- Single-goal-preference-belief structure: individual
Individual versus decision making
multiple decision - Varied goal-preference-belief structure among decision
making individuals: group decision making
makers ! Regardless of the number of individuals actually
involved !
Multi-Criteria
us
Decision maker(s)
rs
Decision problem Includes
Ve
Agent: a computer program characterized by properties as:
- Autonomy
In
- capability of taking independent action
clu
- Decision maker: an entity with the responsibility to
Definition
- "A multi-criteria decision - Reactivity
de
problem involves a set of make decisions - capability of sensing and reacting to its environment and other agents
s
- Individual (e.g., searching for a house or an
Decision Making
Includes
alternatives that are - Rationality
evaluated on the basis of apartment) - capability of acting rationally to solve a problem at hand
conflicting and - A group of individuals (e.g., selecting a suitable site Agents
Includes
incommensurate criteria for housing development) - Humanistic characteristics such as preferences, beliefs and opinions can
according to the decision - An organization (e.g., allocating resources for be part of agent behavior
Elements of Multi-Criteria maker's preferences" Computer-based modeling - This makes it possible to represent human decision makers as agents
Decision Analysis (MCDA) acting in a simulated real-world environment
Three main elements:
Criteria
Inc
lud
es
Decision Objectives
- Decision alternatives: alternative courses of action - Decision alternatives are evaluated on
among which the Alternatives the basis of a set of criteria
decision maker (agent) must choose
Inc
- A geographic decision alternative consists of at least Criteria includes:
s
lud
- Objectives
Include
two elements:
es
- Action (what to do?) - Attributes
- Objective: a statement about the
- Location (where to do it?)
desired state of a system under Hierarchical
18
es Structure
lud
consideration
Includes
Inc
In
- Example: a spatial pattern of
clu
- An alternative is completely accessibility to primary schools
Decision
de
specified by defining the values of - Objective indicated the directions of
s
Alternatives the decision variables improvements of one or more
- Decision variables can be classified attributes - The relationships between objectives
Feasible and infeasible into three categories - Either 'the more of the attribute, the and attributes have a hierarchical
Feasible and infeasible
Includes
decision alternatives - Binary better' or 'the less the attribute, the structure
for two criteria: C1
decision alternatives - Yes/no decision better' - Four levels:
and C2, and constrains for two criteria - Discrete - This implies a maximization or - Goals
Includes
C1 > 10 and C2 > 1.5. - Example: number of minimization of an objective function - Objectives
GIS-MCDA Part 1 See the PowerPoint patrons at a shopping mall - Attributes
for the Decision Matrix - Continuous - Alternatives
- Example: facility size
Includes
- Constraints represent restrictions Attributes
Feasible imposed on the decision variables
(alternatives)
Alternatives - They divide decision alternatives into
two categories:
MCDA basic concepts - acceptable (feasible) - Attribute: a property of an element of a real-world
- unacceptable (infeasible) geographic system
- An alternative is feasible if it satisfies (e.g., transportation system, location-allocation system)
all constraints - Example: For the objective of maximizing physical
accessibility to schools, the attributes such as total
traveling distance, time, cost, or any other measure of
spatial proximity
Inc
lud
es Value Scaling
Inc - Mathematical representation of
lud human judgment
es
Value Function - Worth or desirability of that
alternative with respect to that
- Requirement for transforming the criterion
ula
evaluation criteria to comparable units Form
- The procedures for transforming raw - Standardized score values
data to comparable units are referred range from 0 to 1:
to as the value scaling or - 0: the value of the least-
Includes
Inclu
standardization methods desirable outcome
- Score range procedure is the most - 1: the most-desirable score
des
popular GIS-based method for
standardizing evaluation criteria
Include
Includes
Criterion Weighting
(Part 2)*
s
Piecewise linear - In real-world applications of GIS-
MCDA, the value function is often
form value function approximated by a piecewise linear Form
form ula
Combination Rules
(Part 2)*
- Global value function does not
take into account spatial
heterogeneity of the preferences
Local value function that are represented by the
relationship between the criterion
score and the worth of that
score Fo
rm
- Spatial variation of the value ula
function can be operationalized
by the concept of the local range:
Individual versus decision making
multiple decision - Varied goal-preference-belief structure among decision
making individuals: group decision making
makers ! Regardless of the number of individuals actually
involved !
