Proper way to develop LCL when only given a lower spec limit

marcusja2002

Involved In Discussions
#1
Its been a while since I've done this, so I'm a little rusty.

I've attempting to develop a control chart with lower control limits and I believe I'm calculating everything correctly. This is all force data. There is a minimum force no maximum.

Because of the huge range difference from my max pulls I believe my Standard deviation is being thrown off and in turn making my control limits larger then they should be.

Is there supposed to be a different way I calculate lower control limits when there is only a minimum, or is this the correct / only way to do it, and in turn its telling me my process is out of control since the limits are calculating below the spec limit.

Nothing measured below the spec, but the calculated lower limit is well below the spec.

Just so numbers are available.

Spec min is 20. My lowest reading is 23, highest is 60 mean is 32 with a current stdev of 7.75.

My LCL is 8.96 and UCL is 55.44 (which I don't have a limit for).

If I remember correctly, doesn't this tell me I'm not only still in control under 20 ft lbs, but that statistically I could get a reading below 20?
 
Elsmar Forum Sponsor

Steve Prevette

Deming Disciple
Staff member
Super Moderator
#2
Calculating the Lower Control Limit (or the Upper Control Limit for that matter) is completely independent of the specification. As for the data may be highly skewed towards higher levels (such as with individual income data) one could consider using the Median moving range for the upper control limit (you should be able to find that if you google it, Davis Ballestracci is a proponent of it) or consider some sort of mathematical transformation (such as using Log Normal or Exponential) to deal with "normalizing" the upper tail.
 

Bev D

Heretical Statistician
Staff member
Super Moderator
#3
Can you post your data? And tell us how the samples are taken? (1 part each time? Several parts in a row?). This is the best way for us to help you
 

marcusja2002

Involved In Discussions
#4
1588613242205.png

It was taken in groups of 10, 4 separate operators with 2 separate tools set to the same parameters. Min force is 20, but if I'm calculating everything in theory I could pull below 20 at some point.

Am I just running the wrong calucaltions, should I actualy be running a weible analysis to figure out how likely I'm going to be able to stay above 20 ft lbs?
 

Bev D

Heretical Statistician
Staff member
Super Moderator
#5
When you say groups of 10 and 4 separate operators two tools was each piece tested only once or did each operator test each piece twice?
Were the groups of 10 randomly selected or were they taken in time sequence as they were produced?
 

marcusja2002

Involved In Discussions
#6
They produced 10 crimps which were placed in a pile, then were pulled out of the pile. I dont have minitab or anything like that, so I didnt create a random order for them. I just let them create and pull.
 

Miner

Forum Moderator
Staff member
Admin
#7
So these were in somewhat reverse order? Assuming the pile was not shuffled, the bottom of the pile were the first made and the top the last made then the last made were the first tested? So these were not in a perfect sequence, but there is some type of order to the data. It is not completely random. There does appear to be strata in the data, which may imply potential differences between operators and/or tools. However, you cannot verify this without linking the data to an operator and tool.
 

Bev D

Heretical Statistician
Staff member
Super Moderator
#8
A control chart is intended for data that is in time order of production. Without that you cannot create a usable control chart. If you also can't trace the parts to operator, tool etc. you cannot diagnose any non-homogeneous results and either rationally subgroup or improve the non-homogeneity/'out of control' variation.

There is an old saying about not being able to save a flawed study design with any amount of statistical wizardry. I think that applies here. My real recommendation is to redo the study maintaining time order and who/what was used. I think you can also reduce your subgroup size to 3 sequential parts, but take more subgroups spanning several lots of materials, operators and tools.
 
