trodd_problem_solving_05

Solving Problems &
Making Decisions
Dr. Nigel Trodd
Coventry University
United Kingdom
Introduction
• Analysts solve problems.
• Firstly, you define a problem.
• And then, you ask the right question.
• Only then can you start to think about
how to solve a problem.
Learning objectives
In this lesson you will critically examine 2
approaches to solving problems and/or
making decisions with GIS
– Rational decision making
– Scientific hypothesis testing
st
1
approach:
rational decision making
Decision making
Decision making requires you to:
• generate information
• apply criteria
Decision making in GIS
 Produce maps
 Analyse situations
 Explore data
Rational decision making
A rational person (a.k.a. homo economicus)
• identifies all the consequences of all
possible actions and
• selects an action to optimize his/her wellbeing (minimise adverse consequences)
Process of rational decision making
1. Produce all candidate actions
2. Exclude action by non-compensatory criteria
3. Evaluate utility of remaining candidate
actions using compensatory criteria and
weights.
4. Merge criteria into a single unified criterion.
5. Select ‘best’ action i.e. action with highest
utility or lowest cost.
Example:
site a nuclear waste disposal facility
 What factors?
 Geology, population, transport, conservation
 What criteria?
 favourable deep geology, low population
density, close to road/railway, low conservation
value
Factor maps… suitability
Geology
Population
Transport
Conservation
overlay… and select location
Overlay suitability
maps
Suitable areas
Zoom in
but
Emotional aspects
Experience shows that users ‘play’ with
weights until the solution ‘feels’ right.
nd
2
approach:
hypothesis testing
Making decisions
by testing hypotheses
Set up a hypothesis and try to establish whether
the evidence is sufficient to accept or reject the
idea.
Consider the legal equivalence:
– In a court of law a man is innocent until proven
guilty and a judge or jury would only convict if the
case was proven beyond reasonable doubt
Hypothesis testing
• A research hypothesis consists of 3 parts
– Statement
• e.g. A = B, A > B
– Explanation
• because of x, y and z reasons
– Assumptions
• The hypothesis is valid only under certain conditions
e.g. the world is flat (this makes the physics much easier
when you are modelling!!)
Testing an hypothesis statistically
A sample is analyzed, and
inferences are made about the
population from which the sample
was drawn
but
Hypothesis testing with spatial data
• Frequently the data represent all that
are available
• Tobler’s Law
…and finally
… and finally
• Getting the ‘right‘ answer is all about
asking the right question.
• Ask the right question and you might get
the right answer.