A three point schedule estimate uses optimistic, most likely and pessimistic durations to describe the uncertainty surrounding an activity. The scheduler can convert those inputs into a weighted duration for a deterministic proposal schedule or retain them as a probability distribution for schedule risk analysis.
The method works best when subject matter experts define each estimate against specific assumptions. Simply asking for a best case, normal case and worst case often produces arbitrary numbers. Instead, the proposal team should identify the work, resources, productivity, dependencies and risk conditions behind each duration.
What Is a Three-Point Schedule Estimate?
A three-point estimate contains three values:
- Optimistic duration: The shortest credible duration under favorable conditions. It should still reflect an executable plan, not a miracle scenario.
- Most likely duration: The duration expected under the team’s current assumptions about scope, resources, productivity and normal interruptions.
- Pessimistic duration: A longer but credible duration under adverse conditions. It should reflect plausible problems without assuming an uncontrolled disaster.
The U.S. Department of Energy defines three-point estimating as a technique that uses optimistic, pessimistic and most likely cost or duration estimates to develop a representative estimate. The GAO Schedule Assessment Guide also describes optimistic, most likely and pessimistic duration fields as inputs to schedule risk analysis.
Three-point estimating does not automatically produce a risk-adjusted completion date. It quantifies uncertainty at the activity level. The schedule network, calendars, logic, constraints, risk events and correlations determine how that uncertainty affects program milestones.
How to Develop the Three Duration Inputs
1. Define the activity before estimating it
First, confirm that the activity has a clear scope and completion condition. An activity called “Complete Design” may be too broad to estimate reliably. A task such as “Develop preliminary antenna mechanical design” gives the estimator a more specific unit of work.
The estimate should align with the activity’s level of detail. If the task combines unrelated work performed by several organizations, break it down before applying three-point estimating. The resulting durations will be easier to defend and update.
2. Establish the most likely duration
Start with the most likely value because it represents the current execution plan. Ask how much work must be performed, who will perform it and what calendar applies. Also identify productivity assumptions, review cycles, resource availability and expected rework.
Use historical actuals, engineering judgment, vendor information or analogous work whenever possible. For proposal development, the estimator should identify whether the duration assumes resources available at contract award, after staffing ramp-up or after a predecessor delivers required information.
The companion article on developing proposal schedule durations explains how to connect scope, productivity and resources to executable activity estimates.
3. Define a credible optimistic duration
Next, ask what favorable conditions could shorten the work. Examples include immediate resource availability, first-pass approval, reuse of a mature design or early receipt of supplier data.
However, do not assume overtime, unlimited staffing or waived technical reviews unless those conditions form part of the actual execution strategy. The optimistic estimate should remain feasible within the proposed technical and management approach.
4. Define a credible pessimistic duration
Then identify the conditions that could extend the activity. These may include additional review comments, lower-than-planned productivity, limited test resources, delayed data, supplier response time or modest rework.
The pessimistic value should not combine every imaginable problem. Major discrete events often belong in the risk register and schedule risk model instead. For example, a routine range for fabrication uncertainty should not quietly include a low-probability supplier bankruptcy.
5. Test the range
Finally, confirm that the values follow this relationship:
Optimistic duration ≤ most likely duration ≤ pessimistic duration
Review the width and shape of the range as well. A narrow range implies high confidence. A wide range signals substantial uncertainty, weak definition or exposure to several risks. If the range appears unusually wide, the activity may need further decomposition or separate risk events.
How to Calculate a Three-Point Duration
There is no single calculation that fits every use. The appropriate method depends on whether the team needs one planning duration or a probability distribution for risk analysis.
PERT weighted duration
The traditional Program Evaluation and Review Technique (PERT) formula gives the most likely value four times the weight of each endpoint:
Expected duration = (Optimistic + 4 × Most Likely + Pessimistic) ÷ 6
For example, assume an equipment interface drawing has the following estimates:
- Optimistic: 8 working days
- Most likely: 12 working days
- Pessimistic: 23 working days
The PERT calculation is:
(8 + 4 × 12 + 23) ÷ 6 = 13.17 working days
The team might use 13.2 working days in an estimating repository. If the schedule convention requires whole days, it could enter 13 days after documenting the rounding rule. Avoid repeatedly rounding intermediate calculations because the cumulative effect can distort a long network.
Microsoft documents the same default 1-4-1 weighting for the Optimistic Duration field and PERT analysis in supported desktop versions of Microsoft Project.
Triangular mean
A triangular distribution uses the optimistic value as the minimum, the most likely value as the mode and the pessimistic value as the maximum. Its arithmetic mean is:
Triangular mean = (Optimistic + Most Likely + Pessimistic) ÷ 3
Using the same example, the mean equals 14.33 working days. That result exceeds the PERT value because the triangular mean does not apply extra weight to the most likely estimate.
Do not collapse the range when performing an SRA
For a schedule risk analysis (SRA), the better approach is usually to retain all three values. The risk tool can use them to define a triangular, beta-PERT or other approved distribution. It then samples activity durations across repeated simulations.
This process allows paths to change between iterations. An activity outside the deterministic critical path may become critical when its sampled duration increases. Therefore, a single weighted duration cannot replace a network-level SRA.
The GAO guide also warns that activity duration ranges may combine several unnamed threats and opportunities. A risk-driver analysis can provide more diagnostic value because it assigns identifiable risks to affected activities. Programs often use both methods: duration uncertainty for normal variability and discrete risk events for specific threats or opportunities.
Fictional Proposal Example: Avionics Test Station
Assume a proposal team is developing an integrated master schedule (IMS) for a new avionics test station. The statement of work requires the contractor to develop test procedures, conduct a test readiness review and execute qualification testing.
