Subtask linking

Question

I have 2 tasks like Task A and B. Task A has 2 Subtask named A1 and A2 and the same for Task B(B1 and B2). Task B1 needs Task A1 to be done befor starting and Task B2 needs Task B1 to be done before starting. Actually I have dozens of these type of tasks and I can't link them all one by one. If I just link the Summary tasks together then Microsoft project will calculate the subtasks dependencies for me too ?

Here is an example. In animation studios first we Draw something as reference then we will start modeling of it so I have 2 summary tasks named : Designing and Modeling.

If I have 100 object to Design then how can I create my Project plan !?

Answer 1

Maybe grouping tasks in products instead of tasks could work for you. What I am thinking is, instead of having:

• Design
  • Design product 1
  • Design product 2 ....
• Model
  • Model product 1
  • Model product 2

You could have:

• Product 1
  • Design
  • Model
• Product 2
  • Design
  • Model ...

You could copy and paste the Product tasks with the dependence, making it easy to add a new one. Maybe resources can be assigned before pasting, so you always have a designer for design and modeller for model.

I completely agree on assigning resources at leaf level.

Don't know if this approach can work for you, but this is how I would do it (with the information I have).

Answer 2

Assuming Task B1 has all Tasks A1, A2, ..., An as predecessors then you could set the summary/hammock Task A as the predecessor. This will have Task B1 starting after the last of Task A's subtasks completed. Note that if you have some subtask of Task A (call it Task Ax) that is NOT a predecessor for Task B1, and Task Ax completes AFTER all other subtasks of Task A, then Task B1 will start artificially late. So you will need to do a reality check on your schedule if you start linking things to hammock tasks.

As you enter subtasks under Task B in Project the software's logic should by default set dependencies between them along the lines of B1 is needed to start B2 is needed to start B3, provided you enter them in that order.

Answer 3

While Doug is technically correct, I'd advise against building your logic at the hammock levels. I personally load resources and logic at the leaf level of my schedule, the lowest level of decomposition. While it might be logical in your head about how a summary level will work, the tool is only so smart and you cannot rely on its calculations based on that for things like critical path, or slack, or anything else. The tool will literally expose issues that are not really there or hide issues that are there. In fact, when you load things like FF or SS constraints, it too hoses up the schedule's calculations making the results either invalid or unreliable. My firm has built a ton of add-ins to compensate for the tool's flaws and, even with these tools, we follow strict rules on how to load the schedule.

Do not look for an easy way out if you want your tool to really work for you. If you don't care about that, use a spreadsheet and call it day.

Answer 4

Build the logic at the leaf level, but try to organize the work differently

I agree with David that you have to build the logic at the leaf level.

The number of the Work Breakdown Structure (WBS) levels and the granularity of the tasks is mainly dependent on the size and nature of the project. But, it is also dependent on how the work is divided up among the resources. If you have extreme functional specialization in your team, with each team member doing a small slice of the work and handing it off to the next person, you have to track them all as separate tasks and manage their dependencies.

For example, let us say you have 2 people for the Designing and Modeling, If one of them does the Designing and hands it off to the other person to do the Modeling, you need to track them as separate tasks (especially if they don't get along with each other!). On the other hand, if you can get one resource to do the Designing and Modeling for 50 objects and the other resource the other 50, you can track them at a more aggregate level.