The Central Idea
A mathematical model connects a real or imagined system to mathematical objects that we can analyze.
At the most basic level, modeling asks three questions:
- What quantities matter?
- How are those quantities related?
- What mechanism or process explains the relationship?
These questions form a useful modeling chain:
\[ \boxed{\text{system} \longrightarrow \text{quantities} \longrightarrow \text{relationships} \longrightarrow \text{mechanism} \longrightarrow \text{model}} \]The chain is not always one-way. Modelers often move backward and forward, revising quantities or relationships after discovering that an earlier choice does not represent the system well.
What Is a Mathematical Model?
A mathematical model is a mathematical representation of selected features of a system.
The system might be physical, biological, financial, social, computational, engineered, or entirely theoretical.
A model may use:
- numbers and measured quantities;
- variables and parameters;
- equations and inequalities;
- functions and graphs;
- systems of equations;
- geometric objects;
- probability distributions;
- algorithms and simulations;
- logical or discrete structures.
The mathematics depends on what the model is trying to represent and what questions the model is intended to answer.
A Model Is a Representation, Not the System Itself
A map is not the territory it represents. In the same way, a mathematical model is not the real system.
A model intentionally leaves things out.
Suppose we model the distance traveled by a car moving at constant speed. We may write
\[ d=vt. \]This relationship ignores acceleration, traffic, hills, turns, tire slip, reaction time, and many other details.
That simplification is not automatically a weakness. If the question is simply how far a car travels in \(3\) hours at a steady \(60\) miles per hour, then
\[ d=(60)(3)=180 \]may be exactly the level of detail we need.
Quantities Are the Measurable or Countable Features
A quantity is a feature that can be represented numerically.
Examples include:
- time;
- distance;
- temperature;
- mass;
- population;
- price;
- revenue;
- probability;
- number of customers;
- amount of data;
- rate of flow;
- energy consumed.
A model usually begins by deciding which quantities are important enough to include.
Not Every Feature Needs to Become a Quantity
Real systems contain far more information than most useful models require.
Suppose a coffee shop wants to model daily revenue. Possible quantities include:
- number of customers;
- average purchase per customer;
- hours open;
- number of employees;
- temperature outside;
- advertising spending;
- day of the week;
- inventory availability.
The color of the walls could also be recorded, but it may not be useful for the question being studied.
Modeling therefore begins with selection.
The Question Determines the Quantities
There is no universal list of quantities that must appear in a model.
The quantities depend on the problem.
If we ask,
How much fuel will a trip require?
useful quantities might include distance and fuel economy.
If we ask,
How long will the trip take?
speed, distance, stops, and traffic conditions may matter instead.
The same real-world system can therefore support many different models.
Representing Quantities with Symbols
We usually assign symbols to quantities so that relationships can be written compactly.
For example, let
\[ d=\text{distance traveled}, \] \[ v=\text{speed}, \]and
\[ t=\text{time}. \]Then the verbal relationship
distance equals speed multiplied by time
becomes
\[ \boxed{d=vt}. \]The symbols are useful only because their meanings have been defined.
Always Attach Meaning to a Symbol
In pure algebra, a symbol can be manipulated without much context.
In modeling, context matters.
If \(t\) represents time, we should know whether it means:
- seconds;
- minutes;
- hours;
- days;
- time since a particular starting event.
A model becomes much easier to interpret when each symbol has a clearly stated meaning and unit.
Units Are Part of the Quantity
A numerical value without a unit can be incomplete.
The statement
\[ v=60 \]does not tell us whether the speed is \(60\) miles per hour, \(60\) kilometers per hour, or \(60\) meters per second.
Units also help us check relationships.
If
\[ d=vt, \]then
\[ \frac{\text{miles}}{\text{hour}} \cdot \text{hour} = \text{miles}. \]The units are consistent with distance.
Relationships Connect Quantities
After identifying quantities, we ask how they are connected.
A relationship describes how one or more quantities vary with, constrain, depend on, or interact with other quantities.
