SEQUENCING MODEL

Graphical method is used to determine the minimum total elapsed time.

Graphical method

Algorithm:
Step 1: Use the x-axis to represent the processing time for job 1 and the y-axis to
represent the process time for job 2.
Step 2: Mark the machine time for the two jobs and their corresponding axes as per
the given technological order.
Note that in such a graph moving horizontally will imply that job 1 is
being processed while job 2 remains idle. Moving vertically will imply
that job 2 is processed while job 1 remains idle. Diagonal movement along
45° line to the horizontal (line with slope 1) will imply that both the jobs 1
and 2 are being processed simultaneously. Since each machine can process
only one job at a time, overlapping region for the machines should be
determined first and movement across them should be avoided.
Step 3: An optimal path is the shortest one (that minimises the idle time for job 1 in
the horizontal movement or the shortest one that minimises the idle time for job 2
in the vertical movement) consisting of horizontal, vertical and 45° lines from the
origin to the end. Obviously, we must choose such a combination in which
diagonal movement is maximum possible.
Step 4: Compute the total elapsed time by adding processing time of job 1 + idle
time of job 1.
 

Sequencing Model

The selection of the appropriate order in which waiting customers may be served is
called sequencing. A practical situation may correspond to an industry producing a
number of products, each of which is to be processed through different machines,
of course, finite in number.
 Suppose there are n jobs to perform, each of which requires processing on some
or all of m different machines. The effectiveness (that is, cost, time, mileage, and
henceforth) can be measured for any given sequence of job at each machine, and
the most suitable sequence is to be selected (which optimises the effectiveness
measure) among all (n!)m theoretically possible sequences.

Although theoretically, it is always possible to select the best sequence by testing

each one, in practice, it is impossible because of the large number of computations
involved. For example, if there are 4 jobs to be processed at each of the 5 machines
(i.e., n = 4 and m = 5), the total number of theoretically possible different
sequences will be (4!)5 = 7,962,624. So, easier methods of dealing with such
problems are needed. That is, a technique which helps us arrive at an optimal
sequence without trying all or most of the possibilities is needed.

Definition

Suppose there are n jobs (1, 2, 3, …, n) each of which has to be processed one at a
time at each of m machines A, B, C, … The order of processing each job through
machines is given (for example, job 1 is processed through machines, in the
order A, C, B). The time that each job must require on each machine is known. The
problem is to find a sequence (n!)m number of all possible sequences (or
combinations or order) for processing the jobs so that the total elapsed time for all
the jobs will be minimum.

Mathematically, let

Ai = time for job i on machine A,

Bi = time for job i on machine B, and so on
Ti = time from start of first job to completion of the last job.
 Then, the problem is to determine for each machine a sequence of jobs (i1, i2,
…, in) where (i1 …, in) is the permutation of integers which will minimise T.

Terminology and Notations

Notation:

tij = processing time (time required) for job i on machine j

T = total elapsed time for processing all the jobs. This includes idle time, if
any.
Iij = Idle time on machine j from the end of job (i – 1) to the start of job i.

Terminology:

Number of machines: It means the service facilities through which a job must
pass before it is completed.

For example, a book to be published has to be processed through composing,

printing, binding, and so on. In this example, the book constitutes the job and the
different processes constitute the number of machines.

Processing order: It refers to the order (sequence) in which given machines are
required for completing the job.

Processing time: It is the time required by a job on each machine.

Idle time on a machine: It is the time for which a machine does not have a
process; it is idle time from the end of job (i – 1) to the start of job i.
 Total elapsed time: It is the time interval between starting the first job and
completing the last job including the idle time (if any) in a particular order by the
given set of machines.

No passing rule: It refers to the rule of maintaining the order in which jobs are
to be processed on given machines. For example, if n jobs are to be processed on
two machines M1 and M2 in the order M1 M2, then each job should go to
machine M1 first and then to M2.

Assumptions

1. No machine can process more than one operation at a time

2. Each operation, once started, must be performed till completion.
3. A job in an entity is even though it represents a lot of individual parts; no
lot may be processed by more than one machine at a time.
4. Each operation must be completed before any other operation, which
following it, can begin.
5. Time intervals for processing are independent of the order in which the
operations are performed.
6. There is only one of each type of machine.
7. A job is processed as soon as possible subject to ordering requirements.
8. All jobs are known and are ready to start processing before the period
under consideration begins.
9. The time required to transfer jobs between machines is negligible.

Solution of Sequencing Problems

At present, a solution of following cases are available:

 1. n jobs and two machines A and B, all jobs processed in the order AB.
2. n jobs and three machines A, B and C, all jobs processed in the order ABC.
3. Two jobs and m machines. Each job is to be processed through the
machines in a prescribed order (which is not necessarily the same for both
the jobs).
4. Problem with n jobs and m machines.

Processing n Jobs Through Two Machines

Let there be n jobs, each of which is to be processed through two

machines A and B in the order AB. That is, each job has to pass through the same
sequence of operations. In other words, a job is assigned on machine A first and
after it has been completely processed, it is assigned to machine B. If machine B is
not free, then the job has to wait in a waiting line for its turn, that is, passing is not
allowed. Therefore, machine A will remain busy in processing all the n jobs one-
by-one while machine B may remain idle after completion of one job and before
starting of another job. Thus, the objective is to minimise the idle time of the
second machine. This can be achieved only by determining sequence of n jobs
which are to be processed on the two machines A and B.

