Lecture 28: Project Estimation Techniques - Heuristic
Unit 4: Software Project Management (4353202)
Lecture Agenda
- Recap of Empirical Estimation (COCOMO)
- What is Heuristic Estimation?
- Technique 1: Expert Judgment
- Technique 2: Delphi Technique
- Technique 3: Estimation by Analogy
- Advantages of Heuristic Techniques
- Disadvantages of Heuristic Techniques
- Key Takeaways
Recap of Empirical Estimation (COCOMO)
Empirical estimation models like COCOMO use historical data and mathematical formulas to predict project effort and time. Basic COCOMO provides a rough estimate, while Intermediate COCOMO refines it using cost drivers.
What is Heuristic Estimation?
Heuristic estimation techniques rely on experience, intuition, and rules of thumb rather than strict mathematical models or detailed historical data. They are often used when historical data is scarce or when a quick, preliminary estimate is needed.
These techniques leverage the knowledge and judgment of experienced individuals or teams.
Technique 1: Expert Judgment
This is the simplest and most common heuristic technique. It involves consulting one or more experts who have experience with similar projects.
- Process: An expert (or a small group of experts) provides an estimate based on their knowledge and past experiences.
- Advantages: Quick, inexpensive, and can be highly accurate if the expert is truly knowledgeable and the project is similar to their past work.
- Disadvantages: Highly subjective, prone to bias (e.g., optimism bias), and accuracy depends entirely on the expert's experience and judgment.
Technique 2: Delphi Technique
The Delphi technique is a structured communication method that aims to achieve a consensus among a group of experts. It minimizes the influence of dominant personalities and encourages independent thought.
- Process:
- Experts provide individual, anonymous estimates.
- Estimates are compiled and summarized, and the summary is shared with all experts.
- Experts revise their estimates based on the summary, again anonymously.
- This iterative process continues until a consensus or a narrow range of estimates is reached.
- Advantages: Reduces bias, encourages diverse perspectives, and can lead to more accurate estimates than individual expert judgment.
- Disadvantages: Time-consuming, requires a facilitator, and may not always lead to a strong consensus.
Technique 3: Estimation by Analogy
This technique involves estimating a new project by comparing it to one or more similar past projects (analogous projects) for which actual data is available.
- Process:
- Identify a past project that is similar in size, complexity, and technology.
- Compare the new project's characteristics to the analogous project.
- Adjust the analogous project's actual effort/cost/duration based on the differences.
- Advantages: Relatively quick, easy to understand, and can be accurate if truly analogous projects exist.
- Disadvantages: Highly dependent on the availability of truly similar past projects and the ability to accurately identify and quantify differences.
Advantages of Heuristic Techniques
- Flexibility: Can be applied at any stage of the project, even with limited information.
- Speed: Often provide quick estimates, useful for initial planning and feasibility studies.
- Cost-Effective: Generally less expensive than detailed algorithmic models.
- Leverages Experience: Directly utilizes the valuable knowledge and intuition of experienced professionals.
Disadvantages of Heuristic Techniques
- Subjectivity: Highly dependent on individual judgment and prone to bias.
- Lack of Transparency: The rationale behind the estimate may not always be clear or easily justifiable.
- Inconsistency: Different experts may provide widely varying estimates.
- Limited Accuracy: May not be as accurate as algorithmic models for complex or novel projects.
Key Takeaways
- **Heuristic estimation** relies on **experience and intuition** for quick, preliminary estimates.
- Common techniques include **Expert Judgment, Delphi Technique, and Estimation by Analogy**.
- They are **flexible and fast** but can be **subjective and less transparent**.
- Often used in conjunction with other techniques for more robust estimates.
Next Lecture
Topic: Project Estimation Techniques - Analytical
Q & A
Questions & Discussion

