Skip to content

marksxiety/man-hours-calculator

Repository files navigation

Man-Hours Calculator

Tests Build Lint MIT License

The Man-Hours Calculator is a data-driven framework designed to bridge the gap between technical uncertainty and business requirements. By replacing "gut-feeling" guesses with the Program Evaluation and Review Technique (PERT), it provides a disciplined way to account for the inherent risks in software development.


The PERT Methodology

The core of this system is the three-point estimation technique. Instead of providing a single number, developers provide three distinct data points for every task:

  1. Optimistic (O): The "perfect world" scenario—no blockers or surprises.
  2. Most Likely (M): The most realistic, frequent outcome under normal conditions.
  3. Pessimistic (P): The "worst-case" scenario—unexpected bugs or hidden dependencies.

The Weighted Average (Expected Duration)

The system calculates Expected Duration (E) using a weighted average that deliberately leans toward the most likely estimate while still accounting for outliers:

$$E = \frac{O + 4M + P}{6}$$

The "4M" weight reflects real-world behavior: outcomes cluster around what is most likely. Pure averages treat all scenarios equally, which often overestimates uncertainty. This formula keeps the estimate grounded while acknowledging the "tails" of risk.


Features and Interface

  • Real-Time Computation: All calculations update instantly as you type, including task-level duration, standard deviation, and project-level variance.
  • Probability Analysis: Calculates the statistical chance of finishing by a target deadline based on accumulated project variance.
  • Developer Feedback Loop: Allows developers to self-correct unrealistic estimates by seeing the mathematical impact of their inputs immediately.
  • Modern Tech Stack: Built with Vue 3, TypeScript, Vite, and TailwindCSS for a high-performance experience.

Understanding the Math

Standard Deviation (σ)

A measure of "spread"—how far apart your best and worst-case guesses are. Formula: $$\sigma = \frac{P - O}{6}$$

  • Small σ: Signals high confidence; the task is well-understood.
  • Large σ: Acts as a warning sign; the task is unpredictable or poorly scoped.

Variance (σ²)

Variance is the key to project-level uncertainty. While you cannot add standard deviations together, variances can be summed to find the total project risk. Higher cumulative variance means a wider, less predictable delivery window.

Confidence Intervals

Range Interpretation
E ± 1σ ~68% chance of finishing within this window
E ± 2σ ~95% chance of finishing within this window
E ± 3σ ~99.7% chance of finishing within this window

Glossary of Terms

  • PERT: Program Evaluation and Review Technique—a statistical tool used in project management.
  • Estimation Debt: The accumulated cost of optimistic guesses that ignore technical reality.
  • Sigma (σ): A statistical term for Standard Deviation; used here to represent task-level risk.
  • Expected Duration (E): The final number used for planning, derived from the weighted PERT formula.

Documentation Links


Project Goal

To eliminate "Estimation Debt" by applying a mathematical approach to quoting that accounts for uncertainty while providing full transparency to both developers and stakeholders.

About

Data-driven framework for calculating software project man hours using PERT methodology, Calculate expected duration, variance, completion probability to create quotes you can defend.

Topics

Resources

License

Stars

Watchers

Forks

Contributors