The Datasource object can be configured to provide an interface to connect to batch record servers to retrieve and work with batch records. The Batch Record Datasource supports the following functionality:
- Fetching batch records from various source systems (AspenTech APRM/Batch.21 and PI-Batch with OSIsoft’s PI system currently supported)
- Extending batch records with additional information
- Monitoring the state of active batches
- Reforming of batch record information so it can be inserted into an Asset Effectiveness Tracking system using SQL commands
- Provision of enhanced batch records to the Production Tracking System (upcoming)
The Batch Record sub class of the Datasource object has the following interface options:
- AspenTech APRM (formerly Batch.21)
- OSI PI (PI-Batch) – this type requires the PI-Bridge extension package to be installed on the particular inmation Connector which runs the Batch Record Datasource object(s).
- External Tags – this mode allows to construct pseudo-batch records from any “Live Data” (connected to system:inmation over OPC, OPC UA, native PLC etc. Datasources)
When discussing the Batch Record Datasource, there are a lot of terms that sound similar to one another but have difference in meaning. To avoid confusion and clarify any ambiguity, the following table is provided.
|Batch Record Datasource||The object class in system:inmation to monitor and process batch data (Batch Records) for external systems. Datasources are created below Connector objects and the Batch Record server type is selected at the time of creation.|
|Batch Record||A data document that reflects the data of one production batch. The document may contain a Batch Header, Unit SubBatches, Operations, Phases etc.. The Batch record will alo contain start and end time as well as other elements such as Setpoints, Educts and Products.|
|Batch Data Source||An AspenTech APRM term: A distinct Batch Data Source is configured in the AspenTech Batch Management software. It is a logical name, which itself has a relation to one or multiple Batch Areas|
|Batch Area||An AspenTech APRM term: A distinct Batch Area is configured in the AspenTech Batch Management software. It is a logical name, which defines common characteristics for all Batch Units assigned to the Batch Area|
|Batch Unit||A physical asset which can execute a Batch Process, turning educts into intermediates and/or products according to setpoints in a recipe. Batch Units may operate stand-alone or in parallel and/or a sequential manner|
|Datasource||This can either be an inmation software class (The Batch Record Datasource is one of many specializations of the inmation Datasource class) or an ODBC datasource, which grants access toa database system on Operating System level. As such, the documentation context is an important read in order to distinguish between both variations. If connected to an AspenTech APRM/Batch.21 system the inmation Batch Record Datasource requires a pre-configured and functional ODBC data source to connect to the APRM database.|
All APRM batch records are fetched from the source system through a pre-configured ODBC data source on the computer running the inmation Connector service. The Connector must have access to the particular ODBC data source. This implies:
- The APRM ODBC data source needs to be configured as system data source
- The bit-model of the inmation Connector service and the APRM ODBC data source need to be the same. Currently only a 32-bit APRM ODBC data source is available so an equivalent 32-bit Connector service should be installed on the corresponding host machine
- The inmation Connector must be configured to use Logon credentials which grant access to the APRM system without supplying additional credentials in the Batch Record Datasource object (tested), or alternatively such credentials need to be configured in the Batch Record Datasource and the Connector may run under its default LOCAL_SYSTEM account
More information on configuring APRM ODBC data source drivers and setting service log on credentials can be found here.
The OSI-PI interface requires the installation of the inmationPIBridge service. Currently, the inmationPIBridge service is distributed on request from inmation support (contact: email@example.com or through the support portal), where installation intructions will also be provided.
The tags which are required to be mapped to a virtual batch record must be configured in system:inmation to supply the relevant data. Relevant objects of type I/O Item or Holder Item must exist for each Batch Characteristic. A particular Batch Record Datasource can only access objects which are configured on the same Connector.
A unique advantage of the Batch Record Datasource is that it provides a unified Batch Record for Asset Effectiveness tracking. In order to support this, the main characteristics are defined in a system-independent manner and seperated into different sections.