Johnson’s algorithm for n jobs through 2 machines

Step 1: List the jobs along with their processing times in a table as follows:
Step 2: Examine the columns for processing time on machines A and B, and find
the smallest processing time in each column, that is, find out min (t1j, t2j) for all j
Step 3: (a) If the smallest processing time is for the first machine A, then place the
corresponding job in the first available position in the sequence. If it is for the
second machine, then place the corresponding job in the last available position in
the sequence.

(b) If there is a tie in selecting the minimum of all the processing times, then
there may be three situations:

1. Minimum among all processing times is same for the machine, that is, min
(t1j, t2j) = t1k = t2r, then process the kth job first and the rth job last.
2. If the tie for the minimum occurs among processing times t1j on
machine A only, then select the job corresponding to the smallest job
subscript first.
3. If the tie for the minimum occurs among processing times t2j on
machine B, then select the job corresponding to the largest job subscript
last.

Step 4: Remove the assigned jobs from the table. If the table is empty, stop and go
to step 4. Otherwise, go to step 2.
Step 5: Calculate idle time for machine A and B.

1. Idle time for machine A = (Total elapsed time) – (Time when the last job
in a sequence finishes in machine A)
2. Idle time for machine B = Time at which the first job in a sequence

finishes on machine A +  (time when the jth job in a sequence starts on

 machine B) – (Time when the (j – 1)th j on in a sequence finishes on
machine B)

Step 6: The total elapsed time to process all jobs through two machines is given by
Total elapsed time = Time when nth job in a sequence finishes on machine B.

where, B2j = Time required for processing jth job on machine B.

I2j = Time for which B remains idle after processing (j – 1)th job and before
starting work on the jth job.

Processing n Jobs Through 3 Machines

The problem can be described as

1. Only three machines A, B and C are involved.

2. Each job is processed in the prescribed order ABC.
3. Transfer of jobs is not permitted, that is, adhere strictly the order over the
machine.
4. Exact or expected processing time are given below

There is no general procedure available for obtaining optimal sequence in this case.
Johnson’s method can be extended to cover special cases when either one or both
of the following conditions hold:
 1. The minimum time on machine A ≥ the maximum time on machine B.
2. The minimum time on machine C ≥ the maximum time on machine B.

The method is to replace the problem with an equivalent problem, involving n jobs

and 2 imaginary machines G and H, and corresponding time Gi and Hi are defined
by Gi = Ai + Bi, Hi = Bi + Ci.

If this problem with the prescribed ordering GH is solved, the resulting optimal
sequence will also be optimal for the original problem.

Processing n Jobs Through m Machines

Suppose there are n jobs to be processed through m machines, say M1, M2,

…, Mm in the order M1, M2 … Mm, and let Tij denote the time taken by
the ith machine to complete the jth job. There is no general method available by
which we can obtain optimal sequence(s) in problems of this type. This problem
can be converted to a problem of processing n jobs through 2 machines if no
passing of jobs is permissible and if either or both of the conditions given in step 2
of the procedure that follows is satisfied:

Procedure for obtaining optimal sequence

Step 1: Find (i)  , (ii)  , and (iii)   for j = 1, 2,

… n
Step 2: Check whether

      (i)   for i = 2, 3, …, m – 1

or, (ii)   for i = 2, 3, …, m – 1

Step 3: If inequations of Step 2 are not satisfied, this method fails. Otherwise, go to
next step.
Step 4: Convert the m machine problem into a 2-machine problem considering two
imaginary machines G and H, so that
 
TGj = T1j + T2j + … + T(m – 1)j and
THj = T2j + T3j + … + TMj

Now, determine the optimal sequence on n jobs through 2 machines by using

the optimal sequence algorithm.

Step 5: In addition to conditions given in Step 4, if T2j + T3j + … + T(m – 1) = C (a fixed
positive constant) for all j = 1, 2, …, n, then determine the optimal sequence
for n jobs and two machines M1 and Mm in the order M1 Mm by using the Jhonson’s
algorithm.

Processing Two Jobs on n Machines

Suppose there are two jobs, job 1 and job 2, each of which is to be processed
on n machines M1, M2, … Mn.

Graphical method is used to determine the minimum total elapsed time.

Graphical method

Algorithm:
Step 1: Use the x-axis to represent the processing time for job 1 and the y-axis to
represent the process time for job 2.
Step 2: Mark the machine time for the two jobs and their corresponding axes as per
the given technological order.
 Note that in such a graph moving horizontally will imply that job 1 is being
processed while job 2 remains idle. Moving vertically will imply that job 2 is
processed while job 1 remains idle. Diagonal movement along 45° line to the
horizontal (line with slope 1) will imply that both the jobs 1 and 2 are being
processed simultaneously. Since each machine can process only one job at a
time, overlapping region for the machines should be determined first and
movement across them should be avoided.

Step 3: An optimal path is the shortest one (that minimises the idle time for job 1 in
the horizontal movement or the shortest one that minimises the idle time for job 2
in the vertical movement) consisting of horizontal, vertical and 45° lines from the
origin to the end. Obviously, we must choose such a combination in which
diagonal movement is maximum possible.
Step 4: Compute the total elapsed time by adding processing time of job 1 + idle
time of job 1.
 

General assumptions

1. The technological ordering of each of the two jobs through n machines is

known in advance (ordering may not be same for both the jobs).
2. Each machine can process only one job at a time.
3. The exact processing time on all the n machines are known.