Use the Process Model window to specify that a step be started upon completion of another step, regardless if the other step ran successfully or failed.
For example, you are defining
steps that populate target tables that are unrelated. However, you
want them to run in sequence so that all the tables in the warehouse
are updated during the same time period. You can use the Process
Model window to specify that the second step is to start after the
first step runs, even if the step fails, so that its target table is
updated during the same time period as the first step.
To start a step when another step stops running:
Click 
(the Data Link icon).
Click 
(the On Completion icon).
Click the step that is to run first.
Hold down the mouse button and drag the mouse to the step that is to run after the first step.
Release the mouse button.
A blue arrow that represents the task flow is displayed on the canvas between the two icons.
Click Process --> Save to save the changes that you have made to the process.
Related information