Process Design
What Process Design covers
Process Design is about structuring the process diagram itself: the main page, sub-pages, stage types, decision logic, loops, and exception blocks. Candidates must know the standard stage set - Start, End, Action, Decision, Calculation, Multi-Calculation, Choice, Note, Block, Collection, Loop, Page, Read, Write, Navigate, Code, Exception, Recover, Resume, Anchor - and when each is appropriate.
A central idea is the exception handling model: a stage can throw a System Exception or a Business Exception. Business Exceptions mean the data is invalid and the item should be marked as an exception; System Exceptions mean the application or environment failed and the item should be retried. The Recover stage, paired with Resume, lets a process re-enter a block after handling. Block stages group stages for recovery and clarify intent.
Traps in questions
Common traps: using a Business Exception for an application crash (wrong - that should be a System Exception so the item retries), swallowing exceptions with an empty Recover, or putting recovery blocks around the whole process instead of the risky interaction. Another trap is looping inside a page instead of using the Loop stage over a collection, which hides the iteration count and complicates recovery. Questions may also test whether a Decision stage or a Choice stage is correct: Choice routes to multiple onward paths, Decision is a binary branch.
Worked example
A process reads a queue item, logs into a portal, and enters a reference number. Login failure is transient: wrap the login action in a Block with a Recover that throws a System Exception so the item is retried. If the reference number is rejected as invalid, throw a Business Exception so the item is marked exceptioned and not retried. Everything else - reading the item, marking it complete - sits outside the recovery block.