Why Check an Integral Model?
An integral can be evaluated correctly while still answering the wrong question. The rate might be measured in the wrong units, the interval might not match the stated time period, or a signed integral might be mistaken for a total amount. A numerical result alone cannot reveal all of these problems.
A useful check asks whether the model and its answer agree with the situation in several independent ways: units, interval, sign, and magnitude. When the rate has known bounds, those bounds also give a quick test of whether the computed accumulation is possible.
What the Integral Is Supposed to Represent
Suppose \(r(t)\) is a rate of change on a time interval from \(a\) to \(b\), where \(a<b\). The definite integral \(\int_a^b r(t)\,dt\) represents the signed accumulation of the quantity described by that rate over that interval. For example, if \(r\) is measured in litres per minute and \(t\) in minutes, the integral is measured in litres.
If \(r(t)\) is nonnegative throughout the interval, the signed accumulation equals the total amount added. If \(r(t)\) can be negative, the signed integral is net change: positive and negative contributions can cancel. A gross total that counts amount in either direction instead uses the absolute rate, \(\int_a^b |r(t)|\,dt\). These interpretations are different, so the meaning of the model must be identified before interpreting its value.
For a quantity \(f(x)\) on an interval \([a,b]\), its average value is \(\frac{1}{b-a}\int_a^b f(x)\,dx\). The factor \(1/(b-a)\) divides the accumulated quantity by the interval length. Consequently, the average has the same units as \(f\), not the units of its integral.
Four Checks to Make
Multiply the units of the integrand by the units of the integration variable. For a rate per minute integrated over minutes, the result is an amount. For an average value, divide the integral's units by the interval's units to recover the units of the original function.
Use the interval that matches the stated start and end. If time is measured in hours, convert a duration given in minutes before using it as an integration limit.
A nonnegative rate cannot produce a negative accumulation over an interval in increasing order. If a rate changes sign, decide whether the requested quantity is net change or gross amount.
Compare the result with rate bounds, average bounds, or a simple estimate. A value outside a justified range signals an error in the setup, computation, or interpretation.
Bounds That Test an Integral
Suppose \(r\) is integrable on \([a,b]\) and there are constants \(m\) and \(M\) such that \(m\leq r(t)\leq M\) throughout the interval, with \(a<b\). The integral must then lie between the constant-rate accumulations \(m(b-a)\) and \(M(b-a)\).
Here is the justification. Since \(r(t)-m\geq 0\), its integral over \([a,b]\) is nonnegative. Thus \(\int_a^b r(t)\,dt-m(b-a)\geq 0\), which gives the lower bound. Likewise, \(M-r(t)\geq 0\), so \(\int_a^b (M-r(t))\,dt\geq 0\), giving the upper bound. Dividing all parts of the resulting inequality by the positive number \(b-a\) proves the average-rate bound.
The same reasoning applies to a function \(f\) being averaged: if \(m\leq f(x)\leq M\) on \([a,b]\), then its average value must lie between \(m\) and \(M\). These are necessary checks. Passing them supports a model's plausibility, but does not by itself prove that the model describes the real situation accurately.
Worked Examples
Worked Example: Checking a Flow Accumulation Against Rate Bounds
A flow rate is \(q(t)=3+0.5t\) litres per minute for \(2\leq t\leq 6\), with \(t\) measured in minutes. Check the predicted volume.
First, the units are consistent: litres per minute multiplied by minutes gives litres. The interval lasts \(6-2=4\) minutes. On this interval the rate increases from \(q(2)=3+0.5(2)=4\) to \(q(6)=3+0.5(6)=6\) litres per minute, so the volume must be between \(4(4)=16\) litres and \(6(4)=24\) litres.
Now evaluate the integral:
The answer is positive and lies between \(16\) L and \(24\) L. Its average rate is \(20\text{ L}/4\text{ min}=5\) L/min, which is between the endpoint rates of \(4\) L/min and \(6\) L/min. As a check on the antiderivative, the derivative of \(3t+0.25t^2\) is \(3+0.5t\), the original rate. The model passes these consistency checks.
