Numerical Integrals

Using the Trapezoidal Rule and Simpson’s Rule

Mathematics is not about numbers, equations,

computations, or algorithms: it is about
understanding. “Ah, if we could only do the all the
integrals . . . But we can’t.”
— Beta Mathematica

s
or
t
Tu
i ty
fin
In

Infinity Tutors
Trapezoidal Rule
Suppose we need to approximate the integral
Z b
f (x) dx
a

using the Trapezoidal Rule with n subdivisions. Then:

Z b
hh  i
f (x) dx ≈ Tn = f (x0 ) + f (xn ) + 2 f (x1 ) + f (x2 ) + . . . + f (xn−1 )
a 2

where:
b−a
h= , xi = a + ih, for i = 0, 1, 2, . . . , n

s
n

or
Also,
x0 = a, xn = b

t
Computation Table:
Tu
i xi f (xi )
0
1
2
.. .. ..
ty

. . .
n−1
n
i
fin

Activity 1

Use Trapezoidal Rule to approximate the following integrals with the given n.
In

Z 1
2
(a). e4x dx, n = 7
0
Z 30    
140000
(b). 2000 ln − 9.8t dt, n=5
8 140000 − 2100t
Z 4
(c). x sin x dx, n=5
0
Z 4 √
(d). xex dx, n=7
0
Z 3
(e). ln x dx, n=6
1

2
 Z 2
(f). xe−x dx, n=8
0
(
Z π sin x
(g). f (x)dx, where f (x) = x , x>0
; n=6
0 1, x=0

s
t or
Tu
i ty
fin
In

3
Simpson’s Rule
Suppose we need to approximate the integral
Z b
f (x) dx
a

using Simpson’s Rule with n subdivisions (note: n must be even). Then:

Z b
hh    i
f (x) dx ≈ Sn = f (x0 )+f (xn )+4 f (x1 )+f (x3 )+. . .+f (xn−1 ) +2 f (x2 )+f (x4 )+. . .+f (xn−2 )
a 3

where:
b−a
h= , xi = a + ih, for i = 0, 1, 2, . . . , n

s
n

or
Also,
x0 = a, xn = b

t
Computation Table:
Tu
i xi f (xi )
0
1
2
ty

.. .. ..
. . .
n−1
n
i
fin

Activity 2

Use Simpson’s Rule to approximate the following integrals with the given n.
In

Z 1
2
(a). e4x dx, n = 6
0
Z 30    
140000
(b). 2000 ln − 9.8t dt, n=8
8 140000 − 2100t
Z 4
(c). x sin x dx, n=8
0
Z 4 √
(d). xex dx, n=6
0
Z 3
(e). ln x dx, n=6
1

4
 Z 2
(f). xe−x dx, n=8
0
(
Z π sin x
(g). f (x)dx, where f (x) = x , x>0
; n=6
0 1, x=0

s
t or
Tu
i ty
fin
In

5
Suggested Solutions for Activity 1
(a). We want to approximate the integral:
Z 1
2
e4x dx
0

using the Trapezoidal Rule with n = 7 subdivisions.

Here we have:

a = x0 = 0,
b = x7 = 1,
n = 7,

s
2
f (x) = e4x .

or
Now,
b−a 1−0 1

t
h= = =
n 7 7
Tu
and  
1 i
xi = a + ih = 0 + i = , for i = 0, 1, 2, . . . , 7.
7 7

Thus, we tabulate the data as follows:

ty

2
i xi f (xi ) = e4(xi )
0 0 1.00000000000
i

1
fin

1 7 1.08505714419
2
2 7 1.38615068230
3
3 7 2.08484367736
4
4 7 3.69183066048
In

5
5 7 7.69688981037
6
6 7 18.89269822665
7 1 54.59815003314

Therefore,
Z 1
2
e4x dx ≈ T7
0
h
= [f (x0 ) + f (x7 ) + 2 (f (x1 ) + f (x2 ) + f (x3 ) + f (x4 ) + f (x4 ) + f (x5 ) + f (x6 ))]
2
1
7
= [1 + 54.59815003314 + 2 (1.08505714419 + 1.38615068230 + 2.08484367736
2
+ 3.69183066048 + 7.69688981037 + 18.89269822665)] = 8.94807788827 .

