AutoStore workflow configurations can quickly become large and confusing. Therefore, we recommend to use multiple Tasks in the AutoStore configuration to define different parts of the workflow. Splitting the configuration into e.g. 3 Tasks has proven to be very effective and could be done as follows:
Task 1 - Capture
Task 2 - Process
Task 3 - Route
If possible, the Capture Task should only contain the capture component and the components required to route the job to the next Task. The Process Task should contain all components that are needed for document processing and the Route Task only the ones that are needed for routing the files into the final depository.
To route jobs between Tasks, the Knowledge Package Builder + Send to Folder components must be used as final 2 components in all Tasks (besides in the Route Task) while the Knowledge Package Loader component needs to be used as the capture component (besides in the Capture Task).
This approach has the advantage that it becomes very easy to maintain a workflow. The resulting logs can be clearly tracked back to a specific Task which makes troubleshooting easier, but also reprocessing of failed files or issue reproduction becomes easier as well. Furthermore, these Tasks can be saved as templates and used in other configurations too.
Another advantage of splitting the workflow is that it separates the capture from processing components which might otherwise potentially slow down the scanning process of a connected MFP. It is a technical limitation that for some web service based MFP capture components devices become blocked until the Capture Task is done with processing the scanned job. This obviously takes longer, especially for large scan jobs, if document processing (e.g. OCR) is being done in the Capture Task too and can be kept very short by splitting the Tasks as described above. This applies to some web service based MFP capture components like e.g. Kyocera MFP (HyPAS).