Uploading Production Schedules
Production schedules can be uploaded to an Acuit Pinpoint database, populating the database with preconfigured unit serial numbers, model numbers, and model number aliases. Doing so can help reduce the chance of bar code reader mis-scans causing bad data to be introduced into the database.
Empty unit records (i.e., units with no history) are created for entries in the production schedule. Stations on the line can then be configured so that they are not allowed to create new unit records; a scan of a unit serial number that has not been uploaded in a production schedule (or an incorrect serial number due to a bar code reader mis-scan) will not be accepted and an error message will be shown to the operator.
Production schedules can either be uploaded manually via "Maintenance, Upload Schedule" in the Acuit Pinpoint web interface, or they can be configured to automatically import schedules from a drop folder via the <scheduleUpload> Element line configuration setting.
Model Number Aliases
Uploaded production schedules are also the mechanism used to establish model number aliases. A model number alias represents the equivalent manufacturer model number when the actual model number on the bar code label with the unit is not a normal manufacturer model number (for example, when a unit is manufactured for someone else to be sold under a different name and with a customer-defined model number). When a model number alias exists for a unit, that model number is used internally by Acuit Pinpoint when performing the various checks against the unit.
Unit records that are created via a production schedule upload obtain their model number aliases from the production schedule file. Units that are created during a unit scan on the line (at stations configured to allow this) obtain their model number aliases from a model alias lookup table. This table is updated every time a production schedule file is processed based on the model numbers in the file, adding or updating the model-to-model-alias cross references as needed. It can also be manually updated at any time by manually editing the ModelAliases table in the Acuit Pinpoint databases.
Production Schedule Files
Production schedule files must be comma-separated values (CSV) files, where each line in the file contains information for one unit.
Schedule files must never contain a header line. Instead, the column layout is defined via the
Headers
setting in Schedules Section.
This setting can be defined at any hierarchy level down to the line, but it is typically defined
once for the entire server within the settings.json file. For example:
{
"Acuit.Pinpoint": {
"Schedules": {
"Headers": "SerialNumber,ModelNumber,ModelNumberAlias"
}
}
}
The following header names are used by Acuit Pinpoint:
Name | Description |
---|---|
SerialNumber | The unit serial number. This column is required. |
ModelNumber | The unit model number. This column is required. |
ModelNumberAlias | The unit model number alias. If this column does not exist or this value for a row is blank, then the value from the row's ModelNumber column will be used. |
ProductionGroup | The production group name. If this column does not exist, then the name of the schedule file, without the file extension, is used as the production group name. |
SequenceNumber | The sequence number in the production group for the unit. If this column exists, values must be a signed 32-bit integer, or blank. |
Action | The action to take with the row: "A" to add a unit record, or "D" to delete a unit record (to allow changes in the production schedule). If this column does not exist, then "A" is assumed. When adding a unit record, if it already exists in the database, it will not be affected. When deleting a unit record, only unit records with no history will be deleted. Once a unit has been scanned at any station, it has established some history, and an attempt to delete it by this method will fail. |
Case, spaces, and underscores are ignored when matching each header name. The header names can occur in any order.
The schedule file may contain rows with fewer or more columns that are indicated by this header row. In rows that contain fewer columns than are specified by the header, the missing field values are assumed to be blank. In rows that contain more columns than are specified by the header, the extra fields are ignored.
For example, the Headers setting of "SerialNumber,ModelNumber,ModelNumberAlias"
(from the example
above), could be used for schedule files structured like this:
S00000001,MODEL1234,MODEL1234
S00000002,MODEL1234,MODEL1234
S00000003,TRADEMODELXYZ,MODEL1234
The Headers setting of "ModelNumber,SerialNumber,Action,ModelNumberAlias"
(which is the default),
could be used for schedule files structured like this:
MODEL1234,S00000001,A,MODEL1234
MODEL1234,S00000002,A,MODEL1234
TRADEMODELXYZ,S00000003,A,MODEL1234