Sequencing

Introduction :
The selection of an appropriate order in which to service waiting customers (or jobs)
is called sequencing. Here the effectiveness measure which may be time, cost,
mileage etc. is a function of the order or sequence in which a series of tasks are
performed (or processed). There are the problem in which we have n tasks to be
processed on some or all of m different machines in which the total effectiveness
depends on the order of processing. There are (n!)m possible sequences for the
problem. This is a lengthy and long process. To overcome such situations an
effective method of the sequencing can be adopted.

A Sequencing Problem :
A general sequencing problem can be defined as:
Let there are n – jobs (1,2,3,…….,n) which have to be processed one at a time at
each of m machines A, B, C,……….. . The order of the machines for each job, in
which it should go to the machines is given. The time required by the jobs on each
of the machines is also given. Then the problem is to find the sequence, out of (n!)m
sequences, which optimizes (minimizes) the total time elapsed from the start of the
first job to the completion of last job.
Mathematically
If Ai, Bi,……….., etc. are the times for the job i on machines A, B, C…….and T is the
total times elapsed from the start of the first job to the completion of the last job to,
then the problem of sequencing is to find for each machine a sequence, (i1, i2,
…..,in), where (i1, i2, …..,in) is a permutation of the integers (1, 2,…,n), which
minimizes the T.
There are following four cases of sequencing:
1. n jobs to be processed on two machines A and B, all jobs to be processed in the
order AB.
2. n jobs to be processed on three machines A, B and C, all jobs to be processed in
the order ABC.
3. n jobs to be processed on m machines.

4. Two jobs to be processed on m machines. Each job to be processed through the

machines in a prescribed order which is not necessarily the same for both jobs.
General Assumptions:
We shall solve the sequencing problem under the following assumptions:
1. The processing times of jobs on machines are independent of their processing
order.
2. Each job, once started on a machine, must be performed upto the completion on
that machine.
3. No machine may process more than one job at a time.
4. The time taken by the jobs in going from one machine to another negligible.

5. There is only one machine of each type.

6. All jobs are known and are ready to start
Sequencing Decision Problem for n – jobs on
two machines:
Johnson’s Method

Here we consider the problem of processing n – jobs 1, 2, ……., n on two machines

A and B. under the following assumptions:
1. Each job is processed in order AB.
2. Ai = Processing time of i – th job on machine A. (i = 1, 2, ……., n).
3. Bi = Processing time of i – th job on machine B. (i = 1, 2, ……., n).
The problem is to find the sequence of jobs to be processed on two machines so
that, the total time ( T ) elapsed from the start of the first job to the completion
of the last job is minimized.

The procedure for the solution of the above problem, was developed by Johnson
and Bellman. The method is based on minimizing the idle time for second
machine The Johnson’s procedure for determining an optimal sequence is as
follows:
1. Select the smallest processing time in the lists A1, A2, ………, An (processing times
of jobs on machine A) and B1, B2, ……….., Bn (processing times of jobs on machine
B). If there is a tie then either of these smallest processing times may be chosen.

2. If the smallest processing time is Ar (in the list A1, A2, ………, An) then do the rth
job first. If it is Bs (in the list B1, B2, ……….., Bn) then do the sth job last.

3. Delete the times of the job already assigned from the two lists of Ai’s and Bi’s i.e.
if rth job is assigned previously, then delete Ar and Br both and if sth job is assigned
previously then delete As and Bs both.
4. After assigning one job we are left with (n – 1) jobs. Repeat step 1 to 3 on the
processing times of the remaining (n – 1) jobs.

5. Continue in this way until all the jobs have been ordered. In this way we get an
optimal sequence of jobs. Finally the minimum time T elapsed can be found by
forming a chart.
CL: Six jobs to be processed first over machine 1 and then over machine 2. The
order of the completion of jobs has no significance. The following table gives the
machine times in hours for 6 jobs on the two machines.

Job No. i 1 2 3 4 5 6
Time on machine 1(Ai) 5 9 4 7 8 6
Time on machine 2(Bi) 7 4 8 3 9 5

Find the sequence of the jobs that minimizes the total elapsed time to complete the
jobs. Find the minimum time by using Gantt’s chart or by other method.

Solution:

3 1 5 6 2 4
To find the minimum elapsed time

Machine 1 Machine 2
Idle time of
Job
machine 2
Time in Time out Time in Time out

3 0 4 4 12 4

1 4 9 12 19

5 9 17 19 28

6 17 23 28 33

2 23 32 33 37

4 32 39 39 42* 2

The total time elapsed is 42 hours and the idle time for machine 1 is 42 – 39 = 3 hrs
and for machine 2 is 6 hrs.
CL: Five jobs to be processed first over machine 1 and then over machine 2. The
order of the completion of jobs has no significance. The following table gives the
machine times in hours for 5 jobs on the two machines.