The test lead estimates “Develop Qualification Test Procedures” at 30 working days. During the estimating interview, the scheduler obtains more detail:
- Optimistic, 24 days: Experienced authors are available at authorization to proceed, requirements remain stable and the internal review closes in one cycle.
- Most likely, 30 days: The planned team is available, normal clarification requests occur and the document requires two internal review cycles.
- Pessimistic, 46 days: Specialized authors support another priority during the first week, requirements need clarification and the second review generates moderate rework.
The PERT duration equals 31.67 working days. The proposal scheduler enters 32 days in the deterministic schedule and records the three inputs in the duration basis of estimate.
However, the team also identifies a separate risk that Government-furnished test interface data may arrive late. The scheduler does not bury that event inside the 46-day pessimistic duration. Instead, the proposal risk lead records the event and its impact so the SRA can model it against the affected activities. The related discussion of Government-furnished equipment and schedule risk explains why external dependencies require explicit treatment.
Document the Basis of Estimate
A three-point estimate becomes useful when another professional can understand and update it. Record at least the following information:
- Activity name and stable identifier
- Scope and completion criteria
- Optimistic, most likely and pessimistic durations
- Calendar and duration units
- Resource quantities and skill assumptions
- Productivity or throughput assumptions
- Historical or analogous data
- Review and rework assumptions
- Included uncertainty
- Excluded discrete risks
- Calculation or distribution selected
- Estimator, reviewer and estimate date
NASA’s program planning and control glossary describes a schedule basis of estimate as documentation of the assumptions, methodology, resources, uncertainty, risk and supporting analysis behind duration estimates. NoahsPlan provides a focused template and approach in Basis of Estimate for Schedule Durations.
How to Use Three-Point Estimates in a Proposal Schedule
Proposal teams should separate the planned schedule from the uncertainty analysis. The deterministic IMS normally needs one duration for each activity. The estimate repository or risk model can retain the full range.
Also keep these practices in mind:
- Follow the request for proposal. FAR 15.203 states that a request for proposal identifies the information offerors must submit and the factors used for evaluation. Therefore, the solicitation governs whether the customer expects a schedule BOE, risk analysis, confidence level or specific estimating method.
- Use consistent estimating rules. Define what optimistic and pessimistic mean before interviewing control account managers or technical leads. Otherwise, one estimator may use 10th and 90th percentile values while another uses absolute minimum and maximum values.
- Align the schedule and cost estimate. If the duration assumes two engineers, the cost volume should not assume one engineer unless the team explains the difference. Review schedule-to-cost integration for the broader program-controls workflow.
- Preserve traceability. Connect each estimate to the work breakdown structure, statement of work and estimating source. That traceability helps reviewers determine whether the proposed schedule covers the required scope.
- Revisit high-uncertainty activities. Wide ranges may identify areas where early planning, supplier engagement or risk mitigation can improve the proposal.
Three-point estimating is a recommended estimating and risk-analysis practice, not a universal Federal Acquisition Regulation requirement. It also does not become an Earned Value Management System requirement merely because the program uses an IMS. Contract terms, the solicitation, agency direction and approved program procedures determine the required deliverables and methods.
Common Failure Modes
Using arbitrary percentages
Applying minus 10 percent and plus 25 percent to every activity creates the appearance of analysis without examining actual uncertainty. Standard ranges may help with an initial screen, but subject matter experts should validate them.
Treating optimistic as an accelerated plan
An estimate based on overtime, extra shifts or added resources describes a different execution strategy. Either include that strategy in the proposal or remove it from the optimistic case.
Making the pessimistic value a hidden risk register
If the estimator cannot explain what drives the pessimistic duration, the program cannot manage the underlying risk. Separate normal duration uncertainty from identifiable risk events where practical.
Using the PERT result as a confidence date
A weighted activity duration does not show the probability of meeting a program milestone. That conclusion requires a risk analysis of the integrated network.
Ignoring calendars and resources
Ten working days on a five-day calendar differs from ten days on a seven-day test calendar. Likewise, a duration based on two full-time engineers becomes unreliable if the resource plan provides only one part-time engineer.
Adding pessimistic durations throughout the baseline
Entering worst-case values for every activity can create hidden contingency and an uncompetitive proposal. Instead, build a credible deterministic plan, quantify uncertainty and place visible schedule margin or contingency according to the program’s approved approach.
Practical Takeaway
A useful three-point estimate is not three numbers collected in a spreadsheet. It is a concise model of how scope, resources, productivity and uncertainty affect an activity.
Start with an executable most likely duration. Then define credible favorable and adverse conditions. Document the assumptions, separate discrete risks and choose the calculation based on the intended use. For a deterministic proposal schedule, a PERT-weighted value may provide a practical planning duration. For an SRA, preserve the full range and analyze it within the logic-driven network.
Finally, review the estimates with the same discipline applied to the rest of the proposal IMS. The broader process in Proposal Scheduling for Government Contracts explains how duration development fits into schedule architecture, compliance and proposal execution.
Frequently Asked Questions
Is the most likely estimate the same as the current schedule duration?
Not always. The current duration may be an early placeholder or management target. The most likely estimate should reflect the current scope, resources and execution assumptions.
Should every proposal activity receive a three-point estimate?
No. Focus first on uncertain, critical, near-critical and externally dependent activities. Applying detailed ranges to low-risk administrative tasks may add effort without improving the decision.
Can the pessimistic duration exceed a contractual milestone?
Yes. The range should represent uncertainty honestly. If simulations show an unacceptable probability of missing the milestone, the team should revise the plan, mitigate risk or communicate the exposure rather than manipulate the estimate.
Does a three-point estimate include schedule margin?
No. The estimate describes activity duration uncertainty. Schedule margin or contingency is a separate allowance established from risk exposure and the program’s approved scheduling approach.