MIS 300 Management Information Systems - PowerPoint PPT Presentation

Loading...

PPT – MIS 300 Management Information Systems PowerPoint presentation | free to view - id: 1e1ba5-ZDc1Z



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

MIS 300 Management Information Systems

Description:

... based on a fast computer capable of evaluate one million routes per ... The Bookstore Problem. Decision Problem: How many books to buy to maximize profit? ... – PowerPoint PPT presentation

Number of Views:42
Avg rating:3.0/5.0
Slides: 38
Provided by: marki152
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: MIS 300 Management Information Systems


1
MIS 300 Management Information Systems
  • Model Based Decision Support
  • Expert Systems/AI (Chap 7)

2
Overview
  • OR/MS Video
  • Modeling and Model Examples
  • Expert Systems and AI
  • HW5,6 Help

3
Recall Decision Support Systems
Other CBIS
Internal and External data
Data Management
Model Management
Knowledge Management
User Interface
4
Models
  • Simplified representation or abstraction of
    reality.
  • Capture essence of system without unnecessary
    details
  • Models tailored for specific types of problems
  • Models help us understand the world
  • Prediction (What if?)
  • Optimization (Whats best?)

5
Descriptive vs. Prescriptive Models
  • Descriptive Model
  • Describes a system in terms of parameters and
    variables
  • If we change some input parameter, what will
    happen to our output performance measure?
  • Prescriptive Model
  • Suggests good or optimal solutions
  • Also made up of parameters and variables
  • Searches over many possible solutions to find
    best solution (in some sense)

6
Basic Modeling Concepts
Inputs
relationships
Outputs
roles in model
constraints
Decision Variables
relationships
7
A Simple Modeling Process
Create model
forces detailed examination and thought about a
problem
  • structures our thinking
  • must articulate our assumptions, preconceived
    notions

Validate model - Does it mimic reality well
enough?
Modify model
Use the model to support decision making
  • Searching for general insights
  • Specific numeric answers

8
Exercising the Model Things we might do with a
model
  • Create graphic representation of model parameter
    relationships (visualization)
  • Example GolfClubs_Isken.xls
  • Find values of decision variables that minimize
    or maximize the outputs (optimization)
  • Example SchedulingDSS.xls
  • How do input and/or decision variable values
    affect outputs (sensitivity analysis)?
  • Example GreatThreads_Isken.xls

9
Examples of Specific Questions we might want to
answer
  • How does time spent waiting in line to order a
    Whopper change if we add another staffed cash
    register between 4p-6p?
  • Does the proposed assembly line change increase
    throughput and reduce work in process?
  • How does response time change in our computer
    network if we increase bandwidth by 30?
  • What is the likelihood that this project will
    have a negative NPV?
  • How do different organ allocation policies affect
    the death rates among people on organ transplant
    lists?

10
What is OR/MS? Operations Research/Management
Science
Hyperlink
  • Application of information technology for
    informed decision making.
  • Build models to help understand complex systems
    comprised of people, technology and processes.
  • Related to applied mathematics, MIS, computer
    science, economics, industrial engineering,
    systems engineering
  • Applied broadly in many industries

11
A Few OR/MS Applications
  • Vehicle routing
  • Inventory control
  • Scheduling
  • People
  • Machines
  • Capacity planning
  • Decision analysis
  • Medical
  • investment
  • Financial modeling
  • Design and analysis of telecommunications systems
  • Call centers
  • Military tactics/strategy
  • Healthcare policy

12
The OR/MS Toolbox
  • Statistics
  • Computer simulation
  • Queuing models
  • Forecasting
  • Decision analysis
  • Optimization
  • Computer programming
  • Spreadsheets
  • Databases
  • IT
  • Business knowledge

13
A Shameless Advertisement
  • MIS 446/646 Business Analysis and Modeling
  • Renamed in Fall 2002 (was MIS 436 DSS)
  • I teach every Fall Winter
  • Spreadsheet based modeling course
  • An über business analysis course
  • ExcelAccessVBAmodeling solve tough business
    problems
  • Applications in Finance, Marketing, MIS, POM,
    Accounting, OB

14
Why Spreadsheets?
  • Spreadsheets are the de facto standard platform
    for modeling and analysis in business today
  • The language of business
  • Excel has rich set of modeling and analysis tools
  • Many sophisticated add-ins available
  • Spreadsheet based modeling wave in many business
    schools (Indiana U., Ivey, Dartmouth, Michigan,
    etc.)
  • End user DSS development via VBA
  • A wide open opportunity for stardom
  • Can tie with other products such as DBMS