6
(b). We want to approximate the integral:
Z 30    
140000
2000 ln − 9.8t dt
8 140000 − 2100t
using the Trapezoidal Rule with n = 5 subdivisions.
Here we have:
a = t0 = 8,
b = t5 = 30,
n = 5,
 
140000
f (t) = 2000 ln − 9.8t.
140000 − 2100t

s
Now,
b−a 30 − 8

or
22
h= = =
n 5 5
and  
22

t
ti = a + ih = 8 + i , for i = 0, 1, 2, 3, 4, 5.
5
Tu
Thus, we tabulate the data as follows:

i ti f (ti )
0 8 177.26674301977
ty

1 8 + 225 = 5
62
290.06982595919
2 44
8 + 5 = 84 5 416.06460201532
3 8 + 66
5 = 5
106
557.69124227735
i

4 8 + 88
5 = 5
128
718.13663089723
fin

110 150
5 8 + 5 = 5 = 30 901.67400151124

Therefore,
Z 30
In

f (t) dt ≈ T5
8
h
= [f (t0 ) + f (t5 ) + 2 (f (t1 ) + f (t2 ) + f (t3 ) + f (t4 ))]
2
22
5
= [177.26674301977 + 901.67400151124 + 2(290.06982595919 + 416.06460201532
2
+ 557.69124227735 + 718.13663089723)]
11
= [1078.94074453101 + 2(1981.96230014909)]
5
11
= [1078.94074453101 + 3963.92460029818]
5
11
= × 5042.86534482919
5
= 11094.30376302425

7
(c). We want to approximate the integral:
Z 4
x sin x dx
0

using the Trapezoidal Rule with n = 5 subdivisions.

Here we have:
a = x0 = 0,
b = x5 = 4,
n = 5,
f (x) = x sin x.

s
Now,
b−a 4−0 4

or
h= = =
n 5 5
and  
4 4i
xi = a + ih = 0 + i = , for i = 0, 1, 2, 3, 4, 5.

t
5 5
Tu
Thus, we tabulate the data as follows:

i xi f (xi ) = xi sin xi
0 0 0.00000000000
4
ty

1 5 0.57388487272
8
2 5 1.59931776487
12
3 5 1.62111163332
16
i

4 5 -0.18679725897
fin

5 4 -3.02720998123

Therefore,
Z 4
In

x sin x dx ≈ T5
0
h
= [f (x0 ) + f (x5 ) + 2 (f (x1 ) + f (x2 ) + f (x3 ) + f (x4 ))]
2
4
5
= [0 + (−3.02720998123) + 2(0.57388487272 + 1.59931776487
2
+ 1.62111163332 − 0.18679725897)]
2
= [−3.02720998123 + 2(3.60751701294)]
5
2
= [−3.02720998123 + 7.21503402588]
5
2
= × 4.18782404465
5
= 1.67512961706

8
(d). We want to approximate the integral:
Z 4 √
x ex dx
0

using the Trapezoidal Rule with n = 7 subdivisions.

Here we have:

a = x0 = 0,
b = x7 = 4,
n = 7,
√
f (x) = x ex .

s
Now,

or
b−a 4−0 4
h= = =
n 7 7

t
and  
4 4i
xi = a + ih = 0 + i
7
Tu= ,
7
for i = 0, 1, 2, . . . , 7.

