You are on page 1of 9

KOMPONEN SDSS

ITNY
KOMPONEN SDSS (GIS BASED)
THE MODEL MANAGEMENT

 The model management component (MMC) of SDSS specifically helps to manage, execute, and integrate different
models
 analytical modeling capabilities within GIS programs include :
a) a locational location model in ArcInfo,
b) ideal point analysis in Common GIS,
c) the analytic hierarchy process (AHP)
d) ordered weighted averaging (OWA) capabilities in IDRISI.
(A) IDRISI Macro modeler, (B) ESRI ModelBuilder

TEKNIK MODEL SDSS

 The spatial decision-making process


is often characterized by a range of
attributes and criteria that are often
in conflict. These situations have
often been addressed with multi-
(C) ERDAS Imagine Model Maker, (D) ERDAS Imagine Knowledge Engineer.
criteria evaluation methods.
 Numerous terms are used to discuss
multi-criteria evaluation methods,
including
a) multi-criteria decision-making (MCDM),
b) multi-criteria decision analysis (MCDA),
c) multi-criteria analysis (MCA).
WEIGHTED LINEAR
COMBINATION

1. define the evaluation criteria or map layers,


2. standardize each criterion map layer,
3. define the criterion weights or weight of
relative importance,
4. construct the weighted standardized map
layers,
5. generate an overall score by adding the
weighted standardized map layers, and
6. rank the alternatives according to the
overall performance score.
ANALYTICAL HIERARCHY
PROCESS

1) it allows the comparison of only two


criteria at once and converts subjective
assessments of relative importance into a
linear set of weights,
2) it is based on discrete decisions that do not
take into account the uncertainty
associated with decision makers’ judgment
(Cheng 1996 and 1999; Lee et al., 2008),
3) the subjective judgment, selection, and
preference of decision makers have great
influence on the AHP results
ARTIFICIAL NEURAL
CELLULAR AUTOMATA
NETWORKS (ANN)

 Neural networks are composed of  Cellular automata (CA) use a discrete


individual processing units (neurons) and representation of a regular grid of cells,
joining interconnected weights. each of which has a finite number of
states such as on or off or alive or not
 The ANN method tests created models alive.
by putting new input data through the
network and analyzing output versus  The time step and the neighborhood to
known conditions to get an accuracy be considered for each cell must be
score. defined. The CA data structure is very
analogous to the GIS raster data model.
FUZZY MODELING
TECHNIQUES

 The fuzzy set theory allows objects or


locations to belong partially to multiple
sets instead of being completely discrete
(Malczewski 1999).
 In fuzzy sets, the degree of membership
in a class is expressed on a continuous
scale between 0 and 1, with 0 meaning
not a member and 1 being a full member
(Burrough and McDonnell 1998).

(a) Vector representation of crisp boundaries between


two soil types and
(b) a rasterized version of these crisp boundaries,
(c) fuzzy representation with values for membership in
the Soil A category, and
(d) fuzzy representation with values for membership in
Soil B category.
TUGAS

1. Carilah jurnal (minimal tahun 2015) 1. Pembagian kelompok maksimal 6


yang berkaitan dengan komponen
SDSS diatas orang :
2. Identifikasi metode yang digunakan a) Weighted linear combination (1, 6)
a) Tujuan b) AHP (2, 7)
b) Data
c) ANN (3, 8)
c) Proses
d) CA (4, 9)
d) Output
e) Software e) FUZZY (5, 10)
3. Email ke lulumari@itny.ac.id  Presentasikan minggu depan
4. Format tgs1_sip_2019 menggunakan ppt

You might also like