A relationship might be expressed as:
- an equation;
- an inequality;
- a function;
- a table;
- a graph;
- a recurrence;
- a probability distribution;
- a system of equations;
- an algorithmic rule.
Example: A Direct Relationship
Suppose a worker earns \(\$24\) per hour.
Let
\[ h=\text{hours worked} \]and
\[ P=\text{gross pay in dollars}. \]If the hourly rate is fixed, then
\[ \boxed{P=24h}. \]The relationship tells us that each additional hour adds another \(\$24\) to gross pay.
Relationships Can Include Fixed and Variable Parts
Suppose a service charges a \(\$35\) visit fee plus \(\$80\) per hour of labor.
Let \(h\) be labor hours and \(C\) be total cost.
Then
\[ \boxed{C=35+80h}. \]The \(35\) is a fixed contribution. The term \(80h\) changes with the amount of labor.
A model can therefore combine several relationships or mechanisms in the same expression.
Relationships Can Impose Constraints
Not every relationship tells us that two quantities are equal.
Suppose a truck can carry at most \(2{,}000\) kilograms.
If \(L\) is the load in kilograms, then
\[ \boxed{L\le 2000}. \]This inequality describes a constraint rather than a formula for computing the load.
Many practical models contain equations and constraints at the same time.
Relationships Can Involve Several Quantities
Suppose a rectangular storage area has length \(L\), width \(W\), and area \(A\).
Then
\[ \boxed{A=LW}. \]If the total area is fixed at \(600\) square meters, then
\[ LW=600. \]The relationship does not determine \(L\) and \(W\) individually. Many pairs satisfy the equation.
Modeling often requires us to distinguish between what is known, what is unknown, and what is merely constrained.
A Mechanism Explains How the Relationship Arises
A mechanism is the process, rule, interaction, or structure that produces the relationship represented by the model.
Consider
\[ A=LW. \]The equation is a relationship.
The geometric mechanism is that a rectangle can be covered by rows and columns of unit squares. Multiplying length by width counts the resulting area.
Understanding the mechanism helps us know when the relationship should be trusted.
Mechanisms Can Be Physical
Suppose water enters a tank at a constant rate of \(5\) liters per minute.
If the tank initially contains \(20\) liters, then after \(t\) minutes the volume is
\[ \boxed{V=20+5t}. \]The mechanism is accumulation:
\[ \text{new amount} = \text{initial amount} + \text{amount added}. \]The coefficient \(5\) comes from the inflow rate.
Mechanisms Can Be Financial
Suppose an account starts with principal \(P\) and earns simple interest at annual rate \(r\).
Under simple interest, the interest added each year is based on the original principal:
\[ I=Prt. \]The account value becomes
\[ \boxed{A=P+Prt=P(1+rt)}. \]The formula is not merely an algebraic pattern. It represents a specific financial mechanism: interest is calculated from the original principal, not repeatedly from the growing balance.
Mechanisms Can Be Discrete
Suppose a population begins with \(100\) organisms and doubles once during each time step.
If \(P_n\) is the population after \(n\) steps, then
\[ P_0=100 \]and
\[ \boxed{P_{n+1}=2P_n}. \]The mechanism is repeated multiplication by \(2\).
The same mechanism can also be summarized explicitly as
\[ \boxed{P_n=100(2^n)}. \]Mechanisms Can Be Probabilistic
Some systems do not determine one exact outcome.
Suppose a machine produces an acceptable part with probability \(0.98\) on each independent trial.
If \(X\) is the number of acceptable parts among \(100\) produced parts, then a binomial model may be appropriate:
\[ X\sim\operatorname{Binomial}(100,0.98). \]The model describes a probability mechanism rather than an exact deterministic relationship.
Later lessons will distinguish deterministic and stochastic models more systematically.
Relationship Does Not Automatically Mean Mechanism
Two quantities may move together without one directly producing the other.
A data set might show that \(x\) and \(y\) are strongly associated. That empirical relationship can be useful for prediction.
But the observed association alone does not identify the mechanism.
