Assignment 1

(Decision Support System & Expert System)

Submitted To:

Prof. Dr. Vinay Singh

By
Ishan Kakkar (2019MBA-13)
Assignment 1
Q1. Discuss various modelling and analysis techniques used in Decision support system.
1. Model
A model is a representation of an actual system. Models embody system characteristics that
are important to the model’s users. At the same time, models simplify reality by eliminating
other characteristics that are not important for their purpose. The central idea of a model is that
important relationships that apply to the system being modelled also apply to the model.
2. Models & Modelling Techniques
There are 7 types of DSS models with varying types of modelling techniques. These are:
Category Process and Objective Representative Techniques
1.Optimization of
Find the best solution from a small 1.Decision tables,
problems with
number of alternatives 2.Decision trees
few alternatives
1.Linear and other
Find the best solution from a large or an
2.Optimization mathematical program-
infinite number of alternatives using a
via algorithm ming models,
step by-step improvement process
2.Network models
3.Optimization
Find the best solution in one step using
via an analytic 1.Some inventory models
a formula
formula
Finding a good enough solution or the
4.Simulation best among the alternatives checked 1.Several types of simulation
using experimentation
Find a good enough solution
5.Heuristics 1.Heuristic programming,
using rules
1.Expert Systems Fore-
6.Predictive
Predict the future for a given scenario casting models,
models
2.Markov analysis
1.Financial modelling,
7.Other models Solve a what-if case using a formula
2.Waiting lines

DSS models are broadly classified into Static Models & Dynamic Models on which Static
Analysis & Dynamic Analysis is applied respectively.
3. Static Analysis
1. A Static model takes a single snapshot of situation.
2. In this static snapshot everything occurs in a single interval.
3. Static models do not explicitly take time into account.
4. An example is of a make or buy decision. Also, an annual accounting statement or
financial statement is also an example of static model. Though the financial statements
& activities adhere to an accounting period, the time frame is fixed hence it is static.
5. Stability of the relevant data is assumed.

April 24, 2020 2

Assignment 1
4. Dynamic Analysis:
1. Dynamic models represent scenarios that change over time.
2. A simple example is a 5-year profit-and-loss projection in which the input data, such as
costs, prices, and quantities, change from year to year.
3. Dynamic models are time dependent. This is important because they use, represent, or
generate trends and patterns over time. Dynamic models show average per period,
moving averages and comparative analysis.

5. Situational Analysis is done when decision-making process involves evaluating &

comparing alternatives, during which it is necessary to predict the future outcome of each
proposed alternative. Decision situations are classified on the basis of knowledge of the
decision maker about the forecasted results into 3 categories:

6. Certainty Analysis:
1. Complete knowledge is available.
2. Decision maker knows the outcome of each course of action.
3. Situation involved is often with structured problems with short time horizons.
4. Certain models are relatively easy to develop and solve and they can yield optimal
solutions.

7. Uncertainty Analysis:
1. There are several outcomes for each course of action.
2. Decision maker does not know, or cannot estimate the possible outcomes.
3. This type of analysis is more difficult because of insufficient information. It involves
assessment of the decision maker’s attitude towards risk.

8. Risk Analysis:
1. Decision maker must consider several possible outcomes for each alternative.
2. The decision maker can assess the degree of risk associated with each alternative.
3. Risk analysis can be performed by calculating the expected value for each alternative
and selecting the one with best expected value.

9. Decision analysis is used with decision tables and decision trees. Single-goal situations can
be modelled with decision tables or decision trees.
1. Decision tables are used to organize information and knowledge in systematic tabular
manner.
2. Decision Trees are an alternative representation of the decision table. They show the
relationship of the problem graphically and handle complex situation. Decision Trees
can be cumbersome if there are many alternatives or static nature.
3. Components of decision support mathematical models:
• Result Variables: These reflect the level of effectiveness of a system.
• Decision Variables: These describe alternative course of action.
• Uncontrollable Variables: These represent some factors that affect the result
variables but not under the control of decision maker.
• Intermediate result Variables: These reflect the intermediate outcomes in
mathematical models.