Job No. i 1 2 3 4 5
Time on machine 1(Ai) 5 1 9 3 10
Time on machine 2(Bi) 2 6 7 8 4

Find the sequence of the jobs that minimizes the total elapsed time to complete the
jobs. Find the minimum time by using Gantt’s chart or by other method.

Solution:

2 4 3 5 1
To find the minimum elapsed time

Machine 1 Machine 2
Idle time of
Job
machine 2
Time in Time out Time in Time out

2 0 1 1 7 1

4 1 4 7 15

3 4 13 15 22

5 13 23 23 27 1

1 23 28 28 30 1

The total time elapsed is 30 hours and the idle time for machine 1 is 30 – 28 = 2 hrs
and for machine 2 is 3 hrs.
Sequencing Decision Problem for n – jobs on
three machines:
Modified Johnson’s Method

Here we consider the problem of processing n – jobs 1, 2, ……., n on three

machines A, B and C. under the following assumptions:
1. Each job is processed in order ABC.
2. Ai = Processing time of i – th job on machine A. (i = 1, 2, ……., n).
3. Bi = Processing time of i – th job on machine B. (i = 1, 2, ……., n).
4. Ci = Processing time of i – th job on machine C. (i = 1, 2, ……., n).

The problem is to find the sequence of jobs to be processed on three machines so

that, the total time ( T ) elapsed from the start of the first job to the completion
of the last job is minimized.

The previous method is can be extended to the cases in which either or both of the
following condition hold.
1. The smallest processing time for machine A ≥ the largest processing time for
machine B.
2. The smallest processing time for machine C ≥ the largest processing time for
machine B.
This problem is replaced as the previous problem if we are able to convert three
machines in to two fictitious machines G and H. Then the processing time for
two machines Gi and Hi is given by

Gi = Ai + Bi

H i = Bi + C i

Now find the optimal sequence of jobs in the order GH on these two machines G
and H by the previous method. These resulting optimal sequences on two
machines G and H will also be optimal sequence for three machines A, B and
C.
CL: Five jobs to be processed over machines A, B and C. The order of the
completion of jobs is ABC. The following table gives the machine times in hours for
5 jobs on the three machines.

Job No. i 1 2 3 4 5
Time on machine A (Ai) 5 7 6 9 5
Time on machine B (Bi) 2 1 4 5 3
Time on machine C (Ci) 3 7 5 6 7

Find the sequence of the jobs that minimizes the total elapsed time to complete the
jobs. Find the minimum time by using Gantt’s chart or by other method.
Solution:

Here Min Ai = 5 and Max Bi = 5 Min Ci = 3

Since Min Ai ≥ Max Bi

Therefore we can consider two fictitious machines G and H and their processing
times as follows:
 Processing Times
Jobs
Gi = Ai + Bi H i = B i + Ci
1 7 5
2 8 8
3 10 9
4 14 11
5 8 10

Following are three possible sequences can be obtained:

2 5 4 3 1

5 4 3 2 1

5 2 4 3 1
To find the minimum elapsed time for the first sequence:

Machine A Machine B Machine C

Idle time Idle time
Job Time Time Time Time Time Time of Machine B of Machine C
in out in out in out
2 0 7 7 8 8 15 7 8
5 7 12 12 15 15 22 4
4 12 21 21 26 26 32 6 4
3 21 27 27 31 32 37 1
1 27 32 32 34 37 40* 1+6

Thus the minimum time elapsed is 40 hrs. The time may be verified for the rest of
two sequences. Idle time for machines A, B and C are 8, 25 and 12 hrs:
CL: Five jobs to be processed over machines A, B and C. The order of the
completion of jobs is ABC. The following table gives the machine times in hours for
5 jobs on the three machines.

Job No. i 1 2 3 4 5
Time on machine A (Ai) 4 9 8 6 5
Time on machine B (Bi) 5 6 2 3 4
Time on machine C (Ci) 8 10 6 7 11

Find the sequence of the jobs that minimizes the total elapsed time to complete the
jobs. Find the minimum time by using Gantt’s chart or by other method.
Sequencing Decision Problem for n – jobs on m -
machines:
Now we consider the problem of processing n jobs 1, 2, ………., n on m machines
M1, M2, ………., Mm under the following assumptions.

1. Each job is performed in order M1, M2, ………., Mm .

2. Mij = Processing of i – th job on j – th machine (i = 1, 2, ……….., n and j = 1, 2,
…………., m)

The problem is to find sequence of jobs to be performed on the machines in the

order M1, M2, ………., Mm so that the total time elapsed from the start of the
first job to the completion of the last job is minimized. There is no general
method available for the solution of above problem. The previously discussed
method can be extended as follows.