Thus, we tabulate the data as follows:

√
i xi f (xi ) = xi exi
ty

0 0 0.00000000000
4
1 7 1.33859516201
8
2 3.35222008798
i

7
12
3 7.27020118627
fin

7
16
4 7 14.86565579667
20
5 7 29.43115830362
24
6 7 57.09081634954
In

7 4 109.19630006629

9
 Therefore,
Z 4
√ x
xe dx ≈ T7
0
h
= [f (x0 ) + f (x7 ) + 2(f (x1 ) + f (x2 ) + f (x3 ) + f (x4 ) + f (x5 ) + f (x6 ))]
2
4
7
= [0 + 109.19630006629 + 2(1.33859516201 + 3.35222008798 + 7.27020118627
2
+ 14.86565579667 + 29.43115830362 + 57.09081634954)]
2
= [109.19630006629 + 2(113.34864688509)]
7
2
= [109.19630006629 + 226.69729377018]
7

s
2
= × 335.89359383647

or
7
= 95.96959823956

t
(e). We want to approximate the integral:
Z 3
Tu
ln x dx
1
using the Trapezoidal Rule with n = 6 subdivisions.
Here we have:
ty

a = x0 = 1,
b = x6 = 3,
i

n = 6,
fin

f (x) = ln x.

Now,
b−a 3−1 1
h= = =
In

n 6 3
and  
1 3+i
xi = a + ih = 1 + i = , for i = 0, 1, 2, . . . , 6.
3 3
Thus, we tabulate the data as follows:

i xi f (xi ) = ln xi
3
0 3 =1 0.00000000000
4
1 3 0.28768207245
5
2 3 0.51082562377
6
3 3 =2 0.69314718056
7
4 3 0.84729786039
8
5 3 0.98082925301
9
6 3 =3 1.09861228867

10
 Therefore,
Z 3
ln x dx ≈ T6
1
h
= [f (x0 ) + f (x6 ) + 2(f (x1 ) + f (x2 ) + f (x3 ); +f (x4 ) + f (x5 ))]
2
1
3
= [0 + 1.09861228867 + 2(0.28768207245 + 0.51082562377 + 0.69314718056
2
+ 0.84729786039 + 0.98082925301)]
1
= [1.09861228867 + 2(3.31978199018)]
6
1
= [1.09861228867 + 6.63956398036]

s
6
1

or
= × 7.73817626903
6
= 1.28969604484

(f). We want to approximate the integral:

t
Tu
Z 2
xe−x dx
0

using the Trapezoidal Rule with n = 8 subdivisions.

ty

Here we have:

a = x0 = 0,
i

b = x8 = 2,
fin

n = 8,
f (x) = xe−x .
In

Now,
b−a 2−0 1
h= = =
n 8 4
and  
1 i
xi = a + ih = 0 + i = , for i = 0, 1, 2, . . . , 8.
4 4

Thus, we tabulate the data as follows:

11
 i xi f (xi ) = xi e−xi
0 0 0.00000000000
1
1 4 0.19470019577
1
2 2 0.30326532986
3
3 4 0.35427491456
4 1 0.36787944117
5
5 4 0.35813099608
3
6 2 0.33469524022
7
7 4 0.30410440104
8 2 0.27067056647

Therefore,

s
Z 2
xe−x dx ≈ T8

or
0
h
= [f (x0 ) + f (x8 ) + 2 (f (x1 ) + f (x2 ) + f (x3 ) + f (x4 ) + f (x5 ) + f (x6 ) + f (x7 ))]
2

t
1
4
=
2
Tu
[0.00000000000 + 0.27067056647 + 2(0.19470019577 + 0.30326532986 + 0.35427491456
+ 0.36787944117 + 0.35813099608 + 0.33469524022 + 0.30410440104)]
1
= [0.27067056647 + 2(2.21605051870)]
8
1
ty

= [0.27067056647 + 4.43210103740]
8
1
= × 4.70277160387
8
i
fin

= 0.58809645048

(g). We want to approximate the integral:

(
π
In

Z sin x
f (x) dx, where f (x) = x , x>0
0 1, x=0

using the Trapezoidal Rule with n = 6 subdivisions.