April 24, 2020 3

Assignment 1
10. Sensitivity Analysis attempts to assess the impact of a change in input data on proposed
solution. It is important because it allows flexibility and adaptation to changing conditions.
It also provides a better understanding of the model and the decision-making situation. It is
used for:
1. Revising models to eliminate too-large sensitivities.
2. Adding details about sensitive variables.
3. Obtaining better estimate of sensitive external variables.
4. Altering a real-world system to reduce actual sensitivities.

11. What-If-Analysis addresses the context of what will happen to the solution if an input
variable, an assumption, or a parameter value is changed. With the appropriate user
interface, it is easy for manager to ask a computer model different questions and get the
answers. This analysis is common in expert systems. Through this the user gets an
opportunity to change their answers to some questions.

12. Goal Analysis is used to calculates the values of the inputs necessary to achieve a desired
level of output. It represents a backward solution approach since the objective is defined.

13. Problem solving search methods: The choice phase of problem solving involves a search
for an appropriate course of action. Search approaches are:
1. Analytical Techniques
2. Algorithms
3. Blind Searching
4. Heuristic Searching

14. Simulation Modelling: Simulation is an appearance of reality. Simulation modelling is a

technique for conducting experiments with a computer on a model of a management
system. Its characteristics are:
1. Simulation typically imitative.
2. It is a technique for conducting experiments.
3. Simulation is descriptive rather than a normative.
4. It is used only when a problem is too complex to be treated using numerical optimizing
techniques.

15. Probabilistic Simulation: It is used when there are one or more of the independent
variables. It follows certain probability distributions namely:
1. Discrete distribution
2. Continuous distribution
3. It is conducted with the aid of technique called Monte Carlo simulation.

16. Time-Dependent Vs Time-Independent Simulation: •

1. In time-independent-simulation it is not important to know the exact time of event
2. In time-dependent simulation, it is important to know the precise time of arrival.

17. Visual Simulation: It is used in graphical display of computerized results & includes
animations. It is one of the most successful development in computer-human interactions
and problem solving.

April 24, 2020 4

Assignment 1
Q2. How Business Intelligence (Data Warehousing, Data Acquisition, Data Mining,
Business Analytics, and/or Visualization) helps in developing effective support system?
1. Data Acquisition
All decision-support systems use data, information, and/or knowledge. These three terms are
sometimes used interchangeably and may have several definitions. A common way of looking
at them is as follows:
1. Data: Items about things, events, activities, and transactions are recorded, classified,
and stored but are not organized to convey any specific meaning. Data items can be
numeric, alphanumeric, figures, sounds, or images.
2. Information: Data that have been organized in a manner that gives them meaning for
the recipient. They confirm something the recipient knows, or may have
3. Knowledge: Knowledge consists of data items and/or information organized and
processed to convey understanding, experience, accumulated learning, and expertise
that are applicable to a current problem or activity. Knowledge can be the application
of data and information in making a decision.
Data can be collected/acquired manually or by instruments and sensors. Representative data
collection/acquisition methods are time studies, surveys (using questionnaires), observations
(e.g. using video cameras), and soliciting information from experts (e.g., using interviews). In
addition, sensors and scanners are increasingly being used in data acquisition.
2. Data Warehouse
A data warehouse supports business analysis and decision-making by creating an enterprise-
wide integrated database of summarized, historical information. It integrates data from
multiple, incompatible sources. By transforming data into meaningful information, a data
warehouse allows the managers to perform more substantive, accurate and consistent analysis.
Data warehouses can be described as subject-oriented, integrated, time-variant,
nonnormalized, non-volatile collections of data that support analytical decision-making. In a
data warehouse data are organized based on how the users refer to them. Data in data warehouse
is characterized as:
1. Integrated: All inconsistencies regarding naming convention and value representations
are removed.
2. Non-volatile: Data are stored in read-only format and do not change over time.
3. Time Variant: Data are not current but normally time series.
4. Summarized: Operational data are mapped into a decision-usable format
5. Large Volume: Time series data sets are normally quite large.
6. Not Normalized: DW data can be, and often are, redundant.
7. Metadata: Data about data are stored.
8. Data Sources: Data come from internal and external unintegrated operational systems.

