Spec Limits in SPC charts

JasnahKholin

Starting to get Involved
#11
If it matters, this is semiconductor fabrication. Distributions on most features are pretty tight and all processes are done by machine so there isn't a lot of variation which could be attributed to operator skill or human error. I use the example of hole diameter earlier because the majority of my background is machining.
 
Elsmar Forum Sponsor

Beth Savage

Helping organizations optimize quality
PQ Systems
#14
One argument against adding spec lines to an individuals chart is that the focus changes from process stability and predictive signals to simply acceptability or not. If you do choose to include spec lines on your control charts, make sure that the lines are clearly distinct from the control limits to avoid confusion. Also make sure that anyone who works with the charts or views them has been trained on the difference between the control limits and the specification limits. This applies to management as well as (if not more than) operators.
 

Steve Prevette

Deming Disciple
Staff member
Super Moderator
#15
I many times have added specification (or goal) lines to SPC charts. The Control Limits are calculated at the three standard deviation level based solely upon the data and identified as such. The specification line(s) help you visualize if your process is stable THEN IS IT [original edited] in spec, or out of spec. If there is a trend, is it heading towards or away from the goal/spec.

Now, there are some that propose using the specification limits (or some mathematical interpretation of them) as the "control limits" and from most SPC references, that would indeed be inappropriate.

Sentence above originally read - The specification line(s) help you visualize if your process is stable - is it also stable and in spec, or out of spec. I have modified it due to Beth Savage's valid concern with that sentence. Original intent was to imply "If your process is stable - then you can easily check it against the specification".
 
Last edited:

Beth Savage

Helping organizations optimize quality
PQ Systems
#16
The specification line(s) help you visualize if your process is stable - is it also stable and in spec, or out of spec.
How would you use specification lines to help visualize whether or not a process is stable? The data for spec lines come from the customer, management, or engineering to indicate the allowable spread. Spec limits do not come from the process, so how can they communicate anything about the stability of the process? Control limits help assess process stability because the data is from the process.
 

Steve Prevette

Deming Disciple
Staff member
Super Moderator
#17
How would you use specification lines to help visualize whether or not a process is stable? The data for spec lines come from the customer, management, or engineering to indicate the allowable spread. Spec limits do not come from the process, so how can they communicate anything about the stability of the process? Control limits help assess process stability because the data is from the process.
Indeed, the spec lines come from the customer, or regulations or management's desires to improve (and what takes priority to improve).

BUT we must separate the functions of the specification limits and the control. limits. The control limits tell me if the process is in control or not. The specification limits tell me if I am meeting specs. This gives us four possibilities, as compared to Dr. Deming's Red Bead Experiment:

1. The process is stable and not meeting specs. This is the usual state for the red beads - the average is 10 red beads, the lower control limit 1 and the upper control limit is 19. I plot the stated goal of 3 (which yes may be from pull it out of your you know what) and see the UCL and the average is above 3, so I am never going to meet the goal of 3 without a process change.

2. Let us say we remove some of the red beads from the bucket. In this case, it is "easy" to remove the beads, in reality, it will take some work to identify the proper changes. We get an improving trend - let us say 7 below average (or 8, or 9 depending on your author). We have a process not yet meeting the spec, but we have imparted momentum - we have an improving trend. Not stable, not meeting the spec, but heading in the right direction.

3. Let us say we removed enough beads to have the UCL be less than 3. And we have enough data and knowledge to declare the process stable. Now we have a stable process (GOOD) and a process that is consistently meeting specification (DOUBLE GOOD). We want to stay here. Though should hold out that management could yet decide that less than 3 is not good enough, and desires to lower the spec to zero. Then we go back to step 2..

4. Now let us say (this is a real story from the desert of Washington state) it is a dry day, low humidity, and all red beads are attracted to the paddle and the white beads are repelled. We now exceed the upper control limit established in step 3, perhaps or perhaps not exceeding the specification. I now have an Unstable Process which is heading the wrong direction against the Specification. VERY BAD. But I find a can of anti-cling spray and spray down the beads and all returns to normal (step 3).

Attached is a slide from a publicly-cleared presentation of a real example. Notice the lower right hand chart illustrates an SPC chart with a one sided specification (the dotted green line) and we can see the trend of performance getting worse with time, which we were taking action on (and some of it was being more aware of reporting events). Note the Pareto Chart in the lower left to identify the source of events.
I will admit the stoplight in the upper left corner was less useful, it was just based upon days since last event, not the trend.
 

Attachments

Last edited:

Beth Savage

Helping organizations optimize quality
PQ Systems
#19
Thanks for keeping the red bead experiment alive and available for all to view and learn!

The only statement I take issue with is "specification lines help you visualize if your process is stable." I agree with all of your other statements.

Control limits help you visualize whether your process is stable but spec lines do not. Spec lines do provide information, but they are not connected in any way to process stability. I often encounter folks who confuse spec lines and control limits so I believe this is an important distinction for the OP and others who may follow this thread.
 
