Seder published a paper (published in 2 parts) on Multi-Vari. Get yourself a copy of it (and "the technique of experimenting in the factory" while you're at it). These papers are very quick and easy reads (deceptively simple given their powerful effect) and if you're a member of ASQ they are only $5 each ($10 if not) and worth every penny.

In the old days we took pencil to paper and plotted our data. Now I woudln't trade today's software for it's ability to rapidly plot huge data sets and calculate complex formulas - but I still plot and look at my data first.
As Yogi Berra says: It's amazing the things you can see when you look!
 
Thread starter Similar threads Forum Replies Date
M Design of Experiment (DOE) Help Needed Quality Tools, Improvement and Analysis 8
S DOE (Design of Experiment) with Several Output Variables Quality Tools, Improvement and Analysis 4
T Comparative Design of Experiment to pick better Trimming Blade Supplier Six Sigma 5
F Resolution III Factorial Design for a Screening Experiment Six Sigma 2
B Design of Experiment - 24 Samples per Run Statistical Analysis Tools, Techniques and SPC 16
P Design an experiment to optimize the PCB solder reflow profile Design and Development of Products and Processes 4
X Design stage overview (Product specification) EU Medical Device Regulations 3
gohyl Fixture Setup Design Manufacturing and Related Processes 3
D IS0 13485 - Design perspective to regulatory requirement. ISO 13485:2016 - Medical Device Quality Management Systems 2
T Software item classification and Detailed Design IEC 62304 - Medical Device Software Life Cycle Processes 4
armani 7.1.5 and design and development of product ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 10
Sidney Vianna ISO 9001 News ISO 9001 Design Specification to be developed - May 2022 ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 12
C Design output vs acceptance criteria ISO 13485:2016 - Medical Device Quality Management Systems 8
M Design Control Procedure Medical Device and FDA Regulations and Standards News 4
lisap Design falling outside ISO scope Misc. Quality Assurance and Business Systems Related Topics 11
S Link Between Essential Performance Requirements and Essential Design Outputs 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 3
G Calculating Ppk for Design Verification - Variable Sampling Design and Development of Products and Processes 15
G How many Design FMEAs do you have? FMEA and Control Plans 4
G Updating FMEAs after Design Updates FMEA and Control Plans 4
Stoic Details of Operational Qualification (OQ) test design for plastic extrusion processes, effect of material property noise, and GHTF/SG3/N99-10 US Medical Device Regulations 2
H R&D and Design procedures - can one procedure cover both activities Design and Development of Products and Processes 6
J Design Verification Testing and Statistics Reliability Analysis - Predictions, Testing and Standards 3
K Can I make an exclusion of Design and Development in ISO 13485:2016 if my product is not regulated ISO 13485:2016 - Medical Device Quality Management Systems 12
S Audit Finding - Design History File (DHF) Index: few (3 to 4) reports not identified ISO 13485:2016 - Medical Device Quality Management Systems 3
S Is the Design Service Provider required to be ISO 13485 certified? ISO 13485:2016 - Medical Device Quality Management Systems 13
M How to show the effect of the failure mode on the manufacturing process as a customer of product design process? FMEA and Control Plans 3
C MDR - Annex II _3. DESIGN AND MANUFACTURING INFORMATION_a EU Medical Device Regulations 3
C Items used for Design Verification Design and Development of Products and Processes 7
S Software design document NMPA guidance and consultant China Medical Device Regulations 4
L Corrective and Preventive Actions aligned to design related defects. After Work and Weekend Discussion Topics 5
E DESIGN VALIDATION, USABILITY AND CLINICAL EVALUATION request Medical Device and FDA Regulations and Standards News 0
Watchcat REGULATORY WATCHCAT De Novo Post-Mortem 4 – Design Controls Other US Medical Device Regulations 0
D Finding Optimum Design Parameters using Taguchi method? Using Minitab Software 2
S ISO13485:2016 7.3.9 design and development change ISO 13485:2016 - Medical Device Quality Management Systems 3
J Managing design verification regression testing of design changes Design and Development of Products and Processes 1
K Design History File - Looking for Examples of DHF's. 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 5
Watchcat Software validation vs design V&V? Other US Medical Device Regulations 27
S Design & Development records for Medical Devices Packaging and Labelling ISO 13485:2016 - Medical Device Quality Management Systems 9
D Using electronic lab notebooks in the design control process ISO 13485:2016 - Medical Device Quality Management Systems 3
L Essential Design Outputs for SAMD 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 2
T Justification of "Design Excluded" in IATF 16949 Design and Development of Products and Processes 5
H Contract Manufacturer as Design Owner ISO 13485 ISO 13485:2016 - Medical Device Quality Management Systems 6
B Drawings - Design Rules Inspection, Prints (Drawings), Testing, Sampling and Related Topics 26
R Role of QA during Design and Development of Products and Processes Design and Development of Products and Processes 4
B Design Responsibilities for Mature Acquired Product Lines ISO 13485:2016 - Medical Device Quality Management Systems 3
I RWE Studies and Design Controls US Food and Drug Administration (FDA) 1
G How to record Regulations and Standards as Design Inputs? Design and Development of Products and Processes 15
T Design Traceability - Labelling / Instructions for Use ISO 13485:2016 - Medical Device Quality Management Systems 5
B How to exclude empty rows in full factorial design analysis? Using Minitab Software 0
M Handling design changes with use-as-is inventory disposition Quality Manager and Management Related Issues 6

Similar threads

Top Bottom