Professional Documents
Culture Documents
This article briefly presents only the steps used to make a cartographic expert system
for sustainable development in order to create the necessary maps, used in risks management
and hazards caused by floods, landslides and earthquakes, which incorporates: statistical
analysis of experimental data, optimization in decision-making processes modelling, planning
and network control of complex projects, simulation of complex decision-making processes,
etc.
1. INTRODUCTION
In order to update the topographical and cadastral map, we used the drone.
Taking into account all of these we started the flight mission. To get the most accurate
results of the scanned reality, it is recommended to have a large number of aerial photographs,
because, in this way, when the image will be processed an accurate 3D model will be
obtained, and not an 2,5 D one, like in the majority of cases
2.3. Creating the cartographic data base as support for the expert system
Creating the information database of the integrated global resources, of the environment, on
a certain area of land both inside and outside the city, by realizing the fundamental
components of the informational system of the territory, namely:
• The general topographic map of the territory, including representation of relief.
.
Fig.4. Topographic maps in the buildable pilot – area
• The cadastral map of real estate from buildable and built-up area ( project
application)
• Pedological map and pedological profiles
Fig. 7. Types of data contained in graphic and non-graphic databases of the Expert System
⚫ Knowledge base
⚫ Inference engine
⚫ Facts base
The WEPP (Water Erosion Forecasting Program) was conducted by the US Agricultural
Research Service to estimate the risk of erosion of sloping land. Several versions have been
developed to improve computer algorithms. It is supported by major US agencies, tested in
many laboratories in the US, Australia and Europe. The WEPP program uses a complex
database (75 parameters) that collects climatic, relief, soil, plant, agro-technical, irrigation,
drainage and soil erosion records. The main input and output parameters of the WEPP and
EPIC models are the following (Table 1):
Table 1
INPUTS
1. Relief data: Area, Latitude and longitude of river basin; aspect, slope, slope length; length slope
hydrographic network
3. Pedological data: Number of horizons
2. Climate data: Extreme precipitation (EPIC, (EPIC, WEPP); Texture (EPIC; WEPP);
WEPP) Albedo (EPIC; WEPP); Erosion (EPIC;
Rainfall (EPIC; WEPP) WEPP); Apparent Density (EPIC; WEPP);
Average rainfall (WEPP) Field capacity (EPIC; WEPP); Organic
Torrential Core Intensity (WEPP) Matter Content (EPIC, WEPP); Content of
Maximum and minimum temperatures (EPIC; rock fragments (EPIC; WEPP); Cationic
WEPP) exchange capacity (EPIC; WEPP);
Dew Point Temperature (WEPP) Concentration of organic nitrogen, nitrates,
Wind speed and direction (EPIC; WEPP) mobile phosphorus, organic phosphorus,
organic carbon, calcium carbonate,
vegetable waste, etc. (EPIC)
5. Vineyard data: crop rotation (EPIC;
4. Type of works: (EPIC; WEPP); Number of WEPP); Soil Coefficient with Vineyard
jobs (EPIC, WEPP) Workout Date (EPIC; Plants (EPIC; WEPP); Cutting height at
WEPP); Depth of soil mobilization (EPIC; harvest (EPIC; WEPP); Plane dimensions -
WEPP); Number of combine elements, seed coefficient (EPIC; WEPP); Growth Ratio
drills, etc. (EPIC, WEPP); Irrigation (EPIC, (EPIC; WEPP); Plane spacing (EPIC;
WEPP); Drainage (EPIC; WEPP); Fertilization WEPP); Depth of penetration of roots
(EPIC); Pesticide Treatments (EPIC) (EPIC; WEPP); Frost Temperature (EPIC;
WEPP); Stress Production (EPIC, WEPP)
OUTPUTS
- Soil leakage (EPIC, WEPP);
- Erosion (maximum and average deviation, maximum and average deposition) (EPIC, WEPP);
- Wind erosion (EPIC);
- Sediment Production (WEPP);
Specific sedimentation weight (WEPP);
- Sediment granulometry (WEPP);
- evapotranspiration (EPIC);
- Nitrogen and phosphorus lost through liquid and solid leakage (EPIC);
- Plant-derived nitrogen and phosphorus (EPIC);
- Agricultural production (EPIC);
- Amount of biomass (EPIC);
- Soil and Soil Residues (EPIC), etc.
2.5. The Interference Engine
The inference engine (or mechanism) represents the novelty of the expert systems, it takes
from the knowledge base the facts used to construct the reasoning, the inference mechanism
follows a series of major objectives, such as:
⚫ choosing the control strategy according to the current problem
⚫ developing the plan to solve the problem as needed
⚫ switching from one control strategy to another
⚫ executing the actions provided in the resolution plan
Although the mechanism of inference consists of a set of procedures in the usual sense
of the term, the use of knowledge is not provided by the program but depends on the
knowledge it has at its disposal.