15
The Traveling Salesman Problem Revisited
  • Find minimum cost route through N cities
  • Visit each city exactly once, return home
  • N! possible tours
  • Classic optimization problem with many variants
  • Applications in vehicle routing (e.g. Fed Ex),
    computer chip manufacturing, robot control

16
TSP is a Big Problem
How long would it take just to list all the
possible routes and calculate their distance?
Days to Evaluate is based on a fast computer
capable of evaluate one million routes per
SECOND!
17
Solving The TSP
18
Overview of Computer Simulation
  • A technique for conducting what if? experiments
    with a computer on a model of a management
    system.
  • Imitation of operation of real world process or
    system over time using a computer
  • generates an artificial history of the process
  • statistical inferences can be drawn from analysis
    of this artificial history

19
The Bookstore Problem
  • Bookstore purchases books at start of term to
    sell

Return unsold Books _at_ this price.
  • Demand is uncertain

20
The Bookstore Problem
  • Decision Problem How many books to buy to
    maximize profit? (DECISION VARIABLE)
  • Assume just going to place one order at beginning
    of term.
  • What makes this problem hard?
  • See MIS300_Walton.xls for spreadsheet simulation
    approach.

21
Bookstore Example Using a Spreadsheet Simulation
Model
(1) Inputs
Given these
Predict these
(2) Simulation Model
A computer program or set of Excel formulas
(3) Outputs
22
Why Simulation?
  • Enables study and experimentation with the
    internal interactions of a complex system
  • Often easier, cheaper to experiment with model
    than real system
  • Knowledge gained in designing simulation
  • Assessing sensitivity to input variables
  • Teaching and/or training tool
  • Experiment with new designs and/or policies

23
The Scheduling Problem
  • Staff works 5 consecutive days
  • Can start any day of the week
  • Ex T, W, Th, F, Sa
  • Objective
  • Minimize total amount of staff needed
  • By Finding
  • Number of employees starting their 5-day
    workstretch each day of the week
  • Subject to constraints
  • Daily staffing requirements are met

24
Solving Optimization Problems with the Excel
Solver
  • Solver is Excel add-in (Frontline Systems, Inc.)
  • Pretty sophisticated optimization program
  • Minimize or Maximize some cell
  • By manipulating values in decision variable cells
  • Subject to constraints (on the decision
    variables) in other cells

25
The Solver Interface
Our model of the scheduling problem
Solver
26
About Optimization Problems
  • Can be very, very, very, very hard to solve
    optimally
  • Very good solutions to very difficult problems
    are found routinely using computerized
  • Mathematical models
  • Intelligent heuristics smart rules of thumb
  • Many, many, many business applications
  • Staff scheduling (airlines, hospitals, call
    centers, etc.)
  • Routing (Fed Ex, supply chain)
  • Production scheduling
  • Financial porfolio planning
  • Computer/telecommunications network planning

27
Artificial Intelligence
  • People, procedures, hardware, software, data, and
    knowledge needed to develop computer systems and
    machines that demonstrate characteristics of
    intelligence

28
Branches of AI/Machine Learning
  • Expert Systems
  • Robotics
  • Vision Systems
  • Natural Language Processing
  • Learning Systems
  • Neural Networks

Example Games
29
Conceptual Model of AI
A broad term
30
Intelligent Behavior
  • The ability to learn from experience and apply
    knowledge acquired from experience, handle
    complex situations, solve problems when important
    information is missing, determine what is
    important, and react quickly and correctly to a
    new situation

31
Natural and Artificial Intelligence
32
Components of an Expert System
How did you reach that conclusion?
The thinking part
Facts, rules, relationships
Can be very difficult to acquire knowledge
33
Rules for a Credit Application
34
Neural Networks
  • Train network with historical data
  • Use network to make predictions on new data

From Alter, S., Information Systems A
Management Perspective, 3rd Ed., Addison-Wesley,
1999.
35
Fuzzy Logic
From Alter, S., Information Systems A
Management Perspective, 3rd Ed., Addison-Wesley,
1999.
36
Case Based Reasoning
http//www.mathemedics.com/
37
Bots or Intelligent Agents
  • Software based robots that perform specified
    tasks on behalf of the user with some degree of
    autonomy
  • Different degrees of intelligence
  • search engines
  • user profile based web watchers
  • adaptive, learning agents
  • Email agents, network security, e-commerce
    (procurement), support of desktop applications
  • http//www.salesmountain.com/
  • http//www.watchguard.com/
  • Agents may collaborate with other agents
About PowerShow.com