(i) Min Mi1 ≥ Max Mij for j = 2, 3, ……. , (m – 1)

(ii) Min Mim ≥ Max Mij for j = 2, 3, ……. , (m – 1)
This method is to replace the problem with an equivalent problem involving n jobs
on two machines. If G and H denote the two machines (fictitious) then the
processing times MiG and MiH of ith (i = 1, 2, ….., n) job on these two machines are
given by

MiG = Mi1 + Mi2 + ………. + Mi(m – 1)

MiH = Mi2 + Mi3 + ………. + Mim
Now find the optimal sequence of n – jobs in order GH on the two fictitious
machines G and H by previously discussed method. The resulting optimal sequence
on two machines G and H will also be the optimal sequence on m machines M1,
M2, ………., Mm .
CL: Solve the following sequencing problem when passing time is not allowed.
Processing time in hrs is given below

Machines
Items A B C D E
I 9 7 4 5 11
II 8 8 6 7 12
III 7 6 7 8 10
IV 10 5 5 4 8

Since, Min Mi1 = 7, Min Mi5 = 8

Max Mij = 8, for j = 2, 3, 4

The condition Min Mi5 ≥ Max Mij for j = 2, 3, 4 is satisfied. Therefore we consider
two fictitious machines G and H. The processing times on these two fictitious
machines G and H can be given by:
 4 5
Jobs M iG  M
j 1
ij M iH  M ij
j2
I 25 27
II 29 33
III 28 31
IV 24 22

Proceeding in usual manner we get the following optimal sequence:

I III II IV
Processing two jobs through m - Machines:
Now we shall consider the following problem:
1. There are 2 jobs 1 and 2 to be performed.
2. There are m machines denoted by A, B, C, ………., M.
3. The technological ordering of each of the two jobs 1 and 2 through m machines
is given in advance. This ordering need not to be same for jobs.
4. The exact or expected processing times A1, B1, C1, ……….., M1 for job 1 and
A2, B2, C2, ………, M2 for job 2 on the m machines are given.
The problem is to minimize the total elapsed time T from the start of the first job
on the first machine to the completion of the last job on the last machine.
It is obvious that for each machine there are two sequences of jobs i.e., (1, 2) or (2,
1). Thus, we have to select one sequence out of these two for each machine.
Graphical method is a simple method for solution of this method.
CL: Use graphical method to minimize the time needed to process the following
jobs on the machine shown, i.e., for each machine find the job which should be
done first. Also calculate the total time needed to complete both the jobs.

Job 1 Job 2
Sequence of Machines Time Sequence of machines Time
A 3 B 5
B 4 C 4
C 2 A 3
D 6 D 2
E 2 E 6
 20

14
D
12
11
A
9

3
B

3 7 9 15 17
A B C D E
The Travelling Salesman (0r Routing) Problem :

This problem is about the travelling salesman who want to visit a certain number of
cities. His problem is to select a route that will minimize the total distance travelled
(or cost or the time). In visiting each city once and only once and returning to his
home city (city from where he has started from beginning).

Formation as Assignment Problem:

This problem can be formulated as an assignment problem. Let cij is the distance
(or cost of time) of going from city i to city j. Then these cost can be arranged in the
form of a square matrix.

Here we have two extra restrictions;

(i) We can not go from city i to city i, in tis case we say (cii = ).
(ii) No city is visited twice since the tour is completed.
Cost effectiveness matrix for :

To

A1 A2 ……… Ai ……. An

A1  c12 ……… c1i ……. c1n

A2 c21  ……… c2i …….. c2n

From ……. ……..  …….. …….. ……..

……. …….. …….. …….. …….. ……..

Ai ci1 ci2 ……..  …….. cin

…….. …….. …….. …….. …….. ……..

……… …….. …….. …….. …….. ……..

An cn1 Cn2 …….. cni …….. 

Solution of a Travelling Salesman Problem:
The problem of travelling salesman can be solved by simple assignment problem.
Suppose the solution after solving the assignment problem is like
A1  A 5 , A 5  A1, A 2  A 3 , A 3  A 4 , A 4  A 2

Here the salesman proceed from city A1 to A5, then come back to A1 without
visiting cities A2, A3, A4. If he starts from city A2 then he proceed from A2 to A3,
A3 to A4 and then from A4 to A2 without visiting the cities A1 and A5. Thus this is
not a feasible solution for the problem, In such cases after solving the problem by
assignment problem technique, we use the method of enumeration by assigning the
next minimum element of the matrix in place of 0.
Example:
A machine operator processes four types of items on his machine each week, and
must choose a sequence from them. The set up cost per change depends on the
item presently on the machine and item to be made , according to the following
table.
To
A B C D
A  4 7 3
From B 4  6 3
C 7 6  7
D 3 3 7 

If he produces each type of item once and only once each week how should he
sequence the items on his machine in order minimize the total set up cost.
 A B C D

A  1 0
X 0

B 1  0 0
X

C 1 0  1

D 0 0
X 1 
Here we get,
A  D , D  A , B  C , C  B

But D can not be followed by A until B and C are produced. Now to reach the
feasible solution we assign the next minimum element 1 instead of 0.
X

A B C D

A  1 0 0
X

B 1  0 0

C 1 0  1

D 0 0
X 1 
A  B , B  C , C  D , D  A