Thread starter Similar threads Forum Replies Date
K Proper document of SMPS used in infant warmer for IEC 60601-1 testing IEC 60601 - Medical Electrical Equipment Safety Standards Series 1
L How to write a proper defect description in a NCR Nonconformance and Corrective Action 4
M Proper method to test flow of a regulator Inspection, Prints (Drawings), Testing, Sampling and Related Topics 1
M Proper terms for describing Hardness General Measurement Device and Calibration Topics 5
M Proper ISO standard for Pressure Testing and Leak Rates Inspection, Prints (Drawings), Testing, Sampling and Related Topics 6
C Proper Thickness Gauge for Foam/Felt and Soft Materials General Measurement Device and Calibration Topics 5
D Proper Tool to Measure Wall Thickness (ISO 3611 - Micrometers) Manufacturing and Related Processes 7
X Proper Document Control Requests Document Control Systems, Procedures, Forms and Templates 1
M Proper Gauging - GD&T - Attribute gage to check true positioning on the shop floor Inspection, Prints (Drawings), Testing, Sampling and Related Topics 6
S Proper Record/Document Manipulation Document Control Systems, Procedures, Forms and Templates 2
I Proper gauging of tight tolerance (reamed) holes General Measurement Device and Calibration Topics 4
kedarg6500 Selection of proper modifying agent for Aluminium Silicon Alloy Manufacturing and Related Processes 2
J Proper Amount of Data Points for Cpk Capability, Accuracy and Stability - Processes, Machines, etc. 12
K Proper YTD PPM Supplier Calculation Six Sigma 3
S Proper FDA Classification for a DICOM Viewer 21 CFR Part 820 - US FDA Quality System Regulations (QSR) 29
E Measuring Parts without Proper Tools - Measurement Device Resolution Inspection, Prints (Drawings), Testing, Sampling and Related Topics 10
J Proper Method of Measuring Hex Socket Heads Inspection, Prints (Drawings), Testing, Sampling and Related Topics 4
B Getting the proper information from the supplier - Returning Nonconforming Material Nonconformance and Corrective Action 4
G What is the proper Thanks Giving turkey sampling size? Coffee Break and Water Cooler Discussions 9
B How to Determine the Proper Solder Class Manufacturing and Related Processes 5
S What is the proper flow to feedback the defective found during QA sampling check? Quality Tools, Improvement and Analysis 8
M Proper Format for a Manufacturing Route Sheet Manufacturing and Related Processes 6
K Putting a regulated device on the market without the proper ISO Certs and CE Marking EU Medical Device Regulations 9
L Proper Personnel Ratio - Quality to Manufacturing Manufacturing and Related Processes 6
M ISO14001 nonconformance - no proper treatment of wastewater prior to discharge ISO 14001:2015 Specific Discussions 6
A What is the proper definition for Class 1 & 2 design changes AS9100, IAQG 9100, Nadcap and related Aerospace Standards and Requirements 5
C Proper GD&T for spindle & housing Inspection, Prints (Drawings), Testing, Sampling and Related Topics 1
A Proper term for 'Easy Identification'? ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 8
M How to implement a proper Calibration System? General Measurement Device and Calibration Topics 5
F Is there a proper protocol to follow for Self Certification? ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 23
B True Position without tertiary datum? Establishing a proper rotation Inspection, Prints (Drawings), Testing, Sampling and Related Topics 12
J Proper way to calibrate a check fixture General Measurement Device and Calibration Topics 7
G Proper Poundage for Checking XX Tolerance Plain Plug Gages? General Measurement Device and Calibration Topics 2
Gert Sorensen Labeling Responsibility - Manufacturer vs. Repacker? Proper Component Use ISO 13485:2016 - Medical Device Quality Management Systems 7
L Procedure for proper use and protection of Registration and Accreditation Marks Document Control Systems, Procedures, Forms and Templates 21
C What is the proper method for cleaning dial calipers? General Measurement Device and Calibration Topics 4
N Proper use of the Blueing process when inspecting the taper of bushings General Measurement Device and Calibration Topics 6
J Proper use of gauge pins - Inspecting thru holes of plastic parts General Measurement Device and Calibration Topics 19
I What is the proper date to use as effective date? Old 'procedures', New documents Document Control Systems, Procedures, Forms and Templates 10
D Rockwell vs. Vickers vs. Knoop - Does Thickness Determine the Proper Test? General Measurement Device and Calibration Topics 1
M What is the proper way to use a GO NO/GO guage? General Measurement Device and Calibration Topics 3
B Preventive and Corrective Action Mapping - Proper Sequencing ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 4
T Registered or Certified - What is the proper terminology? QS-9000 - American Automotive Manufacturers Standard 17
O ISO 9001 new certification advice request - Develop a QMS from scratch ISO 9000, ISO 9001, and ISO 9004 Quality Management Systems Standards 9
Sidney Vianna NASA to Develop a Novel Approach for All-Electric Aircraft Using Cryogenic Liquid Hydrogen as Energy Storage World News 2
M IEC 62304 - Develop an Architecture for the Interfaces of Software Items IEC 62304 - Medical Device Software Life Cycle Processes 8
S Help develop sampling plan for prototype samples (medical devices) Statistical Analysis Tools, Techniques and SPC 7
G How to develop a strategy to eliminate an indicator who has been 0 forever? Preventive Action and Continuous Improvement 1
C Who should develop a LSS (Lean Six Sigma) project charter? Six Sigma 2
A How to Develop and Implement 5S +1 Program Lean in Manufacturing and Service Industries 3
Similar threads


















































Top Bottom