Loading

Module 1: Managerial Support Systems

Notes
Study Reminders
Support
Text Version

Components of Decision Support Systems - Database

Set your study reminders

We will email you at these times to remind you to study.
  • Monday

    -

    7am

    +

    Tuesday

    -

    7am

    +

    Wednesday

    -

    7am

    +

    Thursday

    -

    7am

    +

    Friday

    -

    7am

    +

    Saturday

    -

    7am

    +

    Sunday

    -

    7am

    +

Every decision support system has to deal with some amount of data which is stored in
the systems. Is data is related to the different kinds of input that are required by the
models in solving the problems. However, in some decision support systems, the data
retrieval and data update operations are most predominant than in others.
Database management systems, they take care of creation, retrieval, storage,
modification, security as well as the storage structure of the computerised data bases
along with authorization checks and the validation procedures. Data Base Management
Systems provides data independence, what do we mean by that? The ability of an application program and adhoc requests to be independent of the
storage and the access methods for the data. We do not need to know whether the files
are random files, index files, or sequential files, there is no need for knowing the access
methods for the data. The DBMS takes care of it.
(Refer Slide Time: 03:28)

Now, we come to the design cycle of a decision support system. Decision support
systems by its very nature is an interactive and then evolutionary process. Essentially
there are three phases in the design cycle of a decision support system. One is the pre
design phase, the next is the design phase, and the third one is the post design phase.
First, let us concentrate on the pre design phase.

(Refer Slide Time: 04:18)

In this phase, the DSS designer prepares a description of the existing procedures related
to a decision area. The decision problem for which a DSS is required is then identified.
Then, the DSS designer is bothered about developing the required interface.
DSS designer also must be able to foreseen, that whether there are implementation
difficulties related to the proposed design. After all these things are over a cost benefit
analysis needs to be prepared, in order to find out that whether the organization or the
manager should actually go ahead with the development of this decision support system.
(Refer Slide Time: 06:36)

Because, the expected benefits must be more, much more than the cost, that will be
incurred; then only it is wise to going for the development of the decision support
system. A decision support system is likely to be successful, if the problem area has
some of these following characteristics. What are those characteristics? They are should
be a need for managing large amount of data, this is number 1.
Number 2, there must be a need for complex computational and logical tasks to be
performed on that data. The third one is that, there should be a need for quick
computation and data manipulation. And, the fourth one is a need for judgment in
interpreting the results, when managers have the major role and suggest alternative
course for processing the data.
So, once again the following characteristics dictate the need for a DSS. What are those?
There should be a requirement for managing large amount of data, there should be a need
for complex computational and logical tasks to be performed on those data, there should
be need for quick computation and data manipulation. And, the fourth one is that there
should be a requirement for judgment in interpreting the model output that is the results,
and suggest alternative course for processing the data, where the managers have the most
important role to play.
(Refer Slide Time: 09:02)

In the design phase, the problem has been identified. In the pre-design phase, is further
refined for design purposes. Here in there are three major issues that need to be analyzed.

The first one is what is expected of this decision support system right. The second one is
how to know when do we stop this development?
And, the third issue is to determine the priorities which need to be observed in the
development of this DSS. What is expected of DSS? How to know? When to stop the
development? And, the next one is knowing the priorities that must be maintained in the
development of DSS.
(Refer Slide Time: 10:25)

These three issues are very-very important.
(Refer Slide Time: 10:40)

From the viewpoint of the user decision support systems is a conversational system,
which offers him several imperatives. Such as display, find, tabulate, grouping, predict.
Having known that these are the several imperatives, the designer decides which
commands are to be embedded or are to be provided in the decision support system.
Knowing those imperatives is very-very important for a DSS designer.
(Refer Slide Time: 11:52)

Next the designer has to develop the software interface. And, the software that will be
required must support the following characteristics. It should be the interface should be
developed in such a way that it should be communicable, it should be robust, we are
already said that it should not produce absurd results for invalid commands, it should be
flexible and easy to control.

