Domain Limitations in Models — I can explain what it means...

SECTION 1

Historical Context & Motivation

People have used mathematical models to describe the real world for thousands of years. Ancient astronomers created models to predict planetary motion, engineers built formulas for projectile paths, and economists developed equations to forecast market behavior. In every case, the creators eventually discovered the same humbling truth: no model works perfectly everywhere. Each model is valid only within a specific range of inputs — its domain. Outside that domain, the model's outputs become unrealistic, absurd, or even dangerous to rely on.

~240 BC

Eratosthenes Measures the Earth

Eratosthenes used a simple geometric model to estimate Earth's circumference. His model worked brilliantly for the specific latitudes he measured, but it assumed a perfectly spherical Earth — an assumption that breaks down when you need extreme precision.

1687

Newton's Laws of Motion

Newton's equations modeled motion with remarkable accuracy for everyday speeds. However, at speeds approaching the speed of light, these models produce unrealistic predictions, which Einstein later corrected with relativity.

1798

Malthus and Population Growth

Thomas Malthus used an exponential model to predict that human population would outpace food supply. The model was reasonable for short periods but became unrealistic over long time horizons because it ignored factors like technology and resource limits.

2020s

Modern Data Modeling

Today, scientists and analysts routinely specify the domain of validity for every model they build — from climate projections to medical dosage formulas. Understanding domain limitations is a core skill in data literacy.

The central question this lesson addresses is: How do we decide where a model's predictions are trustworthy, and what goes wrong when we push a model past its limits? Learning to identify domain limitations transforms you from someone who blindly follows a formula into someone who thinks critically about what that formula can — and cannot — tell you.

SECTION 2

Core Principles & Definitions

Before diving into examples, let's establish the key vocabulary. A mathematical model is an equation or function that represents a real-world situation. The domain of a model is the set of input values for which the model produces realistic, meaningful outputs. When we say a model is unrealistic outside its domain, we mean that plugging in input values beyond the valid range gives results that don't match reality — predictions that are nonsensical, physically impossible, or dangerously misleading.

Model Domain

The specific range of input values (x-values) for which the model accurately represents the real-world situation. This is often narrower than the mathematical domain of the function itself.

Extrapolation

Using a model to make predictions outside the range of data or conditions that were used to create it. Extrapolation is risky because the real-world relationship may change beyond the observed range.

Interpolation

Using a model to make predictions within the range of known data. Interpolation is generally more reliable than extrapolation because the model was built to fit this region.

Context Constraints

Real-world conditions that limit the domain — such as time that cannot be negative, populations that must be whole numbers, or temperatures that cannot drop below absolute zero.

✦ KEY TAKEAWAY

Think of a model like a map of your neighborhood. The map is incredibly useful for navigating the streets near your house, but if you try to use it to navigate a city three states away, it's useless. The map didn't suddenly become "wrong" — you just went outside its domain. Every model is a map that only covers a certain territory.

SECTION 3

Visual Explanation

The diagram below shows a common scenario: a quadratic model was created to describe the height of a ball thrown into the air. Within the realistic domain (from the moment the ball is thrown until it hits the ground), the model accurately describes the ball's path. Outside that domain, the model predicts that the ball would go underground — a result that makes no physical sense.

The solid cyan curve represents the ball's height within the realistic domain (0 ≤ t ≤ 4 seconds). The dashed red curve shows what the model predicts after the ball has already landed — negative heights that are physically impossible. The vertical red dashed line marks the boundary of the realistic domain.

Notice the key distinction in the diagram: the mathematical function continues beyond t = 4, but the model's validity does not. The equation itself has no problem producing an output when you plug in t = 5, but that output (a negative height) doesn't correspond to anything in the real world. This is the essence of domain limitations: the math keeps going, but the real-world meaning runs out.

SECTION 4

Mathematical Framework

Let's formalize what's happening with equations. When a model is written as a function, we need to distinguish between the mathematical domain (all inputs where the equation can produce a number) and the model domain (the subset of inputs where the output is realistic). The model domain is always equal to or smaller than the mathematical domain.

PROJECTILE HEIGHT MODEL

h(t) = −16t² + 64t

h(t) = height in feet, t = time in seconds. Mathematical domain: all real numbers. Model domain: 0 ≤ t ≤ 4 (ball is thrown at t = 0 and lands at t = 4).

