Overview on the Components of Process Mining

Have a look at the figures below to get a technical overview on the components of Scheer PAS Process Mining:

Scenario 1: Single BRIDGE Installation

If you have set up one single BRIDGE installation on a machine, the full stack of Process Mining services needs to be deployed and executed on that machine.

Figure: Technical Overview on the Components of Process Mining (Scenario 1)

Scenario 2: Distributed BRIDGE Installations

If you have set up BRIDGE installations on multiple nodes to distribute the services, you might want to collect and analyze data from multiple nodes. In that case it is not necessary to deploy the full stack of Process Mining services to each BRIDGE installation. The TrxLogsCollector is able to use an (S)FTP location to forward the collected log files to TrxLogsETL. Therefore, only one node is needed to run the full stack of Process Mining services. The stack of Process Mining services on all other nodes is reduced to the TrxLogsCollector.

Figure: Technical Overview on the Components of Process Mining (Scenario 2)

You will have to install the ETL tools and a related database, the Analytic API and the Process Mining UI.

Overview on the Installation Process

To install Scheer PAS Process Mining, you need to got through the following steps:

1. Setup the database used by the Process Mining services.
   The Process Mining services rely on analytical data that has been collected and stored in a database.
   For a description on how to setup the database for the use of the Process Mining services, refer to Setting up the Database for Process Mining.
2. Deploy the services that will collect and extract the analytical data.
   The analytical data Process Mining relies on have to be collected and extracted to the analytical database.
   For a description on how to deploy the data services, refer to Deploying the Services.
3. Deploy the services that read and provide the data in the analytical database.
   The Node.js services will provide you with an interface to inspect the collected data. These services use the Analytic API service to access the analytical database. For a description on how to do this, refer to Deploying the Services.
   • xUML service (Analytic API service)
   • Node.js services
4. Connect the data collection services to the database.
   To connect the deployed services from steps 2 and 3 to the database configured in step 1, you need to adjust some BRIDGE settings to match your configuration as described in Configuring the Process Mining Services.
5. Start the services.
   It does not matter in what order you start the services.
6. Use the Scheer PAS Administration to create Process Mining users.
   In order to use Process Mining, assign the necessary profiles to the system users. For further information about the needed profiles and how to assign them, consult the ADMIN. See page Configuring the Platform Services for further information about the composition of the Scheer PAS Administration URL.

Updating the Dashboards

Generally, you can just upload the latest service repository to update the Process Mining services. However, with updating TrxLogsETL there are some restrictions and you have to follow the procedure described in Updating the Services.