Drawing of Sample Devices 
The sample devices and spare sample devices are drawn from a lot using random numbers . These devices are then removed and inspected in order to assess the quality of the entire lot. Devices can remain installed for inspection if the Installed devs field (certify installed devices) is selected in the device.
You can determine:
Whether warehouse devices are to be taken into account
This is possible for the total lot procedure at the level of an individual lot or at drawing level.
Whether you require a single drawing or a double drawing
You can only determine this before the first drawing.
If a certain number of devices fails in the first of a double drawing, the devices determined in the second drawing are used. The number is specified by the calibration authority in the case of an official lot and by the utility company in the case of an internal lot. In the case of an official lot, this number determines the calibration authority. In the case of a company lot, it determines the utility company.
During lot device determination all lot devices receive an internal identification number . If the internal identification number of the device corresponds to the random number, this device is drawn as a sample device. Once the fixed number of sample devices has been determined, the spare sample devices are selected in the same way.
The following legal regulations apply to drawing:
Number of devices to be drawn
In Germany the calibration regulation of August 12, 1988 applies (Federal Law Gazette 1988, no. 43, p. 1657ff) §14.
Random number table
In Germany DIN (German Industrial Standard) 57 418 Part 6/VDE 0418 Part 6, June 1983 applies.
Algorithm for determining the random numbers
A maximum of two drawings per year
At the end of the drawing, a log is displayed that you should read line by line. It shows the stages of the drawing and a list of the sample devices and spare sample devices. The drawing is not completed until the log is saved. The devices are then selected in their corresponding master records as sample devices. The log can be printed as proof for the calibration authority.
If you want to use your own algorithm to draw sample devices you can do this by means of BAPI BAPI_ISULOT_INS_SAMPLE_DEVICES .
In Customizing for Device Management the number of sample devices and spare sample devices to be drawn must be defined under .
This number is dependent on the following parameters:
Division
Indicator Single sample lot drawing
Type of certification requirement
Type of sample device (sample device or spare sample device)
Lot size
To carry out a drawing, you need to perform the following activities:
Create lot
Create a lot master record that contains the lot data.
Compilation of lot devices
During this process, the lot number is transferred to the master record of the devices allocated to the lot.
Determination of lot devices
In this process, all devices belonging to the lot are determined using the lot number entered in the device. They are given a consecutive internal identification number and stored in a table. This table serves as a basis for future drawings. The table contents can be printed in the form of a list for the calibration authority.
The random number table consists of 50 lines (j) and 50 columns (i), in which the numbers 0 to 9 are arranged randomly . To display the table, select . The numbers are arranged in pairs so they can be read more easily.
The table is accessed until all required random numbers have been determined. A random number does not produce a sample device or a spare sample device if:
The random number is larger than the lot size
The random number is zero
The random number has already been determined
The device has already been a sample device or a spare sample device (this can be changed via customizing)
The device is in the warehouse (if warehouse devices are not permitted during the drawing)
The device is not allocated to the lot (due to a removal, for example)
The device does not exist
The device cannot be processed during the drawing (due to changes to the device by a different transaction)
The first time you access the table is referred to as initial access and is used to determine the first random number. It consists of a starting line j and as many starting columns i as there are digits in the lot size. You can change this before determination of the first random number.
Initial access is proposed automatically for every drawing and is determined based on the current time. Based on the minute as the starting column and the second as the starting line, two-digit random numbers are determined from the random number table and used as starting lines or starting columns.
The following special features apply.
If the current time produces a first starting line or starting column that is higher than 50, that number is subtracted from 100 and the difference is used.
If the current time produces a first starting line or starting column of 0, the starting line or column 1 is used.
The two-digit numbers for the starting line or starting column are then found based on the starting line and starting column determined by the current time. The following conditions apply:
If the first decimal place of a starting number is greater than four, it is ignored and the next random number is determined.
If a number is repeated during generation of the starting columns, it is rejected and a further random number is determined.
If two drawings have the same starting numbers, they will also have the same random numbers. This allows you to repeat a drawing, for example, for devices that cannot be processed in the short-term (these devices are contained in the log). In this case, the drawing is not saved and the transaction is called again so that the device can be marked as a sample device when it can be processed again.
The first random number is found where the starting line meets the starting columns. The start access is saved with the drawing date in the master record of the corresponding lot so that information about the inspection is available at any time.
The next random number is determined from the point where the line j + 1 meets the predetermined starting columns. If the line becomes j = 50 + 1, it is set back to 1. However, each of the predetermined starting columns is simultaneously increased by 1. If, as a result of this process, a column becomes i = 50 + 1, then it is set back to 1. Then line 2, line 3, and so on are accessed using the starting columns previously increased by 1.
Example
For a lot with 250 devices, a start access was randomly determined with the starting line j = 2 and the starting columns i = 1, 4, 3.
Excerpt from the DIN random number table:
Column i |
|||||
Line j |
1 |
2 |
3 |
4 |
5 |
1 |
2 |
2 |
1 |
7 |
6 |
2 |
1 |
9 |
3 |
6 |
2 |
3 |
1 |
6 |
7 |
7 |
2 |
4 |
7 |
8 |
4 |
3 |
7 |
From this table, the first random number determined is 163, the second is 177, the third is 734, etc. This third random number does not produce a sample device since the number is larger than the lot size.
Using this algorithm, the random numbers are repeated after the table has been accessed 2500 times . Therefore, after the first 2500 times, the first starting column is increased by 1, the second by 2, the third by 3, etc. After an additional 2500 times - if the start access has already been increased - the first column is increased again by 1, the second by 2, and so on. However, the contents of the table are never changed.