Here we have:
a = x0 = 0,
b = x6 = π,
n = 6,
sin x
f (x) = , with f (0) = 1.
x
Now,
b−a π−0 π
h= = =
n 6 6

12
and
π
xi = a + ih = i · , for i = 0, 1, 2, . . . , 6.
6
Thus, we tabulate the data as follows:

sin xi
i xi f (xi ) =
xi
0 0 1.00000000000
π
1 6 0.95492965855
π
2 3 0.82699334313
π
3 2 0.63661977237
2π
4 3 0.41349667157
5π
5 0.19098593171

s
6
6 π 0.00000000000

or
Therefore,

t
Z π
f (x) dx ≈ T6
0
Tu
h
= [f (x0 ) + f (x6 ) + 2 (f (x1 ) + f (x2 ) + f (x3 ) + f (x4 ) + f (x5 ))]
2
π
6
= [1.00000000000 + 0.00000000000 + 2(0.95492965855 + 0.82699334313 + 0.63661977237
2
ty

+ 0.41349667157 + 0.19098593171)]
π
= [1.00000000000 + 2(3.02202537733)]
12
i

π
fin

= [1.00000000000 + 6.04405075466]
12
π
= · 7.04405075466
12
= 1.84431223665
In

13
Suggested Solutions for Activity 2
(a). We want to approximate the integral:
Z 1
2
e4x dx
0

using the Simpson’s Rule with n = 6 subdivisions.

Here we have:

a = x0 = 0,
b = x6 = 1,
n = 6,

s
2
f (x) = e4x .

or
Now,
b−a 1−0 1

t
h= = =
n 6 6
Tu
and
1 i
xi = a + ih = i · = , for i = 0, 1, 2, . . . , 6.
6 6
Thus, we tabulate the data as follows:
ty

2
i xi f (xi ) = e4xi
0 0 1.00000000000
i

1
1 1.11751906874
fin

6
1
2 3 1.55962349761
1
3 2 2.71828182846
2
4 3 5.91669359066
5
In

5 6 16.08324067206
6 1 54.59815003314

14
 Therefore,
Z 1
2
e4x dx ≈ S6
0
h
= [f (x0 ) + f (x6 ) + 4(f (x1 ) + f (x3 ) + f (x5 )) + 2(f (x2 ) + f (x4 ))]
3
1
6
= [1.00000000000 + 54.59815003314 + 4(1.11751906874 + 2.71828182846 + 16.08324067206)
3
+ 2(1.55962349761 + 5.91669359066)]
1
= [55.59815003314 + 4(19.91904156926) + 2(7.47631708827)]
18
1
= [55.59815003314 + 79.67616627704 + 14.95263417654]

s
18
1

or
= · 150.22695048672
18
= 8.34594169371

(b). We want to approximate the integral:

t
Tu
Z 30    
140000
2000 ln − 9.8t dt
8 140000 − 2100t

using the Simpson’s Rule with n = 8 subdivisions.

ty

Here we have:

a = t0 = 8,
i

b = t8 = 30,
fin

n = 8,
 
140000
f (t) = 2000 ln − 9.8t.
140000 − 2100t
In

Now,
b−a 22 11 11 32 + 11i
h= = = , ti = a + i · h = 8 + i · =
n 8 4 4 4

Thus, we tabulate the data as follows:

15
 i ti f (ti )
0 8 177.26674301977
43
1 4 246.33518139249
27
2 2 320.24688864628
65
3 4 399.51653927417
4 19 484.74547257626
87
5 4 576.64231351786
49
6 2 676.05011676092
109
7 4 783.98267835199
8 30 901.67400151124