3. Data mining
Data Mining (DM) is a term used to describe knowledge discovery in databases. Data mining
is a process that uses statistical, mathematical, artificial intelligence and machine-learning
techniques to extract and identify useful information and subsequent knowledge from large
databases.
Data mining is a process that aims to use existing data to invent new facts and to uncover new
relationships. The objectives & characteristics of data mining are:

April 24, 2020 5

Assignment 1
1. Data are often buried deep within very large databases, which sometimes contain data
from several years. In many cases, the data are cleaned and consolidated in a data
warehouse.
2. The data mining environment is usually a client/server architecture or a Web-based
architecture.
3. Sophisticated new tools, including advanced visualization tools, help to remove the
information ore buried in corporate files or archival public records. Finding it involves
massaging and synchronizing these data to get the right results. Cutting-edge data
miners are also exploring the usefulness of soft data (unstructured text stored in such
places as Lotus Notes data-bases, text files on the Internet, or a corporate-wide intranet).
4. The miner is often an end-user, empowered by data drills and other power query tools
to ask ad hoc questions and obtain answers quickly with little or no programming skill.
5. Striking it rich often involves finding an unexpected result and requires end-users to
think creatively.
6. Data mining tools arc readily combined with spreadsheets and other software
development tools. Thus, the mined data can be analysed and processed quickly and
easily.
7. Because of the large amounts of data and massive search efforts, it is some-times
necessary to use parallel processing for data mining.
Data Mining steps are as follows:
1. Define the business problem.
2. Build (find or acquire) the data-mining database.
3. Explore the data.
4. Prepare the data for modelling.
5. Build (or find) the models.
6. Evaluate the models.
7. Act on the results.

4. Data Warehousing, Data Mining in DSS

In a DSS the goal of a data warehouse is to support decision making with data. Data mining is
used in conjunction with a data warehouse to help with various types of decisions. A DSS with
data warehousing and data mining needs a skilled user who will supply the correct data and a
specialist who can make objective conclusions out of the output that is created. Data
warehousing & Data Mining provide answers of many queries to the organization and the user
and helps in decision making. These queries are:
1. Tactical query: A tactical query is a database operation that attempts to determine the
best course of action right now. A tactical query provides information to rank and file
elements in the field that need to respond quickly to a set of unfolding events. Tactical
queries tend to produce a very small result set. It is not uncommon for the result set to
be less than a dozen rows. Usually the result set is designed to fit into a single window
on a display screen.
2. Strategic query: The strategic query provides the information necessary to make long
term business decision. A strategic query is a database operation that attempts to
determine what has happened, why it happened, and/or what will happen next. It
typically accesses vast amounts of detailed data from the warehouse and ranges in
complexity from simple table scans to multi-way joins and sub queries. Applications
that generate strategic queries include; report generation, OLAP, decision support, ad-
hoc, data mining, etc.

April 24, 2020 6

Assignment 1
3. Update query: An update query is a database operation that modifies the state of a
database. Teradata provides a set of bulk load utilities used to load large quantities of
data into the database in an efficient fashion.

5. Business Intelligence/Analytics (With Visualization)

Business Intelligence is a set of methodologies, processes, architectures, and technologies that
transform raw data into meaningful and useful information used to enable more effective
strategic, tactical, and operational insights and decision-making. Business intelligence involves
acquiring data and information from a wide variety of sources and utilizing them in decision-
making. Business analytics adds an additional dimension to business intelligence through
models and solution methods.
Business intelligence methods & tools are highly visual in nature. They provide charts and
graphs of multidimensional data with the click of a mouse. These methods generally access
data from data warehouses and deposit them into a local, multidimensional database system.
Online analytical processing (OLAP) methods allow an analyst, or even a manager to slice and
dice the data, while observing graphs and tables that reflect the dimensions being observed.
Models may be applied to the data for forecasting or to identify opportunities. Data mining
methods apply statistical & deterministic models, and artificial intelligence methods to data, to
identify hidden relationships discover knowledge among the various data or text elements.
Business intelligence systems are the foundation of dashboards, which have evolved from
executive information systems into enterprise information systems that access data warehouses
via OLAP systems.
Dashboards provides managers with exactly the information they need in the correct format
at the correct time. Dashboards can impact on communications and company politics.
Dashboards and scorecards measure and display what is important. Each user, can focus on
what is important to him or her.

