So you have spent a lot of time and effort in developing your ideal “TO-BE” process model and now you are ready to start automating. WAIT! You are not ready just yet.

The Gap
The conceptual process model is NOT the final model that can be executed by your Business Process Management System; there is a gap in the information provided by this model and the executable process model that is required to run the business process.

These process models have two different audiences:

  • Conceptual model – produced by business and process analysts for communication and analysis purposes.
  • Executable model – developed by software developers and solution architects into a precise specification to be read by a BPMS.

Over the last few weeks I have been participating in QUT’s first “Fundamentals of BPM” MOOC which introduces the fundamentals of business process management (BPM) by working through each phase of the BPM lifecycle. Review my last post Automation and the BPM Lifecycle if you need a refresher before proceeding.

During the implementation phase of this lifecycle the conceptual “TO-BE” process model, which was developed during the redesign phase, is now transformed into an executable process model.

This executable process model contains information about:

  • Handling exceptions both technical and process.
  • Data types for data objects such as the format of invoices and purchase orders.
  • Process and task variables.
  • Human resources and their relationship in the organisation.
  • How to connect to external systems such as ERP and records management.

Last weeks course module introduced a “5-step approach” to bridge the gap between the conceptual process model and the executable process model.

The five steps are to use the conceptual process model and:

  1. Identify automation boundaries
    • Determine what steps can be automated and what steps would not lead to an increase in efficiency if automated.
  2. Review manual tasks
    • Establish how these tasks will link to the automated tasks.
  3. Complete the process model
    • Add additional required data elements and tasks related to execution such as exception handling.
  4. Adjust task granularity
    • Maximise efficiency gains through automated task coordination by:
      • Breaking up tasks that are too abstract into more atomic tasks.
      • Aggregating tasks that have been specified in too much detail.
  1. Specify execution properties
    • Data types for specified data objects
    • Data mappings between process variables and task variables

I have previously worked in a project where time was only scheduled for the business analyst to develop the conceptual process model. The particular analyst did not have experience of the specific BPMS being used was asked to develop a model that could be implemented in the BPMS without change. This led to two issues:

  • The conceptual process model was developed as a mix of a conceptual and executable model with assumptions of how the BPMS handled certain situations, which masked the actual tasks in the process.
  • The project plan did not allow any time for the developer to capture the information required to transform the conceptual model into something that could be executed by the BPMS.

A little study of the BPM Lifecycle can be used to great benefit when planning to automate your favourite business process and I can thoroughly recommend the QUT Fundamentals of BPM MOOC