Thread starter Similar threads Forum Replies Date
H Johnson Transformation in Minitab - How upper & lower spec limits are recalculated Using Minitab Software 1
B If I know x-bar, spec limits, and Cpk, can I calculate the control limits? Capability, Accuracy and Stability - Processes, Machines, etc. 16
J API Spec 6D 25th edition release date? Oil and Gas Industry Standards and Regulations 2
T API SPEC Q1 Clause 5.6.1.2.C).2) First Article Oil and Gas Industry Standards and Regulations 16
C Validation of process for production and servicing 5.7.1.5 API Spec Q1, 9th Edition Oil and Gas Industry Standards and Regulations 9
lanley liao How to understand correctly the section of 5.11 MOC in API Spec Q1? Oil and Gas Industry Standards and Regulations 3
U API Spec Q1 - 5.6.1.2 C (3) - Design software Oil and Gas Industry Standards and Regulations 3
lanley liao How to correctly understand the bullet list d) of 6.3 Analysis of Data for API Spec Q1 Oil and Gas Industry Standards and Regulations 7
A API Spec Q1 Purchasing Process - Supplier Reevaluation based on Supplier Risks 5.6.1.4 Oil and Gas Industry Standards and Regulations 17
lanley liao The validity of API Spec Q1 certificate Oil and Gas Industry Standards and Regulations 2
A Design and development procedure for API Spec Q2 Oil and Gas Industry Standards and Regulations 6
Q Capability study with a minimum spec Statistical Analysis Tools, Techniques and SPC 8
S API Spec.Q1 Standards Version in Purchasing Information Oil and Gas Industry Standards and Regulations 1
S API Spec. Q1 clause 5.6.1.2 On site evaluation Oil and Gas Industry Standards and Regulations 10
S API spec Q1 9th edition - Castings & Forgings Oil and Gas Industry Standards and Regulations 4
N API spec Q1 9th edition Oil and Gas Industry Standards and Regulations 2
lanley liao How to understand the term 3.1.21 'Servicing' and the clause 5.7.1.2 'Servicing' of API Spec Q1? Oil and Gas Industry Standards and Regulations 19
A Does anyone have a checklist of API Spec 650 13th Edition? Oil and Gas Industry Standards and Regulations 0
V Thread Plug Gage Pitch GO Diameter out of spec AQL - Acceptable Quality Level 8
M API Spec. Q1, 8th Ed. Section 7.5.2 vs. Q1, 9th Ed. Section 5.7.1.5 ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 0
M Proper way to develop LCL when only given a lower spec limit Reliability Analysis - Predictions, Testing and Standards 7
N Audit non-compliance - API Spec Q1 9th Ed 5.6.1.2 b Oil and Gas Industry Standards and Regulations 25
B Set RH spec from dewpoint spec General Measurement Device and Calibration Topics 2
A API Spec Q1 9th Edition # Re-audit - Disputing with API AARs Oil and Gas Industry Standards and Regulations 1
R Process control documents - Context of API Spec Q1, Clause 5.7.1.3 Other ISO and International Standards and European Regulations 0
G Can calibrated device values be used instead of manufacturers spec? General Measurement Device and Calibration Topics 2
DuncanGibbons OEM specifications such as AIPS spec from Airbus AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 3
S Internal Auditing for API Spec Q1 - auditor qualification requirements Oil and Gas Industry Standards and Regulations 6
S API Spec Q1 - How to define Management Representative competency for QMS Oil and Gas Industry Standards and Regulations 12
R Interesting Discussion API Spec Q1 Addendum 2 Oil and Gas Industry Standards and Regulations 0
F API Spec Q1 Scope - Limiting the scope of certification Oil and Gas Industry Standards and Regulations 3
F API Spec Q1 9th Edition Surveillance Audit - Questions about internal audits. Oil and Gas Industry Standards and Regulations 22
N When to record values if a spec for a mfg process is set Manufacturing and Related Processes 3
M AMS0320 - Electro-deposition of zinc - Anyone come across this spec? AS9100, IAQG, NADCAP and Aerospace related Standards and Requirements 4
A API Spec Q1 9th Edition - 12 month Internal Audit Schedule Audit Nonconformance Oil and Gas Industry Standards and Regulations 10
D What is your Sterile Medical Device Pouch Peel Strength Spec? Other Medical Device Related Standards 3
P Any official doc/spec on different types of Yield calculation? Manufacturing and Related Processes 0
A Capability Analysis for Packaging Seal Strength with spec. >0.1 Kgf using Minitab Using Minitab Software 6
V Plating Spec - Electro plating Manufacturing and Related Processes 3
Moumen H API Spec Q2 Design and Development Requirements for Service Providers Oil and Gas Industry Standards and Regulations 11
A Integrating API Spec Q1 with ISO 9001:2015 Oil and Gas Industry Standards and Regulations 18
S API SPEC 7-1, 7-2 and 5 CT implementation Oil and Gas Industry Standards and Regulations 3
W Shared Responsibility Between Spec Provider and Contract Manufacturer 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 3
K APQP requirement when a new print or spec is received from a customer - Existing Part APQP and PPAP 3
S CpK in Minitab when 0.0 is the lower spec limit Statistical Analysis Tools, Techniques and SPC 7
D US Laboratory which can do Bosch test F 02N P00 089 Contamination Spec VDA Standards - Germany's Automotive Standards 4
W Is the GR&R result caused by the wide spec range? Gage R&R (GR&R) and MSA (Measurement Systems Analysis) 7
A Cosmetic Spec - Convincing Management Visual Inspection Standard is Necessary Manufacturing and Related Processes 5
K Black Anodize Color Specification per MIL Spec Inspection, Prints (Drawings), Testing, Sampling and Related Topics 1
H API Spec Q1 7.5.2 & 7.5.2.1 - Audit Nonconformance Help Oil and Gas Industry Standards and Regulations 15

Similar threads

Top Bottom