PERT key terms

A graphical model showing project activities as boxes or arrows and linking them with dependency lines, so the sequence and logical flow of work can be analyzed.

A network drawing method where each activity is represented by a node, usually a box, and arrows show only logical dependencies between activities rather than duration.

A network drawing method where arrows represent activities and nodes represent events or points in time; dummy arrows are sometimes added to keep the logic correct.

An activity that must begin or finish before another specified activity can start or finish, establishing the required order of work in the network.

An activity whose start or finish depends on the start or completion of another activity earlier in the network sequence.

A dependency type where the successor activity cannot start until the predecessor activity has finished; it is the most common relationship in project schedules.

A dependency type where the successor activity cannot start until the predecessor activity has started, allowing two activities to run in parallel after a common start.

A deliberate waiting period inserted between two dependent activities, expressed in time units, that delays the successor beyond the normal relationship.

A negative lag that allows a successor activity to begin before its predecessor finishes, effectively accelerating the start of downstream work.

A zero-duration, resource-free arrow used in activity-on-arrow networks to show a logical dependency that cannot otherwise be drawn correctly.

An activity with two or more immediate predecessors, where separate paths come together; its start waits for every incoming path to finish.

A significant point or event in the project with zero duration, used to mark completion of a major deliverable or phase.

A technique that uses optimistic, most likely, and pessimistic duration estimates to capture uncertainty and compute an expected activity duration and variance.

The shortest realistic time an activity could take if everything goes unusually well, with no major problems or delays.

The duration an activity is expected to take under normal conditions, assuming typical problems and normal productivity, and the most probable single value.

The longest realistic time an activity could take if major problems occur, excluding extreme catastrophes or highly unlikely events.

The weighted average duration calculated as (Optimistic + 4 times Most Likely + Pessimistic) divided by 6, used as the mean activity time in PERT analysis.

A probability distribution often used in PERT to model activity duration, allowing asymmetry around the most likely value and bounded optimistic and pessimistic limits.

A three-point probability distribution that uses optimistic, most likely, and pessimistic values with simple averaging, producing a less peaked shape than the beta distribution.

A measure of duration uncertainty for one activity, calculated as (Pessimistic minus Optimistic) divided by 6 in classic PERT, showing how spread out possible durations are.

The square of an activity's standard deviation, representing the spread of its possible durations and used to build project-level schedule variance.

The longest-duration path through the network, which determines the shortest possible project completion time and has zero total float when no constraints exist.

The amount of time an activity can be delayed without delaying the project completion date, calculated as late start minus early start or late finish minus early finish.

The amount of time an activity can be delayed without delaying the early start of any immediate successor activity in the network.

A calculation that moves from network start to finish, assigning early start and early finish dates based on predecessor durations and dependencies.

A calculation that moves from network finish to start, assigning late finish and late start dates and revealing where float exists.

The earliest possible date an activity can begin, given the finish dates of its predecessors and the project start date.

The earliest possible date an activity can complete in the schedule, equal to its early start plus its estimated duration.

The latest date an activity can begin without delaying the project completion date, found by subtracting its duration from its late finish.

The latest date an activity can complete without delaying the project completion date, determined from successors' late starts during the backward pass.

A network path whose total float is very small, so modest duration slippage could make it critical and change the project finish date.

A method that converts the difference between a target date and expected project duration into a Z-score, then uses the normal distribution to estimate on-time probability.

The number of standard deviations a target completion date lies above or below the expected project duration, used to look up completion probability.

A bell-shaped probability curve with mean zero and standard deviation one, used with Z-scores to find the chance of finishing by a given date.

A specified deadline or desired project finish date that is compared with the expected duration to compute the probability of finishing on time.

The probability, expressed as a percentage, that the project will finish by a particular date or within a particular duration.

The sum of the variances of activities on the critical path, used as the measure of uncertainty for the overall project duration.

The combined variance of all activities along a particular network path, indicating how much that path's duration may vary.

A computer-based technique that repeatedly samples activity durations from probability distributions to build a distribution of possible project completion dates.

The process of examining uncertainty in activity durations and dependencies to understand the likelihood of meeting schedule targets and identify risky paths.