A third quantity might influence both, the direction of influence might be reversed, or the relationship might arise from selection or measurement.
Good modeling keeps the distinction clear:
\[ \boxed{\text{observed relationship}\ne\text{established mechanism}} \]Mechanisms Suggest Mathematical Form
A mechanism often gives clues about which mathematical relationship to use.
For example:
- constant addition suggests a linear or arithmetic structure;
- constant multiplication suggests an exponential or geometric structure;
- accumulation suggests sums or integrals;
- competition for limited capacity may suggest nonlinear saturation;
- repeated random trials may suggest a probability distribution;
- movement through states may suggest a transition model;
- balance laws often suggest equations linking inflow, outflow, and storage.
The mathematics should reflect the process we believe is operating.
The Same Relationship Can Come from Different Mechanisms
A linear graph does not tell us by itself why the relationship is linear.
The expression
\[ y=50+8x \]could represent:
- a fixed fee of \(50\) plus \(8\) per unit;
- an initial inventory of \(50\) plus \(8\) units arriving each hour;
- a starting elevation of \(50\) meters plus a rise of \(8\) meters per kilometer;
- a baseline score of \(50\) plus \(8\) points for each completed task.
The algebraic form is the same, but the interpretations and mechanisms are different.
The Same Mechanism Can Appear in Different Contexts
Conversely, the same mathematical mechanism can describe many systems.
Repeated multiplication can model:
- compound growth;
- population reproduction;
- repeated percentage change;
- certain decay processes;
- geometric sequences.
Recognizing these shared structures is one of the main reasons mathematical modeling is powerful.
Inputs, Outputs, and State Quantities
A useful way to organize a system is to separate quantities by their role.
An input is information supplied to the model.
An output is information produced by the model.
A state quantity describes the condition of the system at a particular moment or stage.
For a tank model:
- inflow rate may be an input;
- current volume may be a state quantity;
- predicted time until overflow may be an output.
These roles can change depending on the question.
Stocks and Flows
Many dynamic systems are built around two kinds of quantities.
A stock is an amount currently present.
A flow is a rate at which that amount enters, leaves, or changes.
Examples include:
- water volume and water flow rate;
- bank balance and cash flow;
- inventory and shipment rate;
- population and birth or death rates;
- energy stored and power transfer.
A common modeling structure is
\[ \boxed{\text{change in stock} = \text{inflow} - \text{outflow}}. \]Example: Inventory as a Stock-and-Flow Model
Suppose a warehouse begins with \(500\) units.
During one day, \(120\) units arrive and \(175\) units are shipped.
The ending inventory is
\[ 500+120-175=445. \]In symbolic form,
\[ \boxed{I_{\text{new}} = I_{\text{old}} + A - S}, \]where \(A\) is arrivals and \(S\) is shipments.
This balance relationship is reusable across many inventory systems.
Rates Describe How Quantities Change
A rate compares change in one quantity with change in another.
Average speed is
\[ \frac{\Delta d}{\Delta t}. \]Average cost per unit is
\[ \frac{\Delta C}{\Delta q}. \]A population growth rate might compare
\[ \frac{\Delta P}{\Delta t}. \]Rates are central in modeling because many mechanisms are naturally described by how quickly something changes.
Constant Rate Produces a Linear Model
Suppose a quantity begins at \(Q_0\) and changes at a constant rate \(r\) per unit of time.
After time \(t\),
\[ \boxed{Q(t)=Q_0+rt}. \]The mechanism is constant additive change.
If \(r>0\), the quantity increases.
If \(r<0\), the quantity decreases.
The linear form comes from the mechanism, not merely from choosing a line because it looks convenient.
Constant Percentage Change Produces a Different Structure
Suppose instead that a quantity grows by the same proportion during each time period.
If the growth factor per period is \(g\), then
\[ Q_{n+1}=gQ_n. \]Repeating the mechanism gives
\[ \boxed{Q_n=Q_0g^n}. \]This is fundamentally different from adding a constant amount each period.
