Estimating with Fermi Problems and Scaling

1. The Philosophy of "Street-Fighting" Physics

Before we dive into techniques, it's crucial to adopt the right mindset. In theoretical physics, precision is not always the primary goal. The initial, and often most important, step is to grasp the essential physics by knowing what details to ignore. This approach, which emphasizes intuition and clever approximation over brute-force calculation, is a hallmark of effective theoretical work.

Let's begin by watching a discussion with Sanjoy Mahajan, a physicist and educator at MIT, who is a prominent advocate for this "street-fighting" approach to mathematics and physics.

2. Fermi Problems: The Art of Guesstimation

The first technique, named after the physicist Enrico Fermi, involves answering seemingly impossible questions by breaking them down into a chain of smaller, more easily estimated quantities. Fermi was legendary for his ability to get surprisingly accurate answers with very little data.

The video below provides an excellent introduction and a step-by-step guide to solving Fermi problems.

The core of the Fermi method is a five-step process:

1. Identify and Break Down the Problem: What pieces of information do you need?
2. Dare to Be Imprecise: Make educated guesses for each piece. Don't worry about getting them perfect. The power of the method lies in the cancellation of errors.
3. Bound the Problem: Use what you know to set upper and lower limits on your guesses. Is it closer to 10, 100, or 1000? This is often all the precision you need.
4. Do the Math: Multiply and divide your estimates. Keep track of your units.
5. Perform a Sanity Check: Does the final answer seem plausible in the context of what you know about the world?

Your Turn: A Fermi Problem

Let's apply this to a domain relevant to your studies. Given your work in astrophysics and condensed matter, let's try this:

Estimate the total energy (in Joules) released by all core-collapse supernovae in the Milky Way galaxy over a typical human lifespan (say, 80 years).

Try to break this down before looking at the suggested decomposition below.

3. Scaling Arguments: How Things Change

Fermi problems are excellent for estimating static quantities. But much of physics is concerned with dynamics—how one quantity changes in response to another. This is the domain of scaling laws. A scaling argument describes a relationship of the form $Y\propto X^a$, where the core of the analysis is determining the scaling exponent, $a$.

Geometric Scaling

The most intuitive scaling laws are geometric. We know that if we scale up the length of an object, its surface area and volume scale differently.

This square-cube law is fundamental. As you can see in the image of scaling squares, it's a general property of dimensionality, not a specific property of the object's shape.

The resource below provides a good introduction to these basic ideas and lists many fascinating examples of physical scaling laws.

Physical Scaling and Successive Approximation

Scaling arguments become truly powerful when we apply them to physical laws. Often, a rough estimate can be systematically improved by accounting for smaller effects—a method known as successive approximation. This involves identifying the "big part" of a problem, solving that simplified version, and then using that solution to calculate a first correction.

The book Street-Fighting Mathematics by Sanjoy Mahajan is the definitive guide to these techniques. Let's work through one of its classic examples.

Another powerful tool for approximation, especially for analyzing how a system responds to small changes, is the fractional change approximation. For any power-law relationship $y\propto x^n$, a small fractional change in $x$ produces a fractional change in $y$ given by:
$\frac{\Delta y}{y}\approx n\frac{\Delta x}{x}$
This is simply the first-order term of a Taylor expansion, but it's an incredibly useful rule of thumb.

These techniques—Fermi problems, scaling laws, and successive approximation—are not just tricks. They represent a way of thinking that prioritizes physical insight, allowing you to build a conceptual model of a system before getting lost in formal calculations.

Conclusion

In this lesson, we have moved from the formal world of dimensional analysis into the more intuitive and often more practical realm of order-of-magnitude estimation. You've learned how to develop a "feel for the numbers" that govern physical systems, a critical skill for any theorist sizing up a new problem.

Key Takeaways:

• Order-of-magnitude estimation is about knowing what to ignore and focusing on the "big part" of a problem to gain rapid physical insight.
• Fermi Problems allow you to estimate seemingly unknowable quantities by breaking them down into a chain of simpler, guessable parts, relying on the cancellation of errors.
• Scaling Arguments describe how a quantity $Y$ changes as another quantity $X$ changes, typically as a power law $Y\propto X^a$. This is fundamental to understanding how systems behave across different scales.
• Successive Approximation is an iterative method to refine an initial estimate. By solving a simplified version of a problem first and then using that solution to calculate corrections, you can converge on an accurate answer while maintaining physical clarity.
• The fractional change approximation, $\Delta y/y\approx n(\Delta x/x)$, is a powerful linearization tool for analyzing the effect of small perturbations in any power-law system.

We have now seen how to systematically find dimensionless groups and how to use them, along with other heuristic methods, to estimate and scale physical quantities. In our next lesson, we will explore how to extract scaling exponents from data or theoretical models, a crucial step in identifying universal behaviors and classifying physical systems into universality classes.