Therefore,

s
Z 30
f (t) dt ≈ S8

or
8
h


= f (t0 ) + f (t8 ) + 4 f (t1 ) + f (t3 ) + f (t5 ) + f (t7 ) + 2 f (t2 ) + f (t4 ) + f (t6 )
3

t
11
4
=
3
Tu
[177.26674301977 + 901.67400151124
+ 4(246.33518139249 + 399.51653927417 + 576.64231351786 + 783.98267835199)
+ 2(320.24688864628 + 484.74547257626 + 676.05011676092)]
11
= [1078.94074453101 + 4 × 2006.47671253651 + 2 × 1481.04247898346]
ty

12
11
= [1078.94074453101 + 8025.90685014604 + 2962.08495796692]
12
11
i

= × 12066.93255264397
fin

12
= 11061.35483809031

(c). We want to approximate the integral:

In

Z 4
x sin x dx
0

using the Simpson’s Rule with n = 8 subdivisions.

Here we have:
a = x0 = 0,
b = x8 = 4,
n = 8,
f (x) = x sin x.

Now,
b−a 4−0 1 1 i
h= = = , xi = a + ih = 0 + i · =
n 8 2 2 2

16
 Thus, we tabulate the data as follows:

i xi f (xi )
0 0 0.00000000000
1
1 2 0.23971276930
2 1 0.84147098481
3
3 2 1.49624247991
4 2 1.81859485365
5
5 2 1.49618036026
6 3 0.42336002418
7
7 2 -1.22774129691
8 4 -3.02720998123

s
Therefore,

or
Z 4
x sin x dx ≈ S8
0

t
h
= [f (x0 ) + f (x8 ) + 4 (f (x1 ) + f (x3 ) + f (x5 ) + f (x7 )) + 2 (f (x2 ) + f (x4 ) + f (x6 ))]
3
Tu
1h
= 2 0 + (−3.02720998123) + 4(0.23971276930
3
+ 1.49624247991 + 1.49618036026 − 1.22774129691)
i
ty

+ 2(0.84147098481 + 1.81859485365 + 0.42336002418)

1h i
= − 3.02720998123 + 4(2.00439431256) + 2(3.08342586264)
6
i

1h i
fin

= − 3.02720998123 + 8.01757725024 + 6.16685172528

6
1
= × 11.15721899429
6
In

= 1.85953649904

(d). We want to approximate the integral:

Z 4
√ x
xe dx
0

using the Simpson’s Rule with n = 6 subdivisions.

Here we have:

a = x0 = 0,
b = x6 = 4,
n = 6,
√
f (x) = x ex .

17
 Now,
b−a 4 2 2
h= = = , xi = a + ih = i ·
n 6 3 3
Thus, we tabulate the data as follows:

i xi f (xi )
0 0 0.00000000000
2
1 3 1.59031818508
4
2 3 4.38055036042
3 2 10.44970334824
8
4 3 23.50190056938
10
5 3 51.17851092752

s
6 4 109.19630006629

or
Therefore,
Z 4
√ x

t
xe dx ≈ S6
0
Tu
h
= [f (x0 ) + f (x6 ) + 4 (f (x1 ) + f (x3 ) + f (x5 )) + 2 (f (x2 ) + f (x4 ))]
3
2h
= 3 0 + 109.19630006629 + 4(1.59031818508 + 10.44970334824 + 51.17851092752)
3 i
ty

+ 2(4.38055036042 + 23.50190056938)
2
= [109.19630006629 + 4(63.21853246084) + 2(27.88245092980)]
9
i

2
fin

= [109.19630006629 + 252.87412984336 + 55.76490185960]

9
2
= × 417.83533176925
9
In

= 92.85229594872

(e). We want to approximate the integral:

Z 3
ln x dx
1

using the Simpson’s Rule with n = 6 subdivisions.

Here we have:

a = x0 = 1,
b = x6 = 3,
n = 6,
f (x) = ln x.

