Product Engineering - Quality Management

Design Detail : Acceptable Quality Levels

STAEDEAN Quality management offers the internationally recognized Acceptable Quality Levels method as a Sampling method. For each measurement the Inspection ? and Quality level can be set.
This, statistical, method assumes that the inspection results of the sample can be applied to the entire product order. There are many statistical AQL Sampling Plans which are based on different situations like: inspections of lots or production inspection. Some plans allow for reduction of inspection when defect levels decrease or more strict inspection when defects increase.
The STAEDEAN Quality Management AQL is a partial implementation of ISO02859 as well as related international standards like: ANSI/ASQC Z1.4, NF06-022, BS 6001, DIN 40080. It allows the user to select inspection and quality levels.
This standard is also referred to as the "attributes method of inspection". A, randomly selected sample is examined on a number of pre-defined characteristics. The action to be taken is decided by counting the number of non-conforming items in the sample.
An alternative method would be the "variables method". This method selects smaller sample sizes compared to the attributes method. But allows zero defects on the sample. The sample size depends on the combination of Inspection - and Acceptance levels. This method conforms to the BS 6002 standard in accordance with the procedures of ISO 3951 or ISO 8423.

  1. Select the AQL plan with the person responsible: agree with your vendor on the levels you set for their product, or agree with the production manager against which levels you will inspect output.
  2. Set the selected parameters, inspection level and quality level in the measurement template for the items.
  3. When the lot is received through, for example a purchase ? or production order the Measurement will show the sample size as per the parameters in the measurement template.
  4. Randomly select the parts from the lot
  5. Inspect each sample based on the check sheet and record in the findings in the check sheet to determine accepted and rejected samples.
  6. Accept or Reject the remaining part as per the measurement.

To set the AQL levels

  1. Determine the Inspection Level, there are 7 levels split in 2 groups:
  2. General Inspection Levels : I, II, III
    The normal inspection level where III is more strict, leading to a larger sample size
  3. Special Inspection Levels : S1, S2, S3, S4
    A more relaxed inspection level where S1 leads to the smallest sample size

A. INSPECTION LEVELS

Level I II III S1 S2 S3 S4
Lot Size
1 - 8 A A B A A A A
9 - 15 A B C A A A A
16 - 25 B C D A A B B
26 - 50 C D E A B B C
51 - 90 C E F B B C C
91 - 150 D F G B C D E
151 - 280 E G H B C D E
281 - 500 F H J C C E F
501 - 1.200 G J K C D E G
1.201 - 3.200 H K L C D E G
3.201 - 10.000 J L M C D F G
10.001 - 35.000 K M N C D F H
35.001 - 150.000 L N P D E G J
150.001 - 500.000 M P Q D E G J
500.001 N Q R D E H K
  1. Determine the Quality Level represented by a number: 1.0; 1.5; 2.5; 4.0; 6.5 With the Sample Code found in table A select the sample size in table B as well as the acceptable number of rejects in the acceptable quality level column.

B. ACCEPTABLE QUANTITIES

Level Sample 0.4 0.65 1.0 1.5 2.5 4.0 6.5
Code letter
A 2 0 0 0 0 0 0 0
B 3 0 0 0 0 0 0 1
C 5 0 0 0 0 0 1 1
D 8 0 0 0 0 1 1 1
E 13 0 0 0 1 1 1 2
F 20 0 0 1 1 1 2 3
G 32 0 0 1 1 2 3 5
H 50 0 1 1 2 3 5 7
J 80 0 1 2 3 5 7 10
K 125 1 2 3 5 7 10 14
L 200 2 3 5 7 10 14 21
M 315 3 4 7 10 14 21 21
N 500 5 6 10 14 21 21 21
O 800 7 8 14 21 21 21 21
p 1250 10 11 21 21 21 21 21
Q 2000 14 15 21 21 21 21 21

Increase - or Decrease AQL levels

To make the AQL, and Zero Defects, levels dynamic use the Skip Level functionality. For each of the Skip Patterns you can set per pattern line the Inspection - and Quality Level. So when the level increases, decrease the inspection/quality level. When the skip goes back to a rejected level the inspection/quality levels automatically increases again.

Quality Management