Transcript DSS

Decision Support Systems
Yong Choi
School of Business
CSU, Bakersfield
Type of Decision-makings

Structured (Programmed)



routine & repetitive, predictable problems
standard solutions exist
Accounts receivable, order entry, payroll
Type of Decision-makings

Unstructured (Nonprogrammed)



non-routine, unpredictable, “fuzzy” complex problems
no cut-and-dried solutions
Negotiation, Lobbying
Type of Decision-makings

Semistructured (Programmed + Nonprogrammed)
 non-routine, predictable,
 Require a combination of standard solution procedures and
individual judgement
 Production Scheduling, design lay-out of factory
Stages of Decision Making

Stage 1: Intelligence
identify the problems/opportunities and then, collect data or
information

Stage 2: Design
analyze/develop the possible solutions for the feasibility
GO back to stage 1 if there is insufficient data.
Stages of Decision Making

Stage 3: Choice
Choose one alternative
Go back to stage 1 or 2 if there are no satisfactory solutions.

Stage 4: Implementation
Implement the selected alternative
Failure of implementation  go back to stage 1 or 2 or 3
Ex) Buying a new car
Transaction Processing Systems (TPS)




Developed in the early1960s
Serve the operational management level
Performing and recording daily routine and repetitive
transactions
Primary focus: structured decision-makings
Transaction Processing Systems


Lifeblood of an organization
Provide summarized and organized data in the
accounting and finance areas



Account receivable and payable
Sales transactions
Payroll
Management Information Systems



Developed in the 1960s
Intended to serve the operational or middle
management level
Summary and exception reports


monthly production reports
Quarterly travel expense reports
Management Information Systems


Difference between expected sales and actual sales of a
particular product
Primary focus: fairly structured decision-makings
Decision-Support Systems



Developed in the early 1970s
Serve the middle management
Provide alternative-analysis report





investment portfolios
Plant expansion
Primary focus: semistructured and unstructured decisionmakings
See text book for detail examples
Type of DSS

Model driven vs. Data driven
DSS Components

Three Major Components



Data management module
Model management module
Dialog management module
DSS Components
DSS Components

The Data Management Module

Gives user access to databases

Usually linked to external databases
DSS Components

The Model Management Module

Selects appropriate model to analyze data

Linear regression model
DSS Components
A linear regression model for predicting sales volume as a function of dollars spent on
advertising
DSS Components

The Dialog Module

Interface between user and other modules




Prompts user to select a model
Allows database access and data selection
Lets user enter/change parameters
Displays analysis results

Textual, tabular, and graphical displays
Model driven DSS
Primarily stand alone systems
isolated from major org.’s systems
Use models (LP, Simulation)
Sensitivity analysis as a main technique
What-If analysis
Goal Seek Analysis
What-if analysis

Attempt to check the impact of a change in the
assumptions (input data) on the proposed solution

What will happen to the market share if the advertising
budget increases by 5 % or 10%?
Goal-seek analysis


Attempt to find the value of the inputs necessary to
achieve a desired level of output
Use “backward” solution approach


A DSS solution yielded a profit of $2M
What will be the necessary sales volume to generate a profit
of $2.2M?
Tools for Model Driven DSS

Linear Programming



Lindo
Gindo
Spreadsheet Software



Excel
Lotus 1-2-3
Quattro Pro
Data Driven DSS
Many current and the newest DSS
Extract and analyze complex information by analyzing
large pools of data
Support decision makings for the future by discovering
previously unknown patterns
Data mining as main technique
Data Mining
•
Help managers to find hidden patterns and
relationships in large databases to predict future
behavior
–
“If a house is purchased, then new refrigerator will be
purchased within two weeks 65% of the time.”
Web-based DSS for customers

Evaluate and compare real estate prices


Evaluate alternative investment in mortgage
portfolios


fidelity.com (on-line investor center)
Evaluate and compare air fares


Zillow.com: 10402 Loughton Ave. 93311
travelocity.com
Evaluate and compare various automobile prices

aotubytel.com
Executive Information Systems OR
Executive Support Systems





Developed in the late 1980s
Serve the senior management level
Designed mainly to monitor organization’s performance
and address decision makings quickly and accurately
Very user-friendly, supported by graphics
Drill-down capability

EIS drill-down interface design
The Need of EIS




Need for more timely and accurate information for
better decision makings
Need to access internal/external databases to detect
environmental changes
Need to be more proactive due to intensive
competition
Gain computer literacy