Project Management

14 quality checks for your IT project schedule

One way to evaluate whether your project schedule is good is to measure it against this list of quality checks from the Defense Contract Management Agency.

IT project management consultants and Project Management Office (PMO) managers see project schedules that vary in quality. A PMP certified project manager (PM) may have 10 years of experience managing projects, but that isn't an indicator of the PM's ability to build a good schedule.

checklist_1.jpg
Evaluating a project schedule can be subjective unless there is a set of objective criteria to follow. Well-structured PMO organizations may have project schedule quality guidelines, but unless there's a schedule review, the schedule could still have significant defects that impact realistic project delivery. (Remember: The project schedule is a forecasted model of future events, so you want the model to reflect reality as much as possible.)

The Defense Contract Management Agency (DCMA) provides a 14-point assessment that can be applied to evaluate project schedule quality. The DCMS is a Department of Defense (DoD) department that works with suppliers to manage time, cost, and performance for the U.S. federal government.

Pre-assessment metrics

You need to calculate these metrics before applying the quality checks.

  1. Total Tasks is a count of the lowest level tasks, excluding summary, subproject, zero baseline duration, and milestone tasks.
  2. Complete Tasks is a count of Total Tasks with an Actual Finish date.
  3. Incomplete Tasks is a count of Total Tasks without an Actual Finish date.
  4. Baseline Count is a count of all the Total Tasks that should have completed on or before the project status date.
  5. BEI Baseline Count is a count of all the Total Tasks that should have completed on or before the project status date and all the tasks that are missing a baseline start and finish date.
  6. Relationship Count is a count of all the Finish-Start (FS), Start-to-Start (SS), Start-Finish (SF), and Finish-to-Finish (FF) dependencies in the schedule.

The 14 quality checks

  1. Logic Check evaluates if any tasks are missing a successor or a predecessor in the schedule.  Logically, all activities should have at least one predecessor or successor tasks or else you risk creating an orphaned task. A quality schedule should not have more than 5% of its tasks missing logic. The logic check formula is (# of Tasks missing Logic) / (# of Incomplete Tasks) * 100.
  2. Leads Check looks for negative lags (i.e., leads) in the schedule. A quality schedule should have zero leads, as they can impact a project's critical path.
  3. Lags Check ensures the use of lags is minimized in the project schedule, as lags can impact the critical path. A quality schedule would have fewer than 5% of the tasks with lags. The lag check formula is (# of Tasks with Lag) / (Incomplete Tasks).
  4. Relationship Types Check validates 90% of the tasks have Finish to Start relationships.  Using Finish to Start relationships provides a logical flow to the schedule.
  5. Hard Constraints Check is any activity that has a Must Finish On, Must Start On, Start No Later Than, or Finish No Later Than constraint. You often see these constraints in the "pick-a-date" project management. You ideally want your schedule to be a dynamic schedule that flows easily when the project is updated. The hard constraints check identifies any Incomplete task that has a hard constraint. A quality schedule will not have more than 5% of its tasks with hard constraints.
  6. High Float Check identifies all tasks with a total float of more than 44 working days. A task with such a high float usually indicates a predecessor or a successor was missing. The formula is (# of High Float tasks) / (# of Incomplete Tasks). The percentage should not exceed 5%.
  7. Negative Float Check identifies any tasks that have a float less than zero. This check helps identify any tasks that may hinder the completion of other tasks in the schedule. There should be zero tasks with negative float.
  8. High Duration Check identifies any task that has a baseline duration greater than 44 days.  As competent PMs, we know such tasks should be broken down into more discrete tasks. The formula is (# of High Duration tasks) / (# of Incomplete Tasks) and should not exceed 5%.
  9. Invalid Dates Check identifies any total task that has a scheduled start or finish date before the project status date and an actual start or finish date after the project status date.  Forecasted start and finish dates should always appear after the project status date. The schedule also can’t contain an actual start or finish date in the future. There should be zero tasks that fall into this category in a quality schedule.
  10. Resources Check identifies all the tasks that do not have resources (people or costs) assigned. A quality schedule has all resources assigned to tasks in the schedule.
  11. Missed Tasks Check is used to identify any task that was scheduled to complete by the project status date and finished after the baseline finish date. The reality is projects will have late tasks, but you want to ensure your current schedule re-aligns to the baseline schedule. The formula is (# of tasks with forecasted finish before the status date and actual or forecasted finish after baseline schedule date) / (baseline count).
  12. Critical Path Test Check assesses the integrity of the schedule network using the critical path. An intentional delay of 600 days is entered into the remaining duration of a critical path test. The expected result is the project is delayed by 600 days since you added a delay directly on the critical path. The schedule passes the test if the project finish date matches the added duration date finish date.
  13. Critical Path Length Index (CPLI) determines if the project finish date is realistic given the forecasted finish date. The critical path length index is calculated using the formula (Critical Path Length + Total Float) / (Critical Path Length). The value should be greater than 1.0 to pass the test.
  14. Baseline Execution Index (BEI) is similar to Earned Value's Schedule Performance Index as it compares the number of completed activities to date to the number of tasks planned to be completed by the project status date. The BEI is calculated # of Complete Tasks / BEI Baseline Count and should be greater than 1.0.

Summary

You can use these quality checks as a framework to evaluate and review project schedules.  The criteria can also help you evaluate how well a supplier can build and execute a meaningful schedule.

As a project manager, I've seen my share of bad schedules, and I'm sure I've built a poor schedule every now and then. The DCMA criteria serve as a useful schedule quality gauge that eliminates subjective evaluations. 

In my next article, I'll show how to automate some of these quality checks in your Microsoft Project schedule.  

 

About

Dr. Andrew Makar is an IT program manager and is the author of How To Use Microsoft Project and Project Management Interview Questions Made Easy. For more project management advice visit http://www.tacticalprojectmanagement.com.

3 comments
mike.mouser
mike.mouser

I’m trying to create a SQL Query so I can include these metrics in a report. The number of tasks (complete and incomplete) are pretty easy to come up with, but the predecessors, leads, lags are a few bits of information that don't seem to be available, at least not where I can find them. Do you have a SQL query that would include all of these metrics?

sparent
sparent

"you want to ensure your current schedule re-aligns to the baseline schedule." Uh? I don't see how you can re-align to the baseline unless you change your task's duration. That might not always be feasible. It might not even be desirable if the task is not on the critical path.

amakar
amakar

@sparent Thanks for the comment.  With any schedule, there will be variances as the project progresses.  The goal is to realign the forecasted tasks to meet the original baseline dates or the key milestone dates.  As a PM, it is good to recognize a variance but you also want to bring that variance back inline with the baseline commitments.

This may mean reducing scope or skipping some tasks in order to meet the original end date.

I've seen far too many projects with large variances with a naive expectation that they'll meet their original target date.