Anonymous
Most project managers realize that the overall project plan is composed of many components. One of those many components is often the project work plan, developed using some form of project scheduling tool such as Niku, Primavera, MS Project, etc. I know that there are many other tools that enable nice drawings for project plans (Visio, PowerPoint, Excel, etc). The focus of my question below is the use of scheduling tools that allow the project manager to incorporate a WBS, tasks estimates, personnel availability, etc to develop a schedule. I also realize that some projects simply just need a pretty picture. Also, I realize that not all projects are of sufficient size and complexity to warrant the use of a scheduling tool.
As with any tool, it should not be the focus of the project. Rather, the project work plan should aide in the process of planning, not get in the way. Regardless of which software a project manager might use, or the approach used in developing the project work plan, the resulting project work plan ought to result in an effective project work plan.
I’d like to put some better definition around what is meant by “effective”. After all, if I am going to use a tool, I would like to have it clear as to the attributes of an effective plan. Here are some attributes that I would like to hold forth as being reasonable, minimal requirements if I am going to take the time to develop a project work plan using one of these scheduling tools. In no particular order:
1. The WBS is reflected in the project work plan, although often progressively elaborated.
2. The WBS should clearly show the deliverables, whether final or interim.
3. Accountability for each task is assigned
4. Dependencies are used only when absolutely necessary to represent physical constraints.
5. External dependencies, if any, are evident in the work plan.
6. Deadlines must be reflected in the work plan. i.e if the project is due on a certain date, then that’s what the plan should show unless it is physically impossible to meet that date and you are out of options.
7. The scheduled dates should reflect the estimated hours and/or duration. (Even those doing duration-based estimating and planning implicitly are thinking about how many hours might actually be required and the availability of then person doing them)
a. Tasks not started should be scheduled to start in the future, not the past
b. Tasks not complete should have finish dates in the future, not the past
c. Another way of stating 7a and 7b is that no incomplete work would be scheduled in the past.
8. The estimated hours required of person assigned to the work should be within the realm of reason over a given time period. (i.e don’t expect a person to work 80 hours each week over a 3 week period). The project work plan should show that reasonable resource loading.
9. Tasks should be of a reasonable duration/hours such that project progress can be easily tracked and observed. (There are many guidelines, just pick one).
a. Tasks that are too small require too much effort to track
b. Tasks that are too long too often result in surprises and a false sense of progress.
10. The project work plan should clearly demonstrate that sufficient resources existing to accommodate the work not yet completed.
11. The project work plan is able to accommodate unforeseen events as inputs (they always happen), allowing the project manager to use the project plan to assess the impact in any changes to cost, schedule or scope. The project plan should also help the project manager determine the best approach to getting back on track, whether cost, scope or schedule.
I know that this forum is not shy about stating what they don’t like (if so, let me have it!). But I also know that there are people out there who have the experience and skills to provide some constructive input too.
What would you add to the list?
Rich
Saving Changes...
|
|
|
Darrell Markham
Sr. IT Project Manager| Tyler Technologies
Brentwood, Ca, United States
Rich,
I like most of what you have written. I agree with your updating method. However, I disagree with - "4. Dependencies are used only when absolutely necessary to represent physical constraints." I use MS Project where if dependencies are used correctly, with the proper constraint type, you will have a very dynamic schedule that will properly model the current reality of your project. I attempt to use only dependencies, with ASAP constraints, so that an update to one task updates all of the successor tasks as well. So that when I do an update on a task that is behind schedule, and I move the remaining work back into the future where it belongs, all of the successor tasks also move out by the same amount to properly model the current state of the project. Using dependencies keeps the schedule dynamic and reduces maintenance time during execution, when you really don’t have the time to spend changing a long list of start and finish dates.
Darrell
Saving Changes...
|
|
