Which of the following distribution represents the time estimates in PERT ?
Beta distribution
In project management, the Program Evaluation and Review Technique (PERT) is a method used to analyze and estimate the time required to complete a project. Unlike methods that use a single point estimate for activity duration, PERT uses a probabilistic approach, considering uncertainty in time estimates.
Understanding Time Estimates in PERT
For each activity in a PERT network, three time estimates are typically made:
- Optimistic Time (O): The shortest possible time in which the activity can be completed, assuming everything goes exceptionally well.
- Most Likely Time (M): The most probable time required to complete the activity under normal circumstances. This is the estimate that has the highest probability of occurring.
- Pessimistic Time (P): The longest possible time required to complete the activity, assuming unfavorable conditions and reasonable delays.
These three estimates capture the range and likelihood of different completion times for an activity.
Why Beta Distribution is Used in PERT
The Beta distribution is the standard probability distribution used in PERT to model the duration of individual activities. It is chosen for several reasons:
- It is defined over a finite interval, which fits the nature of activity durations bounded by optimistic (O) and pessimistic (P) estimates.
- It can be skewed, reflecting that the 'most likely' time (M) might not be exactly in the middle of the optimistic and pessimistic times. This flexibility allows it to model situations where things are more likely to go wrong (skewed towards the pessimistic side) or more likely to go well (skewed towards the optimistic side).
- While the exact parameters of the Beta distribution (\(\alpha\) and \(\beta\)) are complex to derive directly from O, M, and P, PERT simplifies this by using the mean and variance of the Beta distribution that is thought to approximate the actual distribution of the activity duration.
Calculating Expected Time and Variance using Beta Distribution in PERT
PERT uses simplified formulas derived from the Beta distribution (specifically, often a Beta distribution with parameters \(\alpha = \frac{4M + O - P}{P - O}\) and \(\beta = \frac{4M + P - O}{P - O}\) + 1, or other approximations) to calculate the expected time and variance of an activity duration based on the three estimates (O, M, P).
The formulas commonly used are:
| Measure | Formula (PERT) |
|---|---|
| Expected Time (\(T_e\)) | \(T_e = \frac{O + 4M + P}{6}\) |
| Variance (\(\sigma^2\)) | \(\sigma^2 = \left(\frac{P - O}{6}\right)^2\) |
These formulas provide the average duration and a measure of the uncertainty (spread) around that average, based on the Beta distribution assumption.
Other Distributions Not Typically Used for Individual Activity Time in PERT
- Poisson distribution: Used for counting occurrences of events over an interval, not for measuring duration.
- Normal distribution: Symmetrical and extends from \(-\infty\) to \(+\infty\). While the total project completion time (sum of many activities) in PERT often approaches a Normal distribution due to the Central Limit Theorem, individual activity times are bounded and often skewed, making the Normal distribution less suitable for them.
- Weibull distribution: Often used in reliability engineering for modeling time to failure. Not standard for activity durations in project management.
Therefore, the Beta distribution is the most appropriate choice among the given options to represent the time estimates for individual activities in PERT.
Revision Table: Key Concepts in PERT Time Estimation
| Concept | Description |
|---|---|
| PERT | Project Evaluation and Review Technique; uses probabilistic time estimates. |
| Optimistic Time (O) | Best-case scenario duration. |
| Most Likely Time (M) | Most probable duration. |
| Pessimistic Time (P) | Worst-case scenario duration. |
| Beta Distribution | Probability distribution used to model individual activity duration in PERT. |
| Expected Time (\(T_e\)) | Weighted average duration: \(\frac{O + 4M + P}{6}\). |
| Variance (\(\sigma^2\)) | Measure of uncertainty: \(\left(\frac{P - O}{6}\right)^2\). |
Additional Information on PERT and Probability
While the Beta distribution is used for individual activities, a key benefit of PERT is its ability to estimate the probability of completing the entire project by a specific deadline. This is typically done by calculating the expected duration and variance of the critical path (the sequence of activities that determines the project's shortest possible completion time). The expected duration of the critical path is the sum of the expected durations of the activities on that path. The variance of the critical path is the sum of the variances of the activities on that path (assuming independence). The Central Limit Theorem suggests that the distribution of the project completion time (especially if the critical path involves many activities) can be approximated by a Normal distribution. This allows calculating the probability of completion by a certain date using Z-scores and standard normal tables.
Understanding the underlying distributions helps in assessing project risk and setting realistic deadlines.
Important Questions from PERT and CPM
Which one of the following distributions provides information regarding the uncertainty of duration time estimates is PERT described network?
Negative slack occurs when -
In PERT analysis, the possible number of time estimates for activities linking up two events are -
The amount of time by which an activity can be delayed without affecting project completion time is
- Slack represents the difference between the-