For example, we chose to set up the typical meteorological year for the pilot area required to
determine the flood curves. In order to take into account as many meteorological
characteristics a weight is made. Taking into account the algorithm given above, a
representative month of a given year is obtained.
These 12 months are concatenated and the leveling at the transition from one typical month to
another is done by a weighted mean of the neighboring values with the weights w = [0.3, 0.4,
0.3].
If n is the number of data involved in the leveling operation (for us n = 8, the last 4 days of
a month and the first 4 days of the next month, the calculation sequence is:
for i=2:n-1
date1(i)=w(1)*date(i-1)+w(2)*date(i)+w(3)*date(i+1);
end
where data is the vector of the initial values and data1 is the vector of the leveled values
12
10
0
ian feb mar apr mai iun iul aug sept oct nov dec
atm 7.38 7.82 7.24 10.99 9.17 11.01 10.15 9.84 9.89 8.47 7.44 6.48
2006 5.5 6.5 7 8.2 8.7 9.2 9.2 9.8 8.7 8.7 7.5 6.5
The complexity of economic, biological, technological processes, the race for market
supremacy makes it impossible to solve problems through the art of leadership, with the help
of common intuition and judgment. All this needs to be solved in the conditions of increasing
structural and functional complexity of the systems, the raising of the level of technicality
and corresponding to some accentuated specializations of the professions. It becomes
necessary to develop the scientific decision on the basis of calculations made with measurable
and quantifiable sizes, with the purpose of choosing a variant from which an optimal result is
obtained.
Economic performance in agriculture depends, primarily, on the availability and
quality of natural resources. Environmental damage can affect in a negative form the
agricultural productivity and revenue, which can put pressure on the economic and social
structure in rural communities.
An important category of IT tools used to solve problems related to sustainable
development in agriculture are decision support systems. These tools are also called Decision
Support Systems - SSD (Decision Support Systems ).
The multi-criterial decision model consists in choosing the optimal variant of a finite
set of variants that compare to each other in relation to a finite set of criteria, each pursuing a
goal: the maximum or the minimum.
The decision-making process involves evaluating several decisional variants in order
to choose one. As an optimal variant in relation to a criteria can be suboptimal with respect to
other criteria, it is looking for the variant which matches the best compromise for all criteria.
Starting from the matrix of consequences (aij) with i = 1,2, ..., m and j = 1,2, ..., n
and how each criterion can be expressed in a variety of units of measurement, we need the
normalized matrix the consequences (qij) with (i, j) (1,2, ..., n) x (1,2, ..., m) by following
transformations:
• for the criteria that follow the maximum the applied transformations are:
aij
qij = sum , unde aijsum = aij (1)
aij j
- and for the minimum criteria:
(2)
The type of normalization chosen makes it possible to use formula (3) for this indicator of
information, being fulfilled then the condition h j [0,1]
The degree of diversification of information given by the results of criteria j can be defined
by
d j =1 -hj (4)
dj
pj = n
d
j =1
j (5)
We obtain the coefficients of importance (p0j) where we took into consideration the weights
given by the decident λ = ( 𝝀1, λ2,…, λn ). The evaluation of the importance of the criteria is
done by means of the coefficients of significance j = 1,2,….. n There are situations in which
not all criteria are equally important and a certain degree of subjectivity is introduced by
the intervention of the decision-maker.
Tj .p j with n and n (6)
po j = n
T
j .p j
po j = 1 j = 1
j=1 j =1
j=1
The utility function is calculated by an additive counting method. The optimal variant is
the one that
n
corresponds to the maximum utility
ui = po j .qij with i = 1,2, .., m. (7)
j =1
We click the Multi-Criteria choice. The dialog that serves to input the input data has on the
right the brief presentation of the steps to be followed for entering, correcting, processing the
data, showing the utilities of each variant, keeping the data and the results in a document with
the .txt extension for storage or printing, resuming the algorithm for another processing.
Data Mining: a process of extracting new information from existing data collections,
information that can answer both questions: What is happening? but also why it's happening?
]
Recent computer science achievements have been marked by a reorientation in the
use of accumulated data volumes, moving from retrospective research to research that
addresses future issues, allowed by the evolution of Data Mining technologies
Data Mining classifies and aggregates data from different and even incompatible
systems, looking for new associations.
Advanced analysis techniques are used including statistical algorithms, artificial
intelligence, neuronal networks, fuzzy logic, genetic algorithms, data visualization.
A special place among the tools with which Data Mining operates is Cluster Analysis
We use the linear projection model to compare the 12 months of TMY achieved for the pilot
area. Comparing elements are: air temperature (mean, minimum, maximum), amplitude,
rainfall, duration of sunshine
The result: We notice three clusters (three possible seasons)
Cold one with November, December, January, February and March, warm with May, June,
July, August and the transition season with April, September, October. One month of
transition from cold to hot (April) and the transition from hot period to cold period is done in
September and October.