It is recommended that before implementing a Business Intelligence based DSS, a business

intelligence assessment be done. This assessment includes:
1. Business needs analysis: Analysis of the underlying strategic and tactical business
goals and objectives that are driving the development of the BI solution, including
whether executive sponsorship and funding are available.

2. Organizational analysis: Analysis of the existing business and technical

organizational structures, including the level of IT/business partnering in place, the
organization's culture and leadership style, its understanding of BI concepts, whether
roles and responsibilities have been established, and whether people with the
appropriate amount of time and skills are in place.

3. Technical/methodology: analysis: Analysis of whether the appropriate technical

infrastructure and development methodologies are in place, including all related
hardware and software, the quality and quantity of the source data, and the methodology
and change-control process.

April 24, 2020 7

Assignment 1
Q3. Write comprehensive understanding on developing: (a) Support System (b)Expert
System.
1. Decision Support System Development

1.1.Traditional Approach:
Also known as the Waterfall model for development the traditional approach of DSS
development contains 4 steps and is as follows.:
a. Planning: The needs or problems & opportunities of a business are identified. All
possible requirements of the system to be
developed are captured in this phase and
documented. If required a feasibility study is
conducted pertaining to cost, financial,
technical & organizational feasibility. Once
the project is approved, a project manager is
assigned, who creates a work plan, staffs the
project, and adopts methods for managing it.
b. Analysis: In this phase important inputs are
addressed, such as intended users, objective
of the system & other intricacies. An
analysis is developed & if available an
existing system us analysed & a path
towards migration to new system is defined which leads more information gathering,
development of a process model & a data model.
c. Design: The requirement specifications from first phase are studied in this phase and
the system design is prepared. This system design helps in specifying hardware and
system requirements and helps in defining the overall system architecture.
d. Implementation: The implementation phase brings it all together. This is where the
system is built or purchased. Construction involves not only building the system. but
also testing it to verify that it works. Better planning can lead to systems with fewer
bugs. Installation is the last step and involves actually getting the system up and
running.
Alternative methodologies exist for developing a DSS like RAD (Rapid Application
Development) where a system is developed quickly and users can obtain some functionality as
soon as possible. These include methods of phased development, prototyping, and
throwaway proto typing. The phased development methodology involves breaking a system
up into a series of versions that are developed sequentially. Each version has more functionality
than the previous one, and they evolve into a final system. The advantage is that users gain
functionality quickly; the disadvantage is that the systems with which users start to work are
incomplete by design.
1.2.Prototyping
a. Prototyping involves performing the analysis, design, and implementation phases
concurrently and repeatedly. System prototypes are quickly developed and
demonstrated to users; whose input is used to refine them.
b. The main advantage is that systems are quickly provided to users even if not yet ready
for institutional use. Feedback is obtained, and the system can be modified on moving
to the next prototype. Further analysis may be needed as well.
c. One disadvantage is that changes are introduced quickly and there is no attempt to
correct design decisions early on. Instead they are repaired as the system evolves.