To find the model domain, we ask: For which values of the input does the output still make sense in context? For the height model, h(t) must be greater than or equal to zero, since a ball can't have negative height. Setting h(t) = 0 and solving gives us the boundary values.

FINDING THE DOMAIN BOUNDARY

−16t² + 64t = 0 → −16t(t − 4) = 0 → t = 0 or t = 4

The ball is on the ground at t = 0 (when thrown) and t = 4 (when it lands). Between these values, the height is positive and realistic. Before t = 0 and after t = 4, the model predicts negative heights — unrealistic results.

LINEAR GROWTH MODEL EXAMPLE

P(t) = 200 + 15t

P(t) = population of fish in a lake, t = years after observation began. The model predicts the population grows by 15 fish per year indefinitely. Model domain: perhaps 0 ≤ t ≤ 20 (the range for which data was collected). Beyond t = 20, the lake may reach capacity, making constant growth unrealistic.

📌 Two Types of Domain Limits

Domain boundaries come from two sources. Physical constraints are built into the real-world scenario (time can't be negative, population can't be fractional). Data constraints arise because the model was built from a limited set of observations and may not hold beyond that range.

SECTION 5

How Models Break Down Outside Their Domain

Models can fail outside their domain in several distinct ways. Understanding how a model breaks down helps you anticipate problems before they occur. The diagram below illustrates three common types of model failure when extended past their realistic domain.

Three common types of model breakdown: (1) the model produces physically impossible negative outputs, (2) the model predicts unlimited growth that can't happen in a finite world, and (3) the model's linear trend continues in one direction while real data curves in another. The red dashed lines and red data points indicate the unrealistic zone in each case.

Common types of model breakdown outside the realistic domain

Type of Breakdown	What Happens	Real-World Example
Negative Output	Model gives values below zero when the quantity can't be negative	Height model predicts the ball is underground; profit model predicts −$500 in year 20
Unbounded Growth	Model predicts infinite increase, ignoring real-world limits	Linear population model predicts 10 billion rabbits in a small forest
Wrong Trend	Model predicts increasing values when the real quantity starts decreasing, or vice versa	Linear model of ice cream sales keeps rising even as winter approaches

SECTION 6

Worked Example

A coffee shop models its daily profit based on the number of cups sold using the equation P(c) = −0.01c² + 8c − 200. Let's determine the realistic domain of this model and explain why the model breaks down outside that domain.

Finding the Realistic Domain of a Profit Model

Step 1 — Understand the ContextP(c) represents daily profit in dollars, and c represents the number of cups of coffee sold. In reality, c must be a non-negative whole number (you can't sell −10 cups or 3.7 cups). Also, profit below a certain point may not make physical sense for our question.

Step 2 — Find Where Profit Equals ZeroSet P(c) = 0 and solve: −0.01c² + 8c − 200 = 0. Multiply everything by −100 to clear the decimal: c² − 800c + 20,000 = 0. Using the quadratic formula: c = (800 ± √(640,000 − 80,000)) ÷ 2 = (800 ± √560,000) ÷ 2 = (800 ± 748.33) ÷ 2.

c ≈ 25.8 or c ≈ 774.2

Step 3 — Identify Physical ConstraintsThe coffee shop can't sell negative cups, so c ≥ 0. Also, there's a practical maximum — perhaps the shop can serve at most 500 cups in a day. The profit is positive (the business is making money) only when 26 ≤ c ≤ 774, but the physical capacity caps it at c = 500.

Step 4 — State the Realistic DomainThe model domain is 0 ≤ c ≤ 500 (accounting for physical capacity). Realistically, the shop needs at least 26 cups to break even. Within this range, the model gives plausible profit predictions.

Realistic domain: 0 ≤ c ≤ 500

Step 5 — Explain the BreakdownIf we plug in c = 900 (which exceeds the shop's physical capacity), the model gives P(900) = −0.01(810,000) + 8(900) − 200 = −8,100 + 7,200 − 200 = −1,100. The model predicts a loss of $1,100, which might seem reasonable in the equation but is unrealistic because the shop simply cannot sell 900 cups in a day. The model was never designed for inputs that large.

P(900) = −$1,100 — unrealistic because c = 900 is outside the model domain

SECTION 7

Strengths and Limitations of Models

Recognizing domain limitations doesn't mean models are bad — it means we need to use them wisely. Every model offers a trade-off: it simplifies reality enough to be useful, but that simplification always has limits. The table below contrasts what models do well with where they fall short.

