Scribd
Upload
61 views

Literature and Survey: An Overview of Time Table Generation

This document discusses timetable generation and optimization using genetic algorithms. It provides background on timetable problems, describing them as NP-complete and discussing the use of heuristics. Genetic algorithms are proposed as they can find optimized solutions. The document outlines the components of genetic algorithms and describes how they have been applied to timetable problems, including chromosome representation and using selection, crossover and mutation operators. Constraints like not double-booking resources are also discussed.

Uploaded by

Pratik Patel
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
61 views

Literature and Survey: An Overview of Time Table Generation

This document discusses timetable generation and optimization using genetic algorithms. It provides background on timetable problems, describing them as NP-complete and discussing the use of heuristics. Genetic algorithms are proposed as they can find optimized solutions. The document outlines the components of genetic algorithms and describes how they have been applied to timetable problems, including chromosome representation and using selection, crossover and mutation operators. Constraints like not double-booking resources are also discussed.

Uploaded by

Pratik Patel
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

LITERATURE AND SURVEY

An Overview of Time Table Generation:


Time table is used for processing of managing one of the institution performance
sequences.it is used to give you an idea about events along a given period of
time. This system is easy to build schedule for any institution operation and get
variety of reports to make a quick and good decision.it is efficient way to view
typically a graphic design showing software. All schedules are easy to made and
modify.

In this approach we will be using Genetic Algorithm which comes under


Evolutionary Algorithm. Genetic Algorithm is a fine way to apply to this approach
as it gives an optimised solution to this problem.

History:
Timetabling is known to be a non-polynomial complete problem i.e. there
is no known efficient way to locate a solution. Also, the most striking
characteristic of NP-complete problems is that, no best solution to them is
known. Hence, in order to find a solution to a timetabling problem, a
heuristic approach is chosen. This heuristic approach, therein, leads to a
set of good solutions (but not necessarily the best solution). In a general
educational timetabling problem, a set of events (e.g. courses and exams)
are assigned into a certain number of timeslots (time periods) subject to a
set of constraints, which often makes the problem very difficult to solve in
real-world circumstances. In fact, large-scale timetables such as university
timetables may need many hours of work spent by qualified people or
team in order to produce high quality timetables with optimal constraint

satisfaction and optimization of timetables objectives at the same time.


These constraints are of two types Hard and Soft constraints. Hard
constraints include those constraints that cannot be violated while a
timetable is being computed. For example, for a teacher to be scheduled
for a timeslot, the teacher must be available for that time slot. A solution
is acceptable only when no hard constraint is violated. On the other hand
soft constraints are those that are desired to be addressed in the solution
as much as possible. For example,though importance is given to a

teachers scheduling,focus is on setting a valid timetable and this can lead


to ateacher going free for a time slot.Thus, while addressing the
timetabling problem, hard constraints have to be adhered, atthe same
time effort is made to satisfy as many softconstraints as possible. Due to
complexity of the problem, most of the work done concentrates on

heuristic algorithms which try to find good approximate solutions


Generally, two questions arise

(I) How fast the solution is computed?


(ii) How close the solution is to the optimal one?
Nowadays, in most of the colleges timetable is generated manually which is a
very hard approach and needs a lot of time to generate an optimised time
table.
Most of the time, the generated time table is optimal but in our case, there ae
limited number of classes and laboratories and so it is difficult to get an
optimal solution but the time taken to generate a timetable will decrease.

Scheduling Problem:
The scheduling problems are essentially the problems that deal with
effective distribution of resources. During the scheduling process many
constraints have to be considered. Resources are usually limited and no
two tasks should occupy one particular resource at the same time. For
most of the scheduling problems it has been shown that they are NP-hard,
and that they cannot be solved in polynomial time using a deterministic
algorithm. School timetable scheduling problem presents a set of tasks
(classes) and a set of resources (rooms, groups, instructors). Every task
requests some resources for its realization and has the exact length. The
set of timeslots when a class can be scheduled is also determined. The
goal is to assign those tasks to their resources while satisfying all of the
hard constraints no resource should be allocated by multiple tasks at the
same time.

