DCMA 14-Point Schedule Assessment, Explained

On a construction project of any size, the schedule is the plan everyone works from: the owner, the general contractor, subcontractors, and often a court or an arbitrator if something goes wrong. A schedule full of broken logic or unrealistic dates can’t be trusted to forecast completion. The DCMA 14-point assessment is one of the most widely used ways to check whether a schedule is sound before anyone relies on it.

What is the DCMA 14-point assessment?

The DCMA 14-point assessment is a set of schedule quality checks originally developed by the U.S. Defense Contract Management Agency to evaluate contractor schedules on federal programs. The checks later moved into construction and engineering scheduling, where they now serve as a baseline standard for schedule health on CPM schedules, most often built in Primavera P6.

The assessment doesn’t judge whether a project will finish on time. It judges whether the schedule is built well enough to be trusted as a forecasting tool. A schedule that passes can still slip; one that fails several checks can’t reliably predict the finish.

Why schedule quality matters in construction

Construction schedules get more scrutiny than almost any other kind of project plan, for a simple reason: money and time both ride on them. A few specific pressures make schedule quality a real business issue, not just a scheduling nicety.

• Delay claims and disputes: a late project’s schedule is often the primary evidence for who caused the delay, and broken logic is weaker evidence.
• Owner and legal scrutiny: public agencies, sureties, and legal teams increasingly request DCMA-style metrics with monthly submittals, since the checks are well established.
• Inconsistent scheduler quality: the person building a schedule might be a certified planner or a project engineer doing it as one task among many; the checks catch problems either way.
• The cost of an untrustworthy forecast: crew mobilization, material orders, and subcontractor sequencing all ride on a finish date the schedule may not predict.

The 14 checks, explained

The 14 checks fall into a few groups: logic and relationships, float and duration, date integrity, and higher-level tests on the critical path as a whole. The thresholds below are commonly cited thresholds teams often adapt to their own contract requirements, not a rigid universal law.

1. Logic: activities missing a predecessor or successor. Threshold: 5% or fewer.
2. Leads: negative lag between activities. Threshold: 0%, since leads generally shouldn’t be used.
3. Lags: positive lag between activities. Threshold: 5% or fewer of relationships.
4. Relationship types: share of relationships that are Finish-to-Start. Threshold: 90% or more FS.
5. Hard constraints: activities with constraints like Must Start On or Must Finish On that override logic. Threshold: 5% or fewer.
6. High float: activities carrying more than 44 working days of total float. Threshold: 5% or fewer.
7. Negative float: activities with negative total float. Threshold: 0%.
8. High duration: activities with durations longer than 44 working days. Threshold: 5% or fewer.
9. Invalid dates: actual dates in the future, or forecast dates in the past, relative to the data date. Threshold: 0%.
10. Resources: a completeness check confirming activities meant to carry cost or resource assignments actually have them.
11. Missed tasks: activities that finished late against the baseline. Threshold: 5% or fewer.
12. Critical path test: confirms the critical path is continuous and logic-driven, usually traced by adding a large duration to one activity and confirming the finish date shifts accordingly.
13. Critical Path Length Index (CPLI): critical path length plus total float, versus critical path length alone. Threshold: 0.95 or higher.
14. Baseline Execution Index (BEI): ratio of completed activities to activities that should be complete per baseline. Threshold: 0.95 or higher.

Logic, leads, and lags

The first three checks look at how activities connect. Missing logic means an activity has no predecessor, no successor, or both, so it can float freely or drop out of the real driving path. Leads let a successor start before its predecessor is far enough along, hiding a sequencing decision inside a date instead of showing it in the logic. Lags are less severe, but leaning on lag time instead of real logic makes a schedule harder to update accurately as work progresses.

Relationship types and constraints

Finish-to-Start is the simplest, most predictable relationship type; the other types are more prone to modeling errors. Hard constraints are dates that force an activity to a fixed point regardless of what its logic says. A schedule can look complete and logically sound while a handful of hard constraints quietly override that logic underneath it.

Float and duration

High float usually signals a missing relationship rather than genuine slack. Negative float means the schedule is already behind its own constraints or milestones and shouldn’t appear in an approved baseline. High duration flags activities running longer than roughly two months without a break, which usually means the activity should be split into smaller pieces.

Dates, resources, and missed work

Invalid dates catch a basic but common data error: activities marked complete in the future, or forecast to finish in the past, relative to the data date. The resources check confirms that activities meant to carry cost or resource loading actually have it attached. The missed tasks check compares actual finish dates against the baseline, an early signal of a program’s real execution trend.

Testing the critical path itself

The last three checks matter most for forecasting confidence, because they test the critical path as a system rather than testing individual activities. The critical path test, typically adding a large duration to one activity and confirming the finish date moves in response, proves the path is continuous and logic-driven. CPLI compares critical path length plus total float against critical path length alone, one number for how much cushion the plan carries. BEI compares completed activities against what the baseline said should be complete by now, one of the clearest signals of real execution pace.

The manual problem

Running all 14 checks by hand, every month, on every active schedule is slow and easy to get wrong. On a program with several concurrent contracts, a scheduler might review thousands of activities across multiple schedules in the same week, pulling activity-level detail out of the scheduling software into a spreadsheet built for that one review.

That process is where consistency breaks down. Two reviewers checking the same schedule by hand often produce different findings, simply because of which filters they applied or which activities they happened to spot-check. That inconsistency becomes a real problem at exactly the moment it matters most: a dispute, a claim, or a contract audit.

A faster way to run these checks: Schedule Validator

Manually checking all 14 points is only one option. Schedule Validator, available through Projects Analytics, automates DCMA-style checks using its own DCMA+ scoring method, returning a single Schedule Validation Score (SV Score) plus execution and critical path analysis and an itemized list of what to fix, activity by activity.

Because the checks run the same way every time, two people running the same schedule through it arrive at the same answer. It works with schedules built in Primavera P6, Microsoft Project, and Asta Powerproject, which matters when subcontractors and consultants aren’t all standardized on the same software.

Frequently asked questions

Does a schedule have to pass all 14 checks to be considered acceptable?

Not necessarily. Most contracts and internal standards treat the assessment as a diagnostic tool, not a strict pass or fail gate on every check. A schedule clearing most checks with a couple of minor, explainable exceptions is generally in good shape; one failing several at once, especially negative float, invalid dates, or a weak CPLI, usually needs real rework before anyone should trust its forecast.

Can the 14-point assessment be run on schedules built outside Primavera P6?

Yes. The checks are logic-based rather than software-specific, so they apply to any CPM schedule with activities, relationships, durations, and a baseline, regardless of which tool built it. Primavera P6 is the most common source given how widely it’s used on larger programs, but the same 14 checks apply just as well to schedules built in Asta Powerproject or other CPM scheduling software.

How often should a schedule actually be run through the 14-point assessment?

Most programs that use the assessment run it on every monthly schedule update, since that’s the cadence schedules are typically submitted and reviewed anyway. Owners and legal teams also request it whenever a schedule is submitted as part of a claim or dispute, since it gives a documented, repeatable baseline for how sound the schedule was at that point.