April 24, 2020 8

Assignment 1
DSS development is done through proto typing
for the following reasons:
1. Users and managers are involved in every
phase and iteration. The iterative nature
allows users to be involved in system
design, which is important for DSS.
2. Learning is explicitly integrated into the
design process. Since the users are
involved in system design, both users and
system developers learn about the
decision-making, the ill-structured,
complex problem, and the technologies
that can potentially be applied to it.
3. Requirements evolve as experience is gained. This strategy assumes that the requirements
are only partially known at the beginning of system development, and it attempts to clarify
users' needs by actively involving them in a low-cost, fast-feedback development process.
4. A key criterion associated with prototyping is the short interval between iterations. The
initial prototype must be low-cost. It must fall below the minimum threshold of capital
outlays requiring formal justification. The development of a prototype may be a risky
decision. However, because the benefits of a DSS are often intangible, relating to such
issues as improved decision-making or better understanding, a high initial investment may
result in a decision not to proceed
5. The advantages of prototyping are:
• Short development time
• Short user reaction time (feedback from user)
• Improved user understanding of the system, its information needs, & its capabilities
• Low cost
6. The disadvantage of prototyping is that, the gains obtained from cautiously stepping
through each of the system's life-cycle stages might be lost. These gains include a thorough
understanding of the I.S.’s benefits and costs, a detailed description of the business's
information needs, an easy-to-maintain IS design, a well-tested IS and well-prepared users.
2. Expert System Development
2.1.Knowledge Acquisition
The data is collection of facts. The information is organized as data and facts about the task
domain. Data, information, and past experience combined together are termed as
knowledge. A Knowledge-Base contains domain-specific and high-quality knowledge.
The knowledge base of an ES is a store of both, factual and heuristic knowledge.
• Factual Knowledge − It is the information widely accepted by the Knowledge
Engineers and scholars in the task domain.
• Heuristic Knowledge − It is about practice, accurate judgement, one’s ability of
evaluation, and guessing.
Knowledge is required to exhibit intelligence. The success of any expert system majorly
depends on the quality, completeness, and accuracy of the information stored in the
knowledge base. The knowledge base is formed by readings from various experts, scholars,
and the Knowledge Engineers. A knowledge engineer is a person who acquires
information from subject expert by recording, interviewing, and observing him at work,
etc. He then categorizes and organizes the information in a meaningful way, in the form of
IF-THEN-ELSE rules, to be used by interference machine. The knowledge engineer also
monitors the development of the Expert System.

April 24, 2020 9

Assignment 1
The development of an expert system is similar to that of prototyping & involve 5 core
stages.

a. Identification: The knowledge engineer and expert determine the important features of the
problem. This includes identifying the problem itself (e.g. type and scope), the participants
in the development process (e.g. additional experts), the required resources (e.g. time and
computing facilities), and the goals or objectives of building the expert system (ex. improve
performance or distribute scarce expertise).
b. Conceptualization: The knowledge engineer and expert decide what concepts, relations
and control mechanisms are needed to describe problem solving in the domain. Subtasks,
strategies and constraints related to the problem-solving activity are also explored. At this
time the issue of granularity is addressed. This means considering the level of detail of
represented knowledge. The knowledge engineer normally picks the most abstract level of
detail (coarsest grain) that still provides adequate discrimination between key concepts.
The developers must avoid trying to produce a complete problem analysis before beginning
program implementation, since much learning comes from the 1st implementation that
shapes and directs the conceptualization process.
c. Formalization: Formalization involves expressing the key concepts and relations in some
formal way, usually within a framework suggested by an expert system building language.
Thus, the knowledge engineer should have some ideas about appropriate tools for the
problem by the time formalization begins.
d. Implementation: The knowledge engineer turns the formalized knowledge into a working
computer program. Constructing a program requires content, form and integration. The
content comes from the domain knowledge made explicit during formalization i.e. the data
structures, inference rules and control strategies necessary for problem solving. The form
is specified by the language chosen for system development. Integration involves
combining and reorganizing various pieces of knowledge to eliminate global mismatches
between data structures and rule or control specifications. Implementation must be rapidly
as the rationale for implementing the initial prototype is to check the effectiveness of the
design decisions made during the earlier phases of development. This means that there is a
high probability that the initial code will be revised or discarded during development.
e. Testing: Testing involves evaluating the performance and utility of the prototype program
and revising it as necessary. The domain expert typically evaluates the prototype and helps
the knowledge engineer to revise it. As soon as the prototype runs on a few examples, it
should be tested on. many problems to evaluate its performance and utility. This evaluation
may uncover problems with the representational scheme, such as missing concepts and
relations, knowledge represented at the wrong level of detail, or unwieldy control
mechanisms. Such problems may force the developers to recycle through the various
development phases, reformulating the concepts, refining the inference rules and revising
the control flow.

April 24, 2020 10