Worked Example: Checking an Average Temperature
Along a rod, let \(x\) be distance in metres from one end, and let \(T(x)=18+2x-\frac{x^2}{2}\) degrees Celsius for \(0\leq x\leq 2\). Find and check the average temperature along this interval.
The integral of temperature over distance has units of degree Celsius-metres. Dividing by the rod segment's length, \(2\) m, gives an average in degrees Celsius. The average is
To check the range, \(T'(x)=2-x\), which is nonnegative for \(0\leq x\leq2\). Thus \(T\) increases from \(T(0)=18^\circ\mathrm{C}\) to \(T(2)=20^\circ\mathrm{C}\). The average \(58/3=19\frac{1}{3}^\circ\mathrm{C}\) lies between \(18^\circ\mathrm{C}\) and \(20^\circ\mathrm{C}\), as required. Also, the antiderivative differentiates to \(18+2x-\frac{x^2}{2}\), confirming the integral calculation.
Worked Example: Separating Net Change from Gross Amount
A reservoir's net flow rate is \(r(t)=2-t\) cubic metres per minute for \(0\leq t\leq4\), where positive values mean inflow and negative values mean outflow. Find the net change and the gross amount of flow.
The rate is positive on \(0\leq t<2\), zero at \(t=2\), and negative on \(2<t\leq4\). The signed integral gives net change:
This does not mean no water moved. The positive contribution is \(\int_0^2(2-t)\,dt=[2t-t^2/2]_0^2=4-2=2\text{ m}^3\). The magnitude of the negative contribution is \(-\int_2^4(2-t)\,dt=-[(2t-t^2/2)]_2^4=-(0-2)=2\text{ m}^3\). Therefore the gross amount is \(2+2=4\text{ m}^3\).
The check is consistent with the signs: inflow and outflow cancel in the net change, while both count toward gross flow. The units are cubic metres because the rate's units, cubic metres per minute, are multiplied by minutes. An answer of zero is correct for net change but would be incorrect if reported as the gross amount.
A Common Mistake: Using the Wrong Time Unit
Suppose a machine uses fuel at \(12\) litres per hour for \(15\) minutes. A tempting but incorrect calculation is \(12(15)=180\) litres. This treats the number \(15\) as hours, even though the stated duration is in minutes. The units expose the error: litres per hour multiplied by minutes is not litres unless the minutes are converted.
Convert the interval first: \(15\text{ min}=15/60\text{ h}=0.25\text{ h}\). For a constant rate, the accumulated fuel is \(12\text{ L/h}\times0.25\text{ h}=3\text{ L}\). The result is positive and reasonable for a quarter-hour at \(12\) litres per hour. Checking the units before trusting the arithmetic prevents a factor-of-sixty error.
Using the Checks Together
No single check catches every problem. Units can be right while the interval is wrong; the sign can be right while the magnitude is impossible. A reliable review uses all the information available about the situation.
- Write down what the integral is intended to measure before evaluating it.
- Confirm the integration limits describe the stated interval in compatible units.
- Determine whether the integrand is nonnegative or changes sign, and distinguish net from gross accumulation when necessary.
- Use known lower and upper bounds to test the integral or average.
- Compare the final units and magnitude with the requested quantity.
These checks are especially valuable when interpreting a result in context. A result that passes them is internally consistent with the stated model; additional evidence may still be needed to decide whether the model itself is a good description of the real process.
Check Your Understanding
For each question, explain which consistency check is relevant and show the reasoning needed to answer it.
- A rate is measured in kilograms per minute and integrated over a time interval in minutes. What are the units of the integral?
- If a rate stays between \(3\) and \(7\) units per hour for \(5\) hours, what bounds must hold for its signed accumulation?
- A signed rate is positive for part of an interval and negative for the rest. Does a zero integral imply that no quantity accumulated in either direction? Explain.
- A quantity is between \(10^\circ\mathrm{C}\) and \(16^\circ\mathrm{C}\) everywhere on an interval. What must be true of its average value?
- A constant rate of \(8\) litres per hour acts for \(30\) minutes. What interval in hours should be used, and what accumulation does it predict?