Scheduling a timetable could also be represented like special class of 3D


cutting problems. The timetable could be presented as a 3D structure. The
dimensions of 3D timetables are: days (x-axis), timeslots (y-axis) and
rooms (z-axis). The classes are shown as cubes, which should be placed in
a 3D timetable structure (Fig. 1). The scheduling is a process of placing
those cubes into a timetable, in the way that no conflicting classes (which
allocate the same resource, a student group or an instructor) are placed in
the same timeslot.

Some of the things that need to keep in mind are:


a) Group g can attend only one class at one time.
b) Instructor i can teach only one class at one time.
c) In room r only one class can be taught at one time.
d) All lectures should be kept exactly once.
A GUI (Graphical User Interface) has been developed to facilitate the input
of data. For each class the following can be set:
days and times when the class could be placed;
rooms where the class could be placed;
number of rooms occupied by a class simultaneously
groups of students that attend the class;
Instructors that teach the class.

But in our case, there occur many other problems like one class can utilize more
than one classroom (e.g. a larger group occupies two PC labs at the same time).

Genetic Algorithm:
In various Genetic Algorithms, the method of selecting creatures for
breeding is handled in different ways. Hollands original model uses a

method where the healthiest are most likely to breed. Other methods
select any two creatures at random for breeding. Selective breeding can
be used in conjunction with or in the absence of an Elitist Natural Selection
Operator- in either case the GA can perform evolution. Once parents have
been chosen, breeding itself can then take place. A new creature is
produced by selecting, for each gene in the chromosome, an allele from
either the mother or the father. The process of combining the genes can
be performed in a number of ways. The simplest method of combination is

called single point crossover.


In Hollands founding work on GAs he made mention of another operator,
besides selection, breeding, crossover and mutation which takes place in
biological reproduction. This is known as the inversion operator. An
inversion is where a portion of chromosomes detaches from the rest of the
chromosome, then changes direction and recombines with the

chromosome.
Class timetable of each of odd and even-semesters is prepared manually,
by manipulating those of earlier years, with the only aim of producing a
feasible timetable. Each class timetable is composed of two phases. The
first phase contains the common compulsory classes of all under-graduate
programmes (B.Tech and integrated M.Sc), and the timetable of this phase
is prepared by a central team.

Moreover, the timetable is subject to the following six types of hard


constraints, which must be satisfied by a solution to accept it as a valid

one:
1. A student should have only one class at a time.
2. A teacher should have only one class at a time.
3. A room should be booked only for one class at a time (a set of combined
classes may be treated as a single class).
4. Only one class of a course should be scheduled on a day.
5. A class should be scheduled only in a specific room, if required, otherwise
in a general room which has sufficient sitting capacity for the students of
the class. Due to the requirement of some extra facilities, such as
laboratory apparatus, many classes may need to be scheduled only in
specific rooms.
6. A class should be scheduled only at a specific time-slot, if required. Due to
many reasons, such as involvement of senior teachers in administrative
works, some classes may need to be scheduled only at specific time-slots.

Following three types of soft constraints have been considered :

1. Students should not have any free time-slot between two classes on a
day.
2. Classes of teachers should be well spread over the week.
3. A smaller class should not be scheduled in a room which can be used
for a bigger class.

The basic component of an EA is chromosome which represents a solution in the


search space of an optimization problem. A chromosome is composed of genes,
each of which describes a parameter of a problem. A set of chromosomes forms
a population for an EA, evolution of which takes place over the repeated
application of EA operators, particularly selection, crossover and mutation. The
function of selection operator is to emphasize good solutions and eliminate weak
solutions. Crossover and mutation operators are responsible for the generation of
offspring.

Some features which were used to solve the problems were:


1. Chromosome Representation is a two-dimensional matrix, each
column of which represents a time-slot, and a row represents a
room.
2. Heuristic Approach is used for initializing the chromosomes of the
population.
3. Crowded Tournament Selection Operator (Deb [12]) is used to form
a mating pool of N solutions from the population.
4. Crossover for Valid Resource Allocation (XVRA) is used for
generating a new population of N offspring.
5. Mutation for Reshuffling Resource Allocation (MRRA) is used for
mutating the offspring of the new population, where information of
two random slots of a chromosome is interchanged.

You might also like