(Refer Slide Time: 12:51)

In the design phase, data management is another important task that poses several
problems during the design stage. So, areas that need to be taken care of are what is the
data that need to be held in the database? What is the software that will be required to
manipulate the data? And, how do we collect that data and having collected the data,
how do we maintain that data?
So, data management is the most arduous task in the operation of a decision support
system.
(Refer Slide Time: 13:59)

Next we come to the post design phase. So, pre design phase, design phase, and then we
are discussing about the post design phase. The system has now been developed and
delivered to the user. The system must have certain features that should be useful for the
immediate purposes.
Managers begin using the first version of the system, they learn more about the problem
and the decision support system, then they suggest extensions and improvements. If they
will start using it, while using it, they will also encounter some difficulties, they will also
get a clear idea about the problem that they are going to solve. They will suggest certain
changes, they will suggest certain extensions and improvements of the system, which the
designer has to incorporate.
(Refer Slide Time: 15:32)

The software which will be required in the decision support system should be highly
flexible to incorporate new routines to be embedded such that the capability of the DSS
is enhanced. So, it is an evolutionary process.
And, in the process of evolution it may happen, that after some time the requirements
grow beyond the scope of the system, which was initially envisaged. In that case it is
necessary to go back to the pre-design phase and repeat the sequence of steps which we
have just discussed to design a modified system.

(Refer Slide Time: 16:56)

Now, we will be discussing about the research findings on unsuccessful implementation
of decision support system. First point, what has been found is that the personnel who
were in charge of the design and development of the system or those persons or users,
who had a key role in the operation of the system left the company. And thereby, the
implementation of the system became unsuccessful.
Other reasons for failure in implementation of a DSS are that the management of the
organization where the DSS is going to be get implemented, did not show any major
interest in supporting or using the system. The usage of the system within the
organization was met with a number of obstructions from some employees that is another
reason why many DSS have failed in the past.
And, why the employees they create problems, they create obstructions. The sometimes
they view, the system as being a threat to their position in the organization. Because,
they think that, if the computer system is going to dictate solution for a problem, then the
organizations might chuck them out, they may no longer be needed or their importance
might be curtailed.
So, employees they really do not accept the implementation of such systems, because
they feel threatened. They feel that their role will no longer be needed. And, sometimes
they create problems in the sense that they try to convince others that the system is going
to deteriorate the performance of the organization.

So, these are some of the findings why decision support systems, many decision support
systems have failed in during or implement during implementation.
(Refer Slide Time: 20:23)

So, what are the critical success factors, for successful implementation of a DSS?
Because see we have deliberated with the various reasons why decision support systems
have failed. Now, you must know, what are the critical success factors for successful
implementation of decision support system?
Since, the first point is that before DSS is designed, developed and implement it, the
organizations or the users of those organization of this DSS must feel that this decision
support system is required. A clear need for that decision support system must be felt by
the user for an improvement in the concerned decision area.
Until and unless, the managers, or the users, who will be using the decision support
system they feel the need for it; the acceptability of the decision support systems will not
be there. Next, there must be commitment from the users of the system, for the
development of that system, and that commitment and involvement of the users must be
there right from the beginning.
That means, committed users must be involved from an early stage for DSS
implementation to be successful. Because, if they users they get involved right from the
beginning right at the early stage, then they will be able to specify the requirements of

from the system. They will be able to appreciate and understand the problem. And, the
involvement of the users should not only be there in the requirement analysis stage, but
also users should actively participate in the design process.
Because during design also, the DSS designer might require the help of the user for
constructing or formulating an effective algorithm for solving those problems. And, in
here the role of the designer should be seen as a change agent. But, the actual users must
play the primary role in not only specifying the requirements from the system, but also in
the design and implementation.
(Refer Slide Time: 24:53)

The designer should view his job as if he is delivering a service rather than a product,
because if the designer’s mentality is that he will just deliver a product and you know his
job is finished. Then those kind of decision support systems will never be acceptable,
never will they be successful. Designer should view his job as consisting of delivery of a
service remember, service rather than a product.
The designer must be concerned more with the usefulness of his service rather than
technical perfection. Because, this is very important sometimes some of the system
professionals, or some of these designers, they get engrossed with achieving technical
perfection or achieving some thrill out of developing the system. He tries to utilize his
own knowledge; he tries to get his own satisfaction, in making the system technically
perfect.

But, not bothered about whether the system is user friendly, whether the system we will
serve his intended purpose or not, they are not bothered about the usefulness of this
service. If that is the attitude then DSS cannot be successful. So, designer must be
concerned more with the usefulness of the service that the decision support system is
intended to provide, rather than the technical perfection of the system.
Next the user and the modeler, they should have understanding of each other’s point of
view and areas of expertise. They must compliment each other’s capabilities, each
other’s knowledge, each other’s area of expertise. And, these observations hold for any
modelling job or any information system design effort and particularly most important
for the design of a decision support system, because the designer and modeller must
compliment each other’s area of expertise rather than competing.