Multi-Criteria
us
Decision maker(s)
rs
Decision problem Includes
Ve
Agent: a computer program characterized by properties as:
- Autonomy
In
- capability of taking independent action
clu
- Decision maker: an entity with the responsibility to
Definition
- "A multi-criteria decision - Reactivity
de
problem involves a set of make decisions - capability of sensing and reacting to its environment and other agents
s
- Individual (e.g., searching for a house or an
Decision Making
Includes
alternatives that are - Rationality
evaluated on the basis of apartment) - capability of acting rationally to solve a problem at hand
conflicting and - A group of individuals (e.g., selecting a suitable site Agents
Includes
incommensurate criteria for housing development) - Humanistic characteristics such as preferences, beliefs and opinions can
according to the decision - An organization (e.g., allocating resources for be part of agent behavior
Elements of Multi-Criteria maker's preferences" Computer-based modeling - This makes it possible to represent human decision makers as agents
Decision Analysis (MCDA) acting in a simulated real-world environment
Three main elements:
Criteria
Inc
lud
es
Decision Objectives
- Decision alternatives: alternative courses of action - Decision alternatives are evaluated on
among which the Alternatives the basis of a set of criteria
decision maker (agent) must choose
Inc
- A geographic decision alternative consists of at least Criteria includes:
s
lud
- Objectives
Include
two elements:
es
- Action (what to do?) - Attributes
- Objective: a statement about the
- Location (where to do it?)
desired state of a system under Hierarchical
18
es Structure
lud
consideration
Includes
Inc
In
- Example: a spatial pattern of
clu
- An alternative is completely accessibility to primary schools
Decision
de
specified by defining the values of - Objective indicated the directions of
s
Alternatives the decision variables improvements of one or more
- Decision variables can be classified attributes - The relationships between objectives
Feasible and infeasible into three categories - Either 'the more of the attribute, the and attributes have a hierarchical
Feasible and infeasible
Includes
decision alternatives - Binary better' or 'the less the attribute, the structure
for two criteria: C1
decision alternatives - Yes/no decision better' - Four levels:
and C2, and constrains for two criteria - Discrete - This implies a maximization or - Goals
Includes
C1 > 10 and C2 > 1.5. - Example: number of minimization of an objective function - Objectives
GIS-MCDA Part 1 See the PowerPoint patrons at a shopping mall - Attributes
for the Decision Matrix - Continuous - Alternatives
- Example: facility size
Includes
- Constraints represent restrictions Attributes
Feasible imposed on the decision variables
(alternatives)
Alternatives - They divide decision alternatives into
two categories:
MCDA basic concepts - acceptable (feasible) - Attribute: a property of an element of a real-world
- unacceptable (infeasible) geographic system
- An alternative is feasible if it satisfies (e.g., transportation system, location-allocation system)
all constraints - Example: For the objective of maximizing physical
accessibility to schools, the attributes such as total
traveling distance, time, cost, or any other measure of
spatial proximity
Inc
lud
es Value Scaling
Inc - Mathematical representation of
lud human judgment
es
Value Function - Worth or desirability of that
alternative with respect to that
- Requirement for transforming the criterion
ula
evaluation criteria to comparable units Form
- The procedures for transforming raw - Standardized score values
data to comparable units are referred range from 0 to 1:
to as the value scaling or - 0: the value of the least-
Includes
Inclu
standardization methods desirable outcome
- Score range procedure is the most - 1: the most-desirable score
des
popular GIS-based method for
standardizing evaluation criteria
Include
Includes
Criterion Weighting
(Part 2)*
s
Piecewise linear - In real-world applications of GIS-
MCDA, the value function is often
form value function approximated by a piecewise linear Form
form ula
Combination Rules
(Part 2)*
- Global value function does not
take into account spatial
heterogeneity of the preferences
Local value function that are represented by the
relationship between the criterion
score and the worth of that
score Fo
rm
- Spatial variation of the value ula
function can be operationalized
by the concept of the local range:
