Development Techniques for Management Information Systems

1
Development Techniques for Management Information
Systems

• Methods of Personal Business Problem Solving

2
System Development Techniques

• Classical problem solving approach
• SDLC
• Prototyping
• Applications Packages
• Dynamic approach (end-user)
• Outsourcing

3
Classical vs. SDLC
4
Pros and cons of SDLC

• Advantages
• Suited for large systems
• Understandable by users
• Handles complexity well
• Handles structured requirements
• Handles predefined processes
• Repeatability
• Defensibility
• Process is self correcting?
• Documentation

• Disadvantages
• Too sequential - discourage change
• User acceptance sought
• restrictive
• slow
• Costly
• Complex
• Lengthy
• Poor user participation in early phases
• errors are costly to correct during later stages
• can you manage what the user really wants?
• Little room to experiment

5
Classical vs.Prototyping
6
PROTOTYPING
problem definition feasibility requirements build/
use prototype refine evaluate expand a
ccept document final prototype post
implementation review
7
Pros and cons of prototyping

• Advantages
• Flexibility experimentation
• User involvement
• Fast track
• A visual approach
• Help define needs
• Iterative
• Can replace phases of SDLC

• Disadvantages
• No guarantee of meeting user needs
• Lack of documentation
• Not the final product!
• Not fully tested!
• May be difficult to maintain.
• May be difficult to expand

8
Classical approach Application packages
9
Pros and cons of applications packages approach

• Advantages
• Saves on development and programming costs
• Transfer of responsibility
• Guaranteed expertise
• Broad outside support
• Ease of upgrade
• Sharing of development costs

• Disadvantages
• Special requirements not met
• Lack of proprietary assurance
• Un-needed extras
• Cost of customizing
• Control of the RFP/RFQ process (is there one?)

10
Dynamic approach (a.k.a. End-user approach,
Evolutionary approach)
11
DYNAMIC APPROACH

12
Pros and cons of the dynamic approach (end-user
dev., evolutionary technique)

• Advantages
• Development cost reduction
• Reduced inception-to-use time
• Systems meet user needs
• User involvement and commitment
• Freeing of systems departments from some
  obligations

• Disadvantages
• Need active management control
• Lack of standardization
• Conflicts with existing systems
• 4GL tools can not handle complex systems
• Lack of documentation
• Difficult to maintain and upgrade
• May be unstable

13
Is outsourcing a system development methodology?

• Pros and cons?
• When to outsource?

14
Use of development techniques
15
How would you develop a personal MIS for a small
table linen importer?

• Problem is pricing table linens
• table cloths
• napkins
• runners
• place mats
• Review the MOTLI background on the Web
• Set up a pricing model (see example model)
• Check it!
• Are you done?

16
What SAD process are we using?

• What are the problems at MOTLI?
• Need a reliable and flexible pricing tool
• Need to minimize costs
• Need to maximize profits
• Need an ability to evaluate alternatives
• Need an ability to generate management
  information to support decision making tasks
• Need to be able to update and maintain modules as
  needed

17
What SAD process are we using?

• Is a solution feasible?
• Technically feasible? Describe the technology
  needed.
• Organizationally feasible? Describe how MOTLIs
  organization is structured.
• Financially feasible? What would MOTLI have to
  pay for the technology above?
• Schedule feasibility? Estimate how long it would
  take you to build the actual operating system.

18
What SAD process are we using?

• Where is the user now?
• Do you plan to work with the user?
• (See Web site on MOTLI for the rest of the wishes
  the user has)
• Sketch out a solution build it
• So what SAD process is this?
• What do we do next? (phases?)

19
What data do you need?

• Material costs Slide 13
• Import duty Slide 14
• Transport costs Slide 15
• Overhead costs Slide 16
• Profit rate Slide 17
• Competitions prices Slide 18
• Can we build this prototype?

20
Continue with prototyping process

• Build use prototype (iterations?)
• Refine expand as needed (iterations?)
• Design build outputs
• Basic paste approach
• Linking dynamic output reports
• Linking into complex reports
• Evaluate prototype
• Scenarios
• Goal seek
• Solver
• Accept document the prototype
• Train user field final prototype
• Post implementation review
• Were SDLC phases followed in this SAD technique?

21
End Personal MIS Development
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
(No Transcript)
28
Documentation

• Working with your system
• Assumptions table (if any)
• Using "goal seeking" for solutions
• Using the "solver" to find optima
• "What if?" analysis
• Changing variables
• Saving scenarios
• Reports generated
• Sensitivity analysis
• Appendix
• Illustrations

• Introduction
• Hardware requirements
• Software requirements
• How the system works
• Data requirements
• Sources of data
• Format of data
• Importing data
• Updating database
• Recalculating model after data changes (if
  needed)

29
Scenarios with EXCEL

• See Mongolian Imports 2 write-up Part 5
• Open Pricing Model
• Use Tools-Scenarios
• Select cells to change
• Check prevent changes
• Select result cells
• Display Summary

30
Goal seek with EXCEL

• See Mongolian Imports 2 write-up Part 6
• Open Pricing Model
• Use Tools-Goal Seek
• Identify cell which you need to change
• Set cell to value you need to reach
• Identify cell which can change to reach value
  above
• OK will change to goal Cancel will reset.

31
Solver (Optimize) with EXCEL

• See Mongolian Imports 2 write-up Part 7
• Open Pricing Model
• Use Tools-Solver
• Define objective (Target) cell
• Decide on Max/Min
• Decide cells to change
• Input constraints
• Press Solve
• Set Restore Original Values
• Request Answers
• View Answer tab