Models are powerful within their domain but unreliable outside it

Strengths (Within Domain)	Limitations (Outside Domain)
Provide accurate predictions based on observed data	Predictions become increasingly inaccurate the farther you go beyond the domain
Simplify complex situations into manageable equations	The simplifications ignore factors that become important outside the observed range
Allow interpolation — estimating between known data points	Extrapolation beyond the data range is speculative and often misleading
Help identify trends and make informed decisions	Can produce physically impossible results (negative time, infinite population)

✦ KEY TAKEAWAY

A model is like a weather forecast. A 3-day forecast is usually quite accurate — that's within its domain. A 30-day forecast uses the same general approach, but it becomes wildly unreliable. The model didn't change; the time horizon exceeded its valid domain. Always ask: am I within the range where this model was designed to work?

SECTION 8

Connection to Advanced Modeling

Understanding domain limitations in Math 1 builds the foundation for more sophisticated modeling in future courses. As you advance, you'll encounter techniques that address some of the limitations we've discussed. The table below shows how the basic ideas from this lesson evolve in more advanced settings.

How domain limitation concepts connect to advanced coursework

Math 1 Concept	Advanced Extension
Identifying that a model is unrealistic outside its domain	Piecewise functions that switch equations at domain boundaries to stay realistic across a wider range
Recognizing that linear models can't grow forever	Logistic growth models that include a carrying capacity, naturally leveling off instead of growing without bound
Noting physical constraints like non-negative values	Constrained optimization in calculus, where you formally define boundaries and find the best solution within them
Extrapolation is risky	Statistical confidence intervals that quantify how uncertain a prediction becomes as you move away from the data range

The critical thinking skill you're building now — asking "Is this prediction realistic?" — is exactly what data scientists, engineers, and researchers do every day. The tools get more powerful, but the fundamental question remains the same: Does my model still make sense for this input?

SECTION 9

Practice Problems

PROBLEM 1 — CONCEPTUAL

A model predicts the temperature of a cup of hot chocolate over time using T(t) = 180 − 5t, where T is in degrees Fahrenheit and t is in minutes. Explain why this model is unrealistic for very large values of t. What specific output signals that you've left the realistic domain?

PROBLEM 2 — BASIC CALCULATION

A ball is launched from a 96-foot-tall building. Its height is modeled by h(t) = −16t² + 32t + 96, where h is in feet and t is in seconds. Find the values of t where h(t) = 0, and use them to state the realistic domain of the model.

PROBLEM 3 — INTERMEDIATE

A company's revenue is modeled by R(x) = −2x² + 120x, where x is the price of a product in dollars and R is the revenue in thousands of dollars. (a) Find the x-values where R(x) = 0. (b) Does the domain need further restriction beyond 0 ≤ x ≤ 60? Explain your reasoning.

PROBLEM 4 — APPLIED

A biologist models the number of bacteria in a petri dish with B(t) = 50 + 30t, where B is the number of bacteria and t is the number of hours since the experiment started. The petri dish can hold a maximum of 800 bacteria. (a) State the realistic domain of this model. (b) What does the model predict at t = 30 hours? (c) Explain why this prediction is unrealistic and what real-world factor the model ignores.

PROBLEM 5 — CRITICAL THINKING

Two students are debating about the model C(d) = 0.25d + 2, which represents the cost in dollars of a taxi ride where d is the distance in miles. Student A says the domain should be d ≥ 0 because distance can't be negative. Student B says the domain should be 0 ≤ d ≤ 50 because the taxi only operates within a 50-mile radius and the model was built from data within that range. Who makes the stronger argument, and why? Could there be additional domain restrictions neither student considered?

SUMMARY

Lesson Summary

A mathematical model uses equations to represent real-world situations, but every model has a realistic domain — the range of input values for which its outputs are meaningful. Outside this domain, models can produce negative outputs for quantities that can't be negative, unbounded growth that ignores physical limits, or predictions that follow the wrong trend compared to actual data.

To identify the realistic domain, consider both physical constraints (non-negative time, maximum capacity, temperature floors) and data constraints (the range of observations used to build the model). Interpolation (predicting within the data range) is generally safe, while extrapolation (predicting beyond it) requires caution. The key skill is asking: "Does this output make sense in the real world?" If not, you've likely gone beyond the model's valid domain.