MSA (Measurement Systems Analysis) in Chemical Processing

C

C. Oertli

#1
this is my first time in a forum:
The purpose of this post is really to determine if anybody is intersted to discuss the particularities of the chemical manufacturing industry concerning application of MSA and if this post is conidered OT, to obtain some guidance or redirection.

We are a chemical bach processing business supplying high tech resins to various industries, including automotive. Through this relationship we have been introduced to MSA by customer demand. The merits this system brings to the manufacturing industry are obvious consequentially its application to a processing environment is tempting for guidance in the design of process improvement projects. For about a year I have been trying to adapt MSA principles to our industry. While some sucesses have already been shown, some significant differences in the application of statistical principles between the two industry branches remain unresolved. In the interest of brevity I will be synoptic.

Terminology: The translation of manufacturing terms into processing terms is ambiguous but necessary to design experiments. I found little guidance in the MSA literature. (For example I declared a batch as a 'part' after establishing that the batch is homogenous. The advantage is that destructive tests can be replicated by repetitive sampling).

Process Stability/statistical control: is much lower than manufacturing. Batch operations with short production campaigns (as low as three batches/campaign) where start-up transients do not provide a statisitcal stability comparable to the manunfacturing industry. Economy and ecology prohibits to discard the start-up 'parts' (up to $60,000 per batch) until process stability is reached. Process capabilities are much lower than customary in the maufacturing industry and hence the unquestioned adoption of such criteria may not be prudent. Consequentially it is customary to test all batches (parts) rather than a sample. This is effective for Quality Control to prenvent shipment of non-conforming material but process stability as required to predict process behaviors is limited.

Cost of chemical tetsting is high and significantly more complex than physical testing. Many tests are applications tests rather than by chemical analysis. Gage variation is comparably high.

The above is only a small sample of the differences I see with consequences in the interpretation of the output from a MSA/GageR&R study. As a somewhat critical scientist, I am reluctant to purchase a GageR&R stat pack, 'plug in the numbers' and use the output unquestioned in the justificaiotn of six figure improvement projects. For the time being I enjoy the support of management in the introduction of MSA principles in our operation, but is appears that a rational review of the statistical approach taken is in order.

Comments welcome and thanks in advance!

CO
 
Elsmar Forum Sponsor
#2
Maybe Atul will jump in here and assist as he is far more knowledgable in this area than I. However, I have just completed an MSA on an electronics measurement system, making adaptations to the process suggested in the AIAG MSA manual, as necessary to fit my system. Digital electronic meter accuracies and reading methods differ greatly from the old caliper and ruler measurements of size, around which the MSA manual appears to be written. Take a statistics course, arm yourself, then make more sense applying the numbers to YOUR processes than your customer can fathom. Process variation, efficiency, and cost improvements are the goal, not generating acceptable numbers!
 
D

Dave Strouse

#3
You are not unique

C . Oertli

Welcome to the forum.

One of the most respected names in statistical analysis (statistics types bow toward Wisconsin where he resides when his name is spoken, arguments end when he is cited), Dr. George Box; "grew up" in the chemical industry at Imperial Chenicals in UK.

Please get thyself to thy bookseller and get his book, "Statistics for Experimenters" Box, Hunter, and Hunter , 1978, Wiley ISBN 0-471-09315-7.

Read chapter 17 section three. He uses an example fom a chemical processing plant. The question to resolve is " What part of the observed variation is due to the batch and what to the analytic method. Unless I completely misunderstand your trouble, you should be able to adapt the method.

I've used somewhat similar methods for fibre optics manufacturing type problems.

Let us know how you progress.
 
C

C. Oertli

#4
Thanks for the literature reference.
As an aside, our management is already talking about 'Six Sigma' for which I see statistically responsible MSA as a prerequisite. During the time I have spent in the CMI, I have collected a number of anecdotes that clearly show that an adequate understanding of stability of systems is not widespread.
 

Atul Khandekar

Quite Involved in Discussions
#5
OK, I'll jump in, FWIW!

CO,

Welcome to the Cove!

I agree with what Icy and Dave suggested. Personally, I have no experience in R&R or statistical applications for Chemical analysis, so I am just thinking aloud (or in Text) here.

IMHO, standard procedures of R&R should apply to chemical testing as well. The key in any R&R study is the selection of samples and the amount of variation introduced (mainly) by the equipment & appraisers (and in this case also the batch), how the 'measurement process' is affected by gages and operators.

It is important to know what is being measured, and how is it being measured, also what knowledge you expect to gain in order to carry out process improvement programs.

- In a chemical process, it may not be possible to obtain repeat measurements if the gage is used during operations (in an ongoing process).
- In many cases there is no way an operator can influence the reading. So there is no reproducibility involved.
- Assuming that testing is offline, which I think is the case here, select 'parts' from a homogeneous batch, by splitting the samples if necessary.
- Large variability within the samples can affect process capability. If you know or suspect that this is the case, you can probably use the method described in the second edition of the MSA manual for within-part variation. This involves more tests. Also refer to Chapter V, Section A of third edition manual.
- If you treat a sample from a batch as a part, I think you'll have to use the Nested ANOVA or the Pooled Standard Deviation Approach to GRR (Chapter V, Section E). Dave has already mentioned an excellent reference book. I'd also suggest Douglas Montgomery.

You are right in that a rational review of the statistical approach taken is in order and that you should not blindly believe the output of a statistical software. A software is just a tool that helps you do complex calculations quickly.

Just my 2c. Excuse me if this sounds a little incoherent, but I'm still recovering from a holiday hangover. I'd be glad to discuss this further to learn how to conduct & interpret R&R in chemical testing.

- Atul.

Icy, thanks for that compliment- not sure if I derserve it. :confused:
 
C

C. Oertli

#7
Destructive Testing in CPI

Thanks for the valuable inputs,
One advantage the CPI appears to offer is that ample sample material is available for destructive testing by declaring a batch as a 'part'. The difference to the Manufacturing Industry is that we do not use the entire part for each test, but only a fraction of it. By accepting the 'homogenous batch assumption' along with the 'identical sampling assumption' we should be allowed to safely declare these samples as 'identical' regarding variations and therefore replications on the same part could be performed even for destructive test methods. The analytical chemist can repeat the destructive test on the same part as many times as requried by the experimental design. Therefore I think that the same GageRR principles for tests with replications or the crossed ANOVA may be permitted.
I am using the 'two-way, fixed-effects model with replications' provided by MS Excel, the terms crossed or nested ANOVA are new to me.
...but before going out on a limb any further I want to read the cited texts first.

C.O.
 
K

KMAAA

#8
C. Oertli,

I had to stop and think for a minute to be sure you & I are not employed at the same company....your situation sounds very similar to my own.

I too work at a chemical processing facility where we make polymers by various means. In my case the chemical testing is easy, fast, & cheap. There has been some "instrument capability" work done in the past though I can't verify it was done correctly, or even adequately...I would guess(hope) what has been done is a reasonable approximation of where we need to be. This very question has now filtered down for me to figure out as our entire Operations area is becoming more interested in GRR. In addition to my own longer term interests in setting up a formal GRR system, the interest has gained new support as our customers have begun asking for this information.

One conundrum I need to work through is the past instrument capability estimates have a couple of our key measures coming in at 35-40% "MCI". The traditional guidelines suggest this isn't so good & yet when you look at our process/quality-attribute data (by product, we have a very good & through SPC program in place for all attributes, of all products, from all reactors) the overall process capability indicates we're quite capable at Cpk & Ppk of 1 or more. Mix this observation with our need to qualify new chemical analysis methods & those that develop the methods rightfully ask, "if test X "isn't capable" and yet the products where we use test X have capable processes, why do new test methods have to meet the <30% criteria?" I've just started digging into this, but so far I don't have a good answer.

I'd be interested in what you come up with, please post back as make progress. I'll try to do the same.
 
D

desqae

#9
Hi Folks

I'm facing same dilemma to answering customer in the CpK and MSA study. Do you have sample that can mail to me or post to the forum. 1 day we only produce 2 batch and chemical itself quite sensitive to the heat. pH or viscocity will be varies when chemical temperature raise up or down. Thanks

Desmond
:cool:
 
D

DsqrdDGD909

#10
Desmond,

I'm a little unclear on the question. Can you describe your process, product and analysis a little more?

I think CpK is often misinterpreted for batch processes, but MSAs are very doable in Chem analysis.
 
Thread starter Similar threads Forum Replies Date
A Non-replicable tests in MSA (Measurement Systems Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
W Does TS16949 require MSA on all measurement systems in the Control Plan? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
G Interpreting Type 1 MSA (Measurement Systems Analysis) Results Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
T MSA on Camera Vision System and other Process Measurement Systems Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
P MSA for Counting Measurement Systems Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
A MSA (Measurement Systems Analysis) for Humidity Machine Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
SATHYABALARAMAN MSA (Measurement Systems Analysis) Study plan Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
I Interpretation of MSA (Measurement Systems Analysis) Linearity and Bias Study Results Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
C MSA (Measurement Systems Analysis) Variables Template Document Control Systems, Procedures, Forms and Templates 0
A Measurement Systems Analysis (MSA) - Who Developed it? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
R MSA (Measurement Systems Analysis) on a CMM for Special Characteristics Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
D MSA (Measurement Systems Analysis) on Visual Color Check IATF 16949 - Automotive Quality Systems Standard 13
E MSA for criteria Automatic Video Measurement Systems Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
G Ford - Type 1 Gage Study MSA (Measurement Systems Analysis) acceptable? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 7
D Calibration of Load Cells - ISO 17025 & MSA (Measurement Systems Analysis) ISO 17025 related Discussions 5
G Traceability Studies and Requirements for MSA (Measurement Systems Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
B Nonconformance about MSA (Measurement Systems Analysis) in ISO/TS16949:2009 IATF 16949 - Automotive Quality Systems Standard 9
C MSA (Measurement Systems Analysis) Case - A little help? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
I Measurement Systems Analysis (MSA) for Trading Industry Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
L MSA (Measurement Systems Analysis) Process template - ISO TS 16949 Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
D Any MSA (Measurement Systems Analysis) classes in the Chicago area? Training - Internal, External, Online and Distance Learning 1
kedarg6500 MSA (Measurement Systems Analysis) Fourth Edition Excel Sheet wanted Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
Marc Alan Miner's Measurement Systems Analysis (MSA) Blog Article Series Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 12
N MSA (Measurement Systems Analysis) for Surface Finish Equipment Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
S AIAG MSA (Measurement Systems Analysis) Manual 3rd ed. Chapter IV Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 6
R MSA (Measurement Systems Analysis) for Idiots - Software Quality Assurance and Compliance Software Tools and Solutions 6
Marc MSA (Measurement Systems Analysis) and Gage R&R Spreadsheets Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 0
B MSA (Measurement Systems Analysis) with One Operator using a CMM or Common Calipers Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
Z MSA (Measurement Systems Analysis) for gages used in final inspection Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
Q MSA (Measurement Systems Analysis) Internal Audit checklist vs. TS 16949 Internal Auditing 4
B MSA (Measurement Systems Analysis) on a Video Microscope Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
K MSA - Is Measurement Systems Analysis required in R&D? TS 16949 Clause 7.6.1 Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 14
U MSA Roadmap - Is there an order in which to evaluate a measurement systems Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
P MSA (Measurement Systems Analysis) Gage R&R with 10 Operators Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 6
B Information about MSA (measurement systems analysis) software Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 6
R MSA (Measurement Systems Analysis) - Where to begin when setting up a system? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 14
S Fish Oil Capsule Weight MSA (Measurement Systems Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 9
A How to select equipment for MSA (Measurement Systems Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 9
H Is MSA (Measurement Systems Analysis) only required for SC characteristics? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
J ANSI Bar Code Validator MSA (Measurement Systems Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
B Automated Vision Inspection System MSA (Measurement Systems Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
R Format / Template of all MSA (Measurement Systems Analysis) studied Document Control Systems, Procedures, Forms and Templates 3
W Attribute MSA (Measurement Systems Analysis) for Low Volumes Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 7
H When to conduct a MSA (Measurement Systems Analysis) study Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 5
U How to do MSA (Measurement Systems Analysis) - I need to know the basics Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 2
R MSA (Measurement Systems Analysis) for Visual Standards? IATF 16949 - Automotive Quality Systems Standard 3
P MSA (Measurement Systems Analysis) for pneumatic torque gun Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 1
Y Measurement Systems Analysis - Effect on Process Decisions - MSA Manual page 19 Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 4
B MSA (Measurement Systems Analysis) Positional Tolerances Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3
A How do i start implement MSA (Measurement Systems Analysis) Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 3

Similar threads

Top Bottom