18
 Now,
b−a 3−1 2 1 i 3+i
h= = = = , xi = a + ih = 1 + =
n 6 6 3 3 3

Thus, we tabulate the data as follows:

i xi f (xi )
3
0 3 0.00000000000
4
1 3 0.28768207245
5
2 3 0.51082562377
3 2 0.69314718056
7
4 0.84729786039

s
3
8
5 3 0.98082925301

or
6 3 1.09861228867

Therefore,

t
Z 3
ln x dx ≈ S6
Tu
1
h
= [f (x0 ) + f (x6 ) + 4 (f (x1 ) + f (x3 ) + f (x5 )) + 2 (f (x2 ) + f (x4 ))]
3
1
3
= [0 + 1.09861228867 + 4(0.28768207245 + 0.69314718056 + 0.98082925301) + 2(0.51082562
ty

3
1
= [1.09861228867 + 4(1.96165850602) + 2(1.35812348416)]
9
i

1
= [1.09861228867 + 7.84663402408 + 2.71624696832]
fin

9
1
= × 11.66149328107
9
= 1.29572147567
In

(f). We want to approximate the integral:

Z 2
xe−x dx
0

using the Simpson’s Rule with n = 8 subdivisions.

Here we have:

a = x0 = 0,
b = x8 = 2,
n = 8,
f (x) = xe−x .

19
 Now,
b−a 2−0 1 1 i
h= = = , xi = a + ih = i · =
n 8 4 4 4
Thus, we tabulate the data as follows:

i xi f (xi )
0 0 0.00000000000
1
1 4 0.19470019577
1
2 2 0.30326532986
3
3 4 0.35427491456
4 1 0.36787944117
5
5 4 0.35813099608

s
3
6 2 0.33469524022

or
7
7 4 0.30410440104
8 2 0.27067056647

t
Therefore,
Z 2
Tu
xe−x dx ≈ S8
0
h
= [f (x0 ) + f (x8 ) + 4 (f (x1 ) + f (x3 ) + f (x5 ) + f (x7 )) + 2 (f (x2 ) + f (x4 ) + f (x6 ))]
3
ty

1
4
= [0 + 0.27067056647 + 4(0.19470019577 + 0.35427491456 + 0.35813099608 + 0.304104401
3
+2(0.30326532986 + 0.36787944117 + 0.33469524022)]
i

1
fin

= [0.27067056647 + 4(1.21121050745) + 2(1.00584001125)]

12
1
= [0.27067056647 + 4.84484202980 + 2.01168002250]
12
1
In

= × 7.12719361877
12
= 0.59393271823

(g). We want to approximate the integral:

(
Z π sin x
f (x) dx, where f (x) = x , x>0
0 1, x=0

using the Simpson’s Rule with n = 6 subdivisions.

20
Here we have:

a = x0 = 0,
b = x6 = π,
n = 6,
(
sin x
f (x) = x , x>0
.
1, x=0

Now,
b−a π iπ
h= = , xi = a + ih =
n 6 6

s
Thus, we tabulate the data as follows:

or
i xi f (xi )
0 0 1.00000000000

t
π
1 6 0.95492965855
2
Tu
π
3 0.82699334313
π
3 2 0.63661977237
2π
4 3 0.41349667157
5π
5 6 0.19098593171
6 π 0.00000000000
ty

Therefore,
Z π
i

f (x) dx ≈ S6
fin

0
h
= [f (x0 ) + f (x6 ) + 4 (f (x1 ) + f (x3 ) + f (x5 )) + 2 (f (x2 ) + f (x4 ))]
3
π
In

6
= [1.00000000000 + 0.00000000000 + 4 (0.95492965855 + 0.63661977237 + 0.19098593171)
3
+ 2 (0.82699334313 + 0.41349667157)]
π
= [1 + 4(1.78253536263) + 2(1.24049001470)]
18
π
= [1 + 7.13014145052 + 2.48098002940]
18
π
= × 10.61112147992
18
= 1.85199007154

21