Inspection Results
This discussion continues from our previous example in batch synchronisation.
By the end of the five steps, we've ended up with a frame that looks like the one below…
Property | Value |
---|---|
StartTime | 10:00 |
EndTime | 12:00 |
Order Number | 29238 |
SKU | AR-9233 |
Machine Setting 1 | 22 |
Machine Setting 2 | 34 |
Machine Setting 3 | 11.22 |
Goal
Our next goal is to now add inspection data to this information, so we not only know what settings were used, but how the final product came out.
Combining this sort of high-speed data automated data with low-speed, manually entered data allows us to do some very interesting things in the AI space.
Plan
Inspections come in to ARDI as events. The to make this work, we need to identify what piece of information our existing frame has (see the table above) and what information our inspections have.
If the frame and our inspections have any common identifiers, we can use them to help figure out which inspections are related to which time-frames.
Our inspection data include…
Property | Value |
---|---|
Date | 7/9/2044 |
Result | OK |
Quality | 100 |
OID | 29238 |
In this case, our inspection data also includes the Order Number, so we can use this as our lookup key.
Step 1: Look Up Events
Here, we'll use the lookup_event layer.
It pulls data from the 'Inspection System' source in ARDI. Then it looks for any inspection with a matching order ID that happens during or after the time frame (inspections are always done after the product has been completed.
{ "type": "lookup_event", "source": "Inspection System", "global": "true", "key": "Order Number", "rkey": "OID", "endbuffer": 1200, "prefix": "Lab ", "time": "after", "comment": "Lookup inspection details", "prefix": "Lab " }
Thanks to the prefix, all of the names will have 'Lab' affixed.
Property | Value |
---|---|
StartTime | 10:00 |
EndTime | 12:00 |
Order Number | 29238 |
SKU | AR-9233 |
Machine Setting 1 | 22 |
Machine Setting 2 | 34 |
Machine Setting 3 | 11.22 |
Result | OK |
Quality | 100 |