Skip to Content
Author's profile photo Craig S

QM/Characteristics/Specifications/Classification – Blog 3

QM/Characteristics/Specifications/Classification – Blog 3

Previous Blogs:

QM/Characteristics/Specifications/Classification – Blog 1

QM/Characteristics/Specifications/Classification – Blog 2

GC’s and MIC’s  – Linking

So now you have gone through with the client and you have a list of General Characteristics (GC’s) that you need along with the descriptions, UOM’s and decimal places. So what’s next for this master data? I hate to say “that depends” but yes.. it does..

Before we link the GC’s up with a master inspection characteristic (MIC) we need to determine spec ranges or values for the test.  My suggestion is that regardless of whatever design you ultimately decide on, you should at this point determine the maximum ranges you intend to allow for each test.  This is NOT a spec range even though that term is utilized by many.  These values should be the minimum and maximum values that the test can theoretically have. 

Tests measured in percentages are the easiest ones here.  You normally won’t ever report anything below zero or anything above 100. Yes, there are exceptions.  Going above 100% for a measured test is usually pretty rare but it happens.  Some other tests may exceed 100% but these are often calculated tests that involve measuring some type of change.  In any case, you want to establish the highest upper limit and lowest lower limit. These may or may not be similar to what SAP calls an upper or lower plausibility limit which is maintained in the MIC.  Often times however, the plausibility limits may be set to even narrower ranges at times.

For our examples here I will use a percent test and a PPM test.  We will stick pretty much to standard tests for now.  We’ll use 0 – 100% and 0 – 10,000 for ppm. Why 10,000 you say?  Just because in this case.  In reality the people responsible for testing will know what is reasonable.  Just because a test can have a large value doesn’t mean we need to allow it.  Our goal is to reduce errors at every step possible and minimize potential errors that could impact the customer. 

If a customer has a ppm test where they have never, ever, recorded a result over 500 ppm, you might set the upper range to 500.  Even though theoretically a result of 6000 ppm is physically possible. 

A quantitative characteristic will either be a one-sided or two-sided spec.  In a one-sided spec the value will be written as <= 100.0 or >= 100.0.  A two sided spec as 93.0 – 95.0.  Make sure you include spaces between the operand and the values. A key point to keep in mind is that even though SAP allows you to enter  < and > signs they are not honored when linked to a MIC.    SAP will interpret a < as <= and a > as a >= in the MIC.  So if you really want < 100.0, you must enter <= 99.9 so that in the MIC, 100.0 will be valuated as out of spec.

This can cause various problems when using decimals or rounding and in how values are reported.  For more info on this you can check out this SCN discussion:

When creating the general characteristic, the unit of measure (UOM), total characters, and number of decimal points must be provided.  The decimal point is a character and must be included in the count of characters.  100.0 would be counted as having 5 characters and one decimal space.

For our two characteristics we set up as follows.






Spec Range


Iron, ppm




<= 10000


Assay, %




0 – 100.0

We are now ready to create our general characteristics.   The above info is all you really need to create the general characteristics.  If you are building a spreadsheet for data loading you will need a Data Type field and other fields.  This isn’t a blog about data loading but is primarily about specifications so I’m not going to cover all the fields.  But if you are consulting, some of the fields you should understand and can be valuable to a client are:

Chars Group

Groups characteristics

Often done when multiple businesses share a client.  In which case it often relates to authorization group.

  1. Auth. Group

Limits editors for GC

Interval vals allowed

Allows range to be used

Primarily for one-sided specs.  Can be used for two-sided to put in a lowest detectable level value.

Negative Vals Allowed

Allows values < zero

Single Value

Default for GC’s.

MUST be used for all characteristics you want to link to MIC’s.

Entry required

Value must be recorded

Not normally used when linking to MIC’s

  1. Exp. Display

Allows entry and display in scientific notation

Not often used in most places

On the values screen.. DO NOT click on “additional values” on this screen.

Enter in the range you wish to allow for the test.  For the tests above, this would be <= 10000 and 0 – 100.0.

After making sure the characteristic is released, the new characteristic can be saved.

Qualitative characteristics might be discussed in a different blog since the goal of these blogs is to discuss options for handling spec ranges.  Since qualitative characteristics have no ranges and work by using catalogs, there can be no specs by material or customer.  So they really aren’t relevant to these discussions.

We should now have two general characteristics created.  The next blog will discuss setting up the MIC.

Assigned Tags

      You must be Logged on to comment or reply to a post.
      Author's profile photo Nitin Jinagal
      Nitin Jinagal

      Hello Mr Craig,

      Third blog of the series is now released. Good. I have gone through it once and liked the way you have precisely mentioned everything. I would say it is another value addition as most of won't be this much aware about the specification use.

      I would need spare time to run scenarios with the given details and I'm sure this would come pretty handy when using the specifications in above way. i.e. <= x and all.

      Another blog, nice... Awaited 🙂


      Author's profile photo Corne Ackermann
      Corne Ackermann

      Hi Craig

      I prefer not to use the <= or >= but rather - zzz.zzz where aaa is the minimum and zzz is the maximum allowed value.

      It makes it easier as there is only one standard way of laoding limits, etc. and values less than 0 cannot be accepted by mistake.




      Ps.  I cannot found the next chapter.

      Author's profile photo Craig S
      Craig S
      Blog Post Author

      Yeah.. I didn't get around to writing it.  I should.

      The issue with the use of aaa-zzz as you described is often the publication and definition of the specs. In food, chemicals and pharma this is critical.

      Lets say you have spec range of 67.0 to 69.0.   67 and 69 are both valid values. If I remember, unless you use <= and >=, the value or 67.0 and the value of 69.0 are not allowed.  Yet in most business the spec is inclusive of them.

      That technically would force a spec range of 66.9 to 69.1.  But if you require to print the spec on a COA or document, or need to match to the specs in plant procedures and methods, the spec would be incorrect.