Choosing between linear and exponential models therefore requires thinking about the mechanism of change.
Mechanism First Can Prevent a Bad Model
Suppose a quantity increases by \(5\%\) each year.
A linear model that adds the same numerical amount each year does not represent that mechanism correctly.
Percentage growth means the amount added depends on the current quantity.
If the current amount is \(Q_n\), then
\[ Q_{n+1}=Q_n+0.05Q_n \]so
\[ \boxed{Q_{n+1}=1.05Q_n}. \]The mechanism points naturally toward exponential growth.
Mechanism and Evidence Work Together
A model should not be chosen from theory alone when observations are available.
We may believe a mechanism suggests a certain relationship, then compare that relationship with data.
If the model disagrees substantially with observation, several things may be wrong:
- the mechanism may be incomplete;
- an important quantity may be missing;
- a simplifying assumption may fail;
- measurements may be noisy or biased;
- the mathematical form may be inappropriate.
Later lessons will formalize calibration, validation, sensitivity, and model revision.
From Words to a Model
Consider the statement:
A parking garage charges \(\$6\) to enter and \(\$3\) for each hour parked.
Step 1: identify the quantities.
\[ h=\text{hours parked}, \] \[ C=\text{total cost in dollars}. \]Step 2: identify the relationship.
Total cost equals the entry fee plus an hourly charge.
Step 3: identify the mechanism.
One fixed charge occurs once, while another charge accumulates at a constant rate per hour.
Step 4: write the model.
\[ \boxed{C=6+3h}. \]Example: A Mixing Relationship
Suppose a tank contains \(40\) liters of a solution that is \(25\%\) salt by concentration.
The amount of salt is
\[ 0.25(40)=10 \]liters of salt equivalent.
The basic relationship is
\[ \boxed{\text{amount of component} = \text{concentration} \times \text{total amount}}. \]This structure can be reused in mixture, chemistry, finance, and weighted composition problems.
Example: A Resource Constraint
Suppose a workshop has \(240\) labor hours available.
Product \(A\) requires \(3\) labor hours per unit and product \(B\) requires \(5\) labor hours per unit.
If \(x\) and \(y\) are the numbers produced, then labor use is
\[ 3x+5y. \]The capacity constraint is
\[ \boxed{3x+5y\le240}. \]The mechanism is resource consumption: each unit of production uses some of a limited resource.
Example: Two Relationships Form a System
Suppose two ticket types are sold.
Adult tickets cost \(\$12\), student tickets cost \(\$8\), \(300\) tickets are sold, and total revenue is \(\$3{,}200\).
Let
\[ a=\text{adult tickets} \]and
\[ s=\text{student tickets}. \]The count mechanism gives
\[ a+s=300. \]The revenue mechanism gives
\[ 12a+8s=3200. \]Together,
\[ \boxed{ \begin{aligned} a+s&=300,\\ 12a+8s&=3200. \end{aligned}} \]One system can therefore contain several relationships generated by different aspects of the same situation.
Models Can Be Built at Different Levels of Detail
Consider modeling the motion of a falling object.
A simple model might assume constant gravitational acceleration and ignore air resistance.
A more detailed model might include:
- air drag;
- changing air density;
- wind;
- rotation;
- shape and orientation.
More detail does not automatically mean a better model.
The useful level of complexity depends on the purpose of the model.
Start Simple, Then Add Structure When Needed
A practical modeling strategy is:
- identify the main question;
- choose the smallest useful set of quantities;
- write the simplest plausible relationships;
- connect those relationships to a mechanism;
- compare the model with what the situation requires;
- add complexity only when it improves the model's usefulness.
This keeps the mathematics aligned with the problem instead of allowing the mathematics to become an end in itself.
A Diagram Can Clarify the Model Before Algebra Begins
Sometimes a simple dependency diagram is useful.
For a basic revenue model,
\[ \text{customers} \longrightarrow \text{units sold} \longrightarrow \text{revenue}. \]For a tank,
\[ \text{inflow rate} \longrightarrow \text{volume} \longrightarrow \text{overflow time}. \]Such diagrams help identify which quantities influence others and can expose missing links before equations are chosen.
