Configuring Predecessors in forward scheduling vs backward scheduling

George Speier

I have a user who wants to schedule some projects in a forward-looking manner (top to bottom) and schedule other projects in a backward-looking manner (bottom to top).

To explain to her how Predecessors would need to be configured in either scenario, I created the little example shown below. In the FORWARD section, the manually entered Start date on row 3 (highlighted) triggers the automatic population of all the other dates. In the BACKWARD section, the manually entered Start date on row 23 (highlighted) is the trigger.

These examples show how to achieve the exact same schedule using either strategy (forward or backward).

In the BACKWARD section, everything seems pretty straight-forward to me, except when Duration values of zero are involved. To get the schedule to match the schedule in the FORWARD section, I had to add the lag times shown above. While this works, I have no idea why it does, which makes explaining the process a lot more difficult. 😉 Can anyone explain to me why these lag times were necessary? Many thanks!

Evan Barker

This has to do with how a milestone is represented (ie. a 0 day duration). If you hard type milestone without a predecessor or successor, it will show up as the beginning of that day instead of the end of the day. If you have a milestone that is following a standard finish to start logic, it will show up at the end of the previous activity.

So, Task B4 in the backwards version is actually saying that the completion happened in the 00:00 hr of that day. if you put any activity in that has a duration and uses the start to finish relationship, that activity will naturally finish at the end of the previous day. So, to get your example above to work, the finishing task would need to be 1 day long so it takes up the full work day. I found it easier to visualize if you zoom all the way in on the gantt chart so you can see the diamond symbol of the milestone at the beginning and end of the days when used in different set ups.

George Speier

Thank you, @Evan Barker!

Evan Barker

This has to do with how a milestone is represented (ie. a 0 day duration). If you hard type milestone without a predecessor or successor, it will show up as the beginning of that day instead of the end of the day. If you have a milestone that is following a standard finish to start logic, it will show up at the end of the previous activity.

So, Task B4 in the backwards version is actually saying that the completion happened in the 00:00 hr of that day. if you put any activity in that has a duration and uses the start to finish relationship, that activity will naturally finish at the end of the previous day. So, to get your example above to work, the finishing task would need to be 1 day long so it takes up the full work day. I found it easier to visualize if you zoom all the way in on the gantt chart so you can see the diamond symbol of the milestone at the beginning and end of the days when used in different set ups.

George Speier

Thank you, @Evan Barker!