One month of transition from cold to hot (April) and the transition from hot period to cold
period is done in September and October.
On a multicriteria analysis model defined on information from the knowledge base
achieved on IMEV project with the goal of choosing the optimal rootstock for the
Sauvignon variety, we tested the correct functioning of the two decision making
software created by the Mathematics group: VIP1 and the Nonlinear Projection,
applied here as a preprocessing data procedure as well as an alternative.
Planning risks management activities is important for the level, type and visibility of risk
management to be proportionate to their risks and importance.
The result will be a risk management plan - that will address methodological issues (defining
approaches, tools and sources of information to be used in risk management processes),
responsibilities, budgeting, programming of interventions, etc.
. A risk management information system includes the following modules::
Entreprise Relationship Planning-ERP, which integrates one of the main processes in the field
of economics and finance: funding sources, material and human resource management,
planning and monitoring of activities, etc.
- Customer Relationship Management (CRM) - the entire set of interactions between
decision makers and operators). CRM facilitates the provision of services via the Internet,
telephone, ATM - Automated Teller Machine / Kiosk, etc .;
- E-Business - the web access of ERP systems. It can take one of the forms: direct links
between information from risk areas (B2B), historical information from existing and up-to-
date databases (B2C), knowledge bases about the risk factors (natural calamities - SCM) –
fig.9,
- Business Intelligence (BI) – applications for data collection, storage and processing
in order to make decisions. BI applications include activities such as:
Decision Suport Systems- SSD
Online Analytical Processing – OLAP
Data Warehouse
Data Mining
As it can be seen, there is a relationship between ERP, CRM, E-Business and
Business Intelligence components Business Intelligence.
Fig 9. The Relationship between ERP, CRM, E-Business and Business Intelligence components for
Hazard and Risk in Natural Disasters
CRM systems interact with ERP components to perform the desired functions. In this
respect (B2C or B2B), calls ERP`s for processing demands or ERP processed data are
periodically archived in warehouse repositories and processed using OLAP tools etc..
The results of analyzes are used by BI components to make decisions and develop
action plans to minimize the effects of the risks.
3.CONCLUZII
This project seeks to provide the decision makers the tools needed to achieve the
objectives of sustainable development of the area defined by the ATU by:
• develops a formal information system for the conservation of the natural resources of
the habitat, the economic and social development, the protection of the environment
and the prevention of disasters;
• it integrates within the Rural Center for Sustainable Development and it is the
possibility of studying and informing both decision makers and citizens on the
possibilities for identifying and mitigating the effects of natural disasters and poverty
facing the studied area by knowing the potential for sustainable development of the
area represents a complex informational integrated system by creating a support-
digital map- in which the environmental information is represented spatially,
overlapping cadastral data, data on the underground with the specification
of the geomorphological structure and the geological resources, data concerning
the potential of the land by pedological soil mapping, ambient environment, data
about climate, hydrography, etc., which allows an impact analysis of
disruptive climate factors, soil pollution factors, water, air pollution, allows
the economic evaluation of resources in order to eradicate poverty, create
opportunities for development, create jobs and encouraging development
initiatives of the local economy, and correlated with social data and information, the
system can establish through a specific zoning the vulnerabilities to discrimination
concerning the equality of opportunities, access to education, health insurance, or
fight against poverty.
The main results obtained through this project are the following:
making the updated digital map of the area;
the implementation of the information system for monitoring climate change;
creating an interface between the computer system and users, allowing active participation in
the collection of meteorological data and information, their introduction nto the system and
processing them with specialized programs, in order to zone the study area to obtain the
following spatial information:
• flood risk areas
• areas with risk of landslide
• polluted zones
• areas affected by erosion
• deforested areas
• classification of land by use categories and soil quality classes, etc.
• identifying the real possibilities for removing the "disparities" by highlighting the
economic resources of the area
• NGO collaboration with representatives of the local administration to identify factors
of economic progress
Addressed to: Decision makers for the economic and social development of the area, all
citizens in the proposed area.The project will be a pilot project for generalized application in
other areas.
Bibliography
http://www.scritub.com/tehnica-mecanica/SISTEME-EXPERT-SI-NEUROEXPERT441911164.php
http://www.utgjiu.ro/math/mbuneci/book/exp/cap08.pdf
http://www.mpt.upt.ro/doc/curs/gp/Sisteme_inteligente_in_electrotehnica/Inteligenta_artificiala_si_Retele
_neuronale_cap1.pdf
http://www.matrixrom.ro/romanian/editura/domenii/cuprins.php?cuprins=SF90
http://steconomiceuoradea.ro/anale/volume/2005/statistica-si-informatica-economica/13.pdf
http://safetyinknowledge.blogspot.ro/2015/01/dezvoltarea-unui-sistem-expert-pentru.html
http://www.revista.forter.ro/2009_3_t/06-arm/01.htm
https://www.icpa.ro/proiecte/Proiecte%20nationale/ExET/ExET14.pdf