Common Error: Starting with an Equation Too Soon
A familiar equation can tempt us to force a situation into a form that does not match the mechanism.
For example, not every increasing quantity should be modeled by
\[ y=mx+b. \]Before choosing the equation, ask:
- What is changing?
- What causes the change?
- Is the change additive, multiplicative, constrained, random, or something else?
Common Error: Using Undefined Symbols
Writing
\[ y=4x+10 \]is not a complete model unless the meanings of \(x\) and \(y\) are known.
A modeling statement should identify what each important symbol represents, preferably with units.
For example:
\[ x=\text{hours of service}, \] \[ y=\text{total charge in dollars}. \]Common Error: Ignoring Units
Suppose distance is measured in miles and time in minutes.
If speed is \(60\) miles per hour, directly multiplying
\[ 60\cdot30 \]mixes hours and minutes incorrectly.
Convert first:
\[ 30\text{ minutes}=0.5\text{ hour}. \]Then
\[ d=(60)(0.5)=30\text{ miles}. \]Unit consistency is one of the fastest ways to catch modeling errors.
Common Error: Confusing Description with Explanation
A fitted curve may describe data very well without explaining why the data behave that way.
For example, a polynomial may approximate observed measurements over a limited interval.
That does not mean the underlying system operates by a polynomial mechanism.
Models can be descriptive, predictive, explanatory, or some combination of these. The purpose should be stated clearly.
Common Error: Including Everything
A model can become less useful when too many quantities are included without a clear reason.
Every added quantity can create:
- more data requirements;
- more parameters to estimate;
- more possible interactions;
- more opportunities for error;
- greater difficulty interpreting results.
Include complexity when it serves the modeling objective.
A Reliable First-Pass Modeling Procedure
- State the question the model should help answer.
- Identify the system or process being represented.
- List the quantities that appear directly in the question.
- Add other quantities that appear necessary to explain the process.
- Define symbols and units for the important quantities.
- Describe the relationships in words before writing equations.
- Identify the mechanism that could generate each relationship.
- Translate the relationships into mathematical form.
- Check units, signs, magnitudes, and obvious boundary cases.
- Ask whether the model is simple enough to use and detailed enough for its purpose.
Example: Full First-Pass Model
A delivery company charges a fixed dispatch fee of \(\$18\) plus \(\$2.40\) per mile.
We want a model for the total delivery charge.
1. Identify the quantities
\[ d=\text{delivery distance in miles}, \] \[ C=\text{total charge in dollars}. \]2. Describe the relationship
Total charge equals a fixed dispatch fee plus a distance-based charge.
3. Identify the mechanism
The dispatch fee occurs once. The mileage charge accumulates at a constant rate of \(\$2.40\) for each mile.
4. Write the model
\[ \boxed{C=18+2.40d}. \]5. Check the units
\[ \frac{\text{dollars}}{\text{mile}} \cdot \text{mile} = \text{dollars}. \]6. Interpret the model
At \(d=0\), the model gives \(C=18\), representing the fixed dispatch fee. Each additional mile increases the charge by \(\$2.40\).
Practice 1: Identify the Quantities
A bakery wants to estimate daily revenue from the number of loaves sold and the selling price per loaf.
Identify three natural quantities for the model.
Practice 2: Write the Relationship
A streaming service charges \(\$12\) per month plus \(\$4\) for each premium rental.
Let \(r\) be the number of premium rentals and \(C\) be the monthly cost. Write a model for \(C\).
Practice 3: Identify the Mechanism
The model
\[ Q(t)=75+6t \]represents a quantity that begins at \(75\).
Describe the change mechanism represented by the model.
Practice 4: Stock and Flow
A reservoir contains \(1{,}200\) cubic meters of water. During one hour, \(90\) cubic meters enter and \(140\) cubic meters leave.
Find the amount of water after the hour.
