Soft Modelling and Problem Formulation

1
Soft Modelling and Problem Formulation

• Simon French
• simon.french_at_mbs.ac.uk

2
Problem FormulationSetting the Context

• Decisions do not arrive with the options and
  their pros and cons catalogued.
• Rather issues arrive
• need to identify objectives and think creatively
  about the means to achieve them

3
Building an understanding

• How can we catalyse the process of giving shape
  to the issues and relevant data?
• How can we share and record that understanding?
• Process
• open brainstorming and discussion
• focus on objectives
• draw together issues into several perspectives
  and frame problem
• build model(s) and conduct analyses.

4
Building an understanding

• How can we catalyse the process of giving shape
  to the issues and relevant data?
• How can we share and record that understanding?
• Process
• open brainstorming and discussion
• focus on objectives
• draw together issues into several perspectives
  and frame problem
• build model(s) and conduct analyses.

What we shall try to cover in this talk
5
Menegitis Scenario

• Over the past year the incidence of meningitis in
  small children has been on the increase. It has
  become apparent over the past months that the
  cause of this increase is that the bacteria which
  cause the disease, always present in the throats
  of much of the population, have crossed over into
  the dog population. Puppies and dogs licking
  children are transmitting the disease. The
  Ministry of Health in your country are debating
  what measures to take to deal with the issue.
• Use some of the techniques that we have been
  using to formulate the problem, identify
  objectives, key uncertainties, stakeholders,
  possible actions, etc.
• This is a hypothetical example!!!

6
Brainstorming

• Simply get a group to list ideas that seem
  relevant without evaluation
• Built on the notions
• on idea triggers another
• all ideas have equal value a priori
• Write up on whiteboard or Post-its

7
Electronic Brainstorming
8
Electronic Brainstorming
http//www.groupsystems.com/default.htm
9
Value focused thinking

• Early on brainstorm values and objectives
• Values are what we care about. As such, values
  should be the driving force for our decision
  making. They should be the basis for the time
  and effort we spend thinking about decisions.
  But this is not the way it is. It is not even
  close to the way it is.
• Keeney (1992)
• More creative
• alternative focused thinking closes down the mind
• value focus thinking opens it up

10
Check-lists

• Simply an aide-memoire
• Used to prime brainstorming
• Used to structure reports

11
PEST and 7 Ss

• External environment
• Political
• Economic
• Social
• Technical

• Internal Environment
• Strategy
• Structure
• Systems
• Style
• Shared values
• Skills
• Staff

12
SWOT
13
Simple two dimensional plots

• Easy to draw on paper or flip charts
• Even better use post-its

14
Stakeholder Identification
15
Stakeholders involved in Asthma Drug Scare
16
Uncertainty identification
17
Networks

• Can show inter-relations

18
Mindmaps
19
Cognitive Mapping
20
Preference ModellingAttribute Hierarchies
From the International Chernobyl Project
21
Building Hierarchies

• Top down
• Bottom up (brainstorming and gathering)
• Iterative, both top down and bottom up
• Checks
• Why do you prefer this to that?
• Best of all worlds and worst of all worlds

22
Types of Attribute

• Natural
• Subjective or constructed
• Proxy

An objective is an attribute plus a direction of
preference
23
Decision Tree Model
Building tree helps dms think about key
uncertainties and contingencies
24
Influence diagrams
IDs allow DMs to see dependencies
25
Role Activity Diagram
26
Rich Picture Diagrams

• A picture speaks a 1000 words ...

27
Rich picture diagram of hole in the ozone layer
issues as perceived in 1988
From Daellenbach (1994)
28
(No Transcript)
29
(No Transcript)
30
Combine plots ..
31
Invent PlotsStrategic Issues in a CS Dept.