Probability Distributions

Correct Answer: Triangular Distribution with a peaked mode

The Triangular Distribution with a peaked mode is most appropriate for modeling cost outcomes that concentrate near the best estimate because: 1. It reflects a concentrated probability around the most likely value (mode) 2. It provides a practical and intuitive representation of project costs 3. It accounts for both optimistic and pessimistic scenarios while maintaining focus on the most likely estimate Why other options are incorrect: The Uniform Distribution assumes equal probability across all values, which is unrealistic for project costs as some values are typically more likely than others. The Beta Distribution with extreme positive skewness would shift probability mass away from the best estimate, contradicting the requirement for concentration near the expected value. The Log-normal Distribution with extended right tail parameters would spread the probability too far to the right, not maintaining the desired concentration near the best estimate.

Correct Answer: Uniform distribution

The Uniform distribution is the correct choice for this scenario. A Uniform distribution represents a situation where all values between a minimum and maximum have an equal probability of occurring. This matches exactly with the question's description of 'equal chance of any value occurring between a minimum and maximum value'. The other distributions have different characteristics: - Normal distribution concentrates values around a mean with decreasing probability toward the extremes - Beta distribution is flexible but typically skewed and bounded between 0 and 1 - Triangular distribution has a defined mode (peak) point where probability is highest, with decreasing probability toward the minimum and maximum values Only the Uniform distribution maintains the same probability across all values in its range, making it the perfect match for this scenario.

Correct Answer: Lognormal Distribution

The Lognormal Distribution is the most appropriate choice for modeling duration of multiple sequential activities with random variations because: 1. It reflects the natural behavior of activity durations where: - Activities can't have negative durations - Most activities cluster around a mode - There's a positive skew for potential delays 2. Central Limit Theorem suggests that when multiple independent random variables are multiplied, their product tends toward a lognormal distribution 3. Sequential activities often have multiplicative effects on duration variations, which lognormal distribution models accurately Other distributions are less suitable because: - Chi-square distribution is better suited for quality control and hypothesis testing scenarios - Exponential distribution models time between independent events, not cumulative durations - Uniform distribution assumes equal probability across all outcomes, which doesn't reflect real-world project duration variations