Streamlining BPM Modeling – Reduce Initial BPM Process Rollout Time by 80%

The objective of modelling a BPM process is to test process templates prior to rollout to a production environment.

Process templates consist of steps, sometimes interconnected in complex ways, where each step requires an indicated level of performance skill and, usually, the collection of context-situation appropriate data to validate performance of steps and for use downstream from such steps.

Some process steps carry out calculations on recorded data, accordingly, modeling includes validation of calculation rules.

Other steps found in process templates include branching decision boxes which require a selection by a user. Some branching decision boxes can be automated in which case modeling includes testing and validation of rule sets at decision box options.

Usual defects in process templates include unnecessary steps, missing steps, steps connected in the wrong order, steps with bad routings, bad forms, bad calculation routines and bad rules.

Other modeling objectives include testing of the User Interface. No point putting in place facilities for doing the right things, the right way, at the right time, using the right resources if users refuse to use the run time software made available to them.

Process modeling can take up a lot of time.

Here is a tip on how to dramatically reduce the time and cost of initial rollout of a BPMs processes for modeling purposes.

Whereas it takes ½ an hour to one hour, or more (depending on complexity), to “paint” a data collection form for production use, you can image 20-40 forms per hour using an ordinary camera on tripod and use the images for modeling your processes.

When you substitute images of forms for real forms in a 40-50 step workflow that has, say, 20 forms, it takes 2 hours to build the workflow and put in place form images, instead of 12 hours to build/roll out a template that features real forms. This gives you an initial 80% rollout time savings, with no extra cost beyond the cost of a camera and copy stand.

Prove it to yourself using the following protocol:

1. Drag and drop steps on your canvas/sheet and connect these using directional arcs to build your test workflow.
2. Assign routings at each step.
3. Attach to each step a data collection form that features an image control plus a memo control.
4. Do not spend time automating branching decision boxes – instead, attach a placeholder form to each decision box option and describe the rules that will later be used by a rule set when you automate the processing at the decision box.
5. Park one or more of your data collection and placeholder forms at each workflow step.
6. Compile your workflow and piano-play one or more instance of your template in a test run time environment with a small group of stakeholders.
7. As stakeholders comment on the sequencing of steps, routings, forms at steps, record their comments at the step forms and then immediately update and re-issue your template.
8. Now replace your forms with data collection/branching decision box forms and calculation algorithms/rules, recompile your template, repeat the piano play.

Compare your time/cost for items #1 through #7 with the turnaround time for initial modelling using the traditional approach of generating paper maps, marking these up by hand, going away to generate new versions of your paper maps, and then hosting another session with the process documentation team.

N.B. Actual initial rollout time using traditional approaches to modeling is a lot higher because of the need to host multiple mapping sessions with stakeholders. Each additional session requires a “settling in” time. You are likely to find that months can be reduced to a few days using “two-pass” rollout of processes.