The following table supplies an overview of the inmation standard fields which deal with the general Batch Context:
|Batch ID||This is the unique batch identifier in the external system. ). The field must be usable to retrieve exactly one Batch from the external system.||The format of the Batch ID differs depending on the external system. APRM uses a running number, Pi-Batch uses a GUID string, External Tags can be anything).|
|Batch Number||The logical Batch number as shown on a DCS screen while the Batch is in active production.|
|Batch Status||Detemines if batch is active or completed. Only required to be configured with External Tags Interface to a low level system. Batch status is automatically determined with APRM and Pi-Batch systems.||Batch status is True if batch is active and False if batch is completed|
|Unique Batch ID||The Unique Batch ID is created by system:inmation for use in Asset Effectiveness/Production Tracking. It is created from underlying field components from the external system (Batch ID, Batch Number, Start and End times) and should not be configured manually|
|Production Version||Information about the production information||System and user specific|
|Start Time||The UTC start time of a particular Batch or Unit Batch. In case the data acquisition is related to APRM or PI-Batch via PI-Bridge the system uses default time interpretation. In case the time is taken from External Tags it can be either an ISO8601 time string or a POSIX millisecond numerical value||Example for an ISO8601 UTC string:
2018-07-01T13:45:12.145ZMilliseconds can be omitted. If the ‘Z’ is ommited it will be assumed to be present. Local time stamps will only be supported in a later release.
|End Time||The UTC end time of a particular Batch or Unit Batch. Details see above.||See above|
The following table supplies the material classification for products and educts. For products the Material Type will default to “Product”. For educts the Material Type will default to “Educt”. In case no field assignment is defined in order to determine a specific material type, the defaults will be used.
|Material Type||The readable name of a material type.||The following types are supported in the current version: Invalid = 0, Product = 1, Educt = 2, Reinsert = 3, Waste = 4, Cleaning = 5|
|Material Type Code||Numerical code related to the material type||See above|
In case the actual configuration contains only the Type, the Type Code will be determined from the Type (string). If the Type string is not in the list of the supported types it will be assigned the code for ‘Invalid’ (0). If the actual configuration supplies only the Material Type Code, the Material Type is determined from the Material Type Code. As such, it is sufficient to implement a logic to determine the Material Type Code alone (in case it deviates from the standard for one or multiple units – which is automatic Educt / Product assignment based on the selected target field and does not require any specfic logic).
In general, the system will not be able to derive the Material Type for anything other than Educt or Product. In case you want to specify a ‘Reinsert’ type as an example, you have to provide a custom script, e.g. analyzing the material name and assigning specific ones to be of ‘Reinsert’ type.
This table contains the Educt-related target characteristics. Each Batch or Unit Batch may specify zero, one or multiple educts related to a particular Batch.
|Educt Actual||The actual consumption of the educt. Either a volatile value to be read from a tag or a one-shot report at the end of the Batch.|
|Educt Code||The ERP/MES code for the educt|
|Educt EU||The engineering unit to measure the consumed mass||e.g. "kg", "t" or "tons"|
|Educt Name||The readable name of the educt||e.g. "VE-Wasser"|
|Educt Setpoint||The recipe setpoint for the target educt consumption|
The Product-related target characteristics are in the following table. Each Batch or Unit Batch might specify zero, one or multiple products related to a particular Batch. Each distinct product related to a Batch will create a distinct record for the Asset Effectiveness ‘Main Product’ table. Additional products related to Batch Units will create additional records in the related ‘Additional Product’ table.
|Product Actual||The actual produced mass of the product. Either a volatile value to be read from a tag or a one-shot report at the end of the Batch.|
|Product Code||The ERP/MES code for the product||User Specific|
|Product EU||The engineering unit to measure the produced mass per Batch||e.g. “kg”, “t” or “tons”|
|Product Name||The readable name of the product||e.g. “Toluran 45”|
|Product Setpoint||The recipe setpoint for the target production for that particlular product|
The following target characteristics reflect on recipe data.
|Master Recipe Version||The version of the master recipe which was used for a particular Batch.||System-specific, user-specific|
|Control Recipe Name||The name (or number) of the Control Recipe used for a particular Batch||System-specific, user-specific|
For more information on creating and configuring a Batch Record Datasource, please visit the DataStudio documentation.