Practice 5: Resource Constraint
A machine has at most \(480\) minutes available. Product \(A\) requires \(6\) minutes per unit and product \(B\) requires \(10\) minutes per unit.
Let \(x\) and \(y\) be the numbers of units produced. Write the capacity constraint.
Practice 6: Units Check
A bicycle travels at \(18\) miles per hour for \(40\) minutes.
Find the distance traveled, showing the unit conversion.
Practice 7: Additive or Multiplicative?
A population grows by \(3\%\) each year.
Is the underlying year-to-year mechanism constant addition or constant multiplication? Write a recurrence for the population.
Practice 8: Relationship versus Mechanism
A data set shows that two variables have a strong positive correlation.
Does that fact alone establish a mechanism by which one variable causes the other? Explain.
Practice 9: Build a Simple Model
A storage company charges a one-time setup fee of \(\$25\) and \(\$9\) per week.
Define quantities, describe the mechanism, and write a model for total cost after \(w\) weeks.
Practice 10: Diagnose the Model
A student models a \(5\%\) annual growth process with
\[ Q(t)=Q_0+0.05t. \]Explain why this relationship does not generally represent \(5\%\) growth of the current quantity.
Practice Answers
Practice 1
A natural set of quantities is:
\[ n=\text{number of loaves sold}, \] \[ p=\text{price per loaf}, \]and
\[ R=\text{daily revenue}. \]A basic relationship would be \(R=np\).
Practice 2
\[ \boxed{C=12+4r}. \]The \(12\) is the fixed monthly charge and \(4r\) is the rental-dependent charge.
Practice 3
The model begins at \(75\) and adds \(6\) units for each additional unit of time. The mechanism is constant additive change.
Practice 4
\[ 1200+90-140=1150. \]Therefore,
\[ \boxed{1150\text{ cubic meters}}. \]Practice 5
\[ \boxed{6x+10y\le480}. \]Each unit consumes machine time, and total use cannot exceed the available \(480\) minutes.
Practice 6
\[ 40\text{ minutes} = \frac{40}{60}\text{ hour} = \frac{2}{3}\text{ hour}. \]Therefore,
\[ d=18\left(\frac{2}{3}\right)=12. \]The bicycle travels
\[ \boxed{12\text{ miles}}. \]Practice 7
A \(3\%\) increase means multiplication by
\[ 1+0.03=1.03. \]Therefore,
\[ \boxed{P_{n+1}=1.03P_n}. \]Practice 8
No. Correlation describes an observed relationship, but the relationship alone does not establish a causal mechanism. Additional evidence and reasoning are needed to determine why the variables move together.
Practice 9
Let
\[ w=\text{number of weeks} \]and
\[ C=\text{total cost in dollars}. \]The mechanism combines a fixed setup fee with a constant weekly charge. Therefore,
\[ \boxed{C=25+9w}. \]Practice 10
The term \(0.05t\) adds a fixed numerical amount based on time. A \(5\%\) growth process instead adds \(5\%\) of the current quantity, so the amount added changes as the quantity changes.
For discrete annual growth, the appropriate recurrence is
\[ \boxed{Q_{n+1}=1.05Q_n}. \]What to Remember
Mathematical modeling begins with structure, not with formula hunting.
The core sequence is:
\[ \boxed{\text{identify quantities} \longrightarrow \text{describe relationships} \longrightarrow \text{explain mechanisms} \longrightarrow \text{write mathematics}}. \]- Quantities are measurable or countable features of the system.
- Relationships describe how quantities depend on or constrain one another.
- Mechanisms explain the process that generates those relationships.
- Units help define quantities and test whether equations are sensible.
- The same mathematical relationship can describe different mechanisms.
- The same mechanism can appear in many different real-world contexts.
- A useful model is intentionally selective and purpose-driven.
Where We Go Next
Once the important quantities and relationships have been identified, they need precise mathematical roles.
The next lesson, Defining Variables and Parameters, develops the distinction between quantities that vary within a model and quantities that are treated as fixed for a particular modeling task.