Design a rule based expert system for eia

DESIGN A RULE-BASED EXPERT SYSTEM FOR
ENVIRONMENTAL IMPACT ASSESSMENT (EIA)
B. S. PANDA1*, RAHUL ABHISHEK2
1

Asst. Professor, Dept. of CSE, MITS Engg. College, Rayagada, Odisha.
5 semester IT student, MITS Engineering College, Rayagada, Odisha.
E-mail: bspanda@sify.com1, rahulmithu.abhishek@gmail.com2

2 th

ABSTRACT:
Knowledge or Rule based Expert systems systems are widely used in engineering
applications and in problem-solving. Rapid development today has brought with it
environmental problems that cause loss or destruction of natural resources.
Environmental impact assessment (EIA) has been acknowledged as a powerful
planning and decisionmaking tool to assess new development projects. It requires
qualified personnel with special expertise and responsibility in their domain. Rulebased EIA systems incorporate expert’s knowledge and act as a device-giving
system. The system has an advantage over human experts and can significantly
reduce the complexity of a planning task like EIA.
Keywords: Expert systems, EIA, Knowledge engineering, Problem-solving.

1. INTRODUCTION
Rapid development today has given rise to
environmental problems that cause loss or
destruction of natural resources. Decisions
on environmental issues and on sustainable
development is about the better management
and decisionmaking from a higher authority,
for the current environmental impact
assessment (EIA) activity in harmony with
the environment [1]. EIA can be defined as a
systematic activity designed to identify,
predict and evaluate the environmental
impacts of a particular action [2]. It
interprets and communicate information
about these impacts, and details out the
mitigating measure prior to approval and
implementation [3, 4].
The process of EIA seeks to avoid costly
mistakes in project implementation, either of

environmental damage or of subsequent
modifications in order to make the project
environmentally acceptable. It involves the
collection, analysis, interpretation and
review of an extensive amount of
information. EIA is then needed as a
powerful tool to aid decision-makers and to
quantify the impacts prior to any new
development for better problem-solving.
2. Rule-Based Expert System
Rule-Based Expert systems are computer
programs, designed to make available some
of the skills of an expert to nonexperts.
Since such programs attempt to emulate the
thinking patterns of an expert, it is natural
that the first work was done in Artificial
Intelligence (AI) circles [5, 6, 7].

AI's scientific goal is to understand
intelligence by building computer programs
that exhibit intelligent behavior. It is
concerned with the concepts and methods of
symbolic inference, or reasoning, by a
computer, and how the knowledge used to
make those inferences will be represented
inside the machine. AI programs that
achieve expert-level competence in solving
problems in task areas by bringing to bear a
body of knowledge about specific tasks are
called knowledge-based or expert systems.
The area of human intellectual endeavor to
be captured in an expert system is called the
task domain. Task refers to some goaloriented, problem-solving activity. Domain
refers to the area within which the task is
being performed. Typical tasks are
diagnosis,
planning,
scheduling,
configuration and design.
One of the most intelligent features of expert
system structure is the implementation of
knowledge modelling. Knowledge based
systems may reduce significantly the
complexity of a planning task such as EIA.
This task requires qualified personnel with
special expertise and responsibility to be
involved in decision-making within their
domain. The development of an expert
system for such applications can incorporate
the expert’s knowledge and act as a devicegiving system. challenge of dealing with real
problems in an EIA, requires more
flexibility in its domain. Knowledge sources
and the solution space are usually structured
hierarchically as they would be in a human
expert.
In this approach, various sources ofexpert
knowledge in EIA can participate in forming
and modifying the emerging solution. Expert
knowledge from different domains is shared
in the system to infer an adequate
information and accurate decision, and

provide a powerful explanation facility
“WHY” and “HOW” to the end-user.
3. EIA PROCEDURE AND
STRUCTURE
Results from an EIA study should be
presented in a form that is readily
understood and can be utilised in project
planning and implementation. Under the
EIA Act [8], nineteen categories of activities
are prescribed and these include those
related to agriculture, airport, drainage and
irrigation, land reclamation, fisheries,
forestry, housing, industry, infrastructure,
ports, mining, petroleum, power generation,
quarries, railways, transportation, resort and
recreational development, waste treatment
and disposal, and water supply. In order to
conduct an effective EIA study, three major
steps should be taken. The preliminary
assessment stage should be initiated at the
pre-feasibility stage of the development of
an activity. Project options and any
significant residual impacts are identified at
this stage. A detailed assessment is
undertaken for those projects for which
significant residual environmental impacts
have been predicted in the preliminary
assessment. The review stage of EIA is
carried out internally by the DOE for
preliminary assessment reports and by the
ad-hoc Review Panel for detailed
assessment
reports.
Recommendations
arising out of the review are transmitted to
the relevant project approving authorities for
consideration in making a decision on the
project. The outcomes of the EIA reports
and format are divided into chapters and
sections. These chapters include the
Introduction, Title of the Project, Project
Initiator, Statement of Need, Project
Descriptions, Project Options, Description
of The Existing Environment, Potential
Significant
Impacts,
Mitigation and
Abatement Measures, Residual Impacts,

Summary of Conclusions, Data Sources and
List of References. This paper discusses on
expert system developed to produce EIA
reports according to the chapters mentioned.
The system is able to organise all
information in each chapter to produce a
complete reports.

the developer or the project manager is
actually complying with the approved
guidelines. For these reasons, the
development of a comprehensive expert
system to aid in preparing EIA reports as
well as for decision-making will be helpful.
5.1 The Expert System Kernel

4. SYSTEM COMPONENTS
At the development stage of the expert
system [9], four major components were
constructed. The generic system includes the
domain knowledge, the expert system
kernel, solution methods and strategies and
the user interface. Fig. 1 shows the general
components of the expert system structure.

The central part of the expert system is the
kernel, and it contains three main modules
as shown in Fig. 2. The first module is the
working memory and it is connected to an
intelligent user interface unit. Through the
interface unit, the end-user can communicate
with the expert system by selecting any of
the EIA prescribed activities. The selected
activity will be recorded in the working
memory and the inference engine
mechanism and control procedures initiate
the knowledge base and load all information
for that category into the working memory.
Inputs of the problem parameters will be
acquired from the user and the reasoning
mechanism activates rules to search the
knowledge base and fires the activated one
to form solution as an output.

Fig. 1: General Components of the Expert System
Structure

5. SYSTEM OBJECTIVE
Currently, EIA approval is mandatory for
most of the development projects to be
undertaken in Malaysia. However, there are
no standard procedures to follow by the
consultants in preparing the EIA reports.
The Department of Environment (DOE)
encounters problems in evaluating (to
approve, reject or approve with conditions)
the EIA reports. If a report is approved, the
DOE also faces problems of verifying that

Fig. 2: Problem/Solution to Expert System
Kernel

5.2 The System Domain Knowledge
Knowledge is the key to expert systems
functionality. One of the most difficult tasks
in developing expert systems involves
information gathering and knowledge
extraction. The system domain knowledge
was acquired from various disciplines. The
main source was the DOE through the
approved EIA reports and the EIA experts
and consultants. To maintain knowledge in
the expert system, a knowledge acquisition
unit was developed. A database system was
also attached to the expert system.
Data was gathered from sources such as
environmental agencies, the literature,
interviewing experts in their domain and
subsequently translated to form the “rule of
thumb”. These rules were used to generate
the expert system rules using the C
Language Integrated Production System
(CLIPS).
6. The knowledge base structure and
Eia selector
Owing to the categorisation of the EIA
prescribed activities schedule, the design of
the knowledge-based expert system structure
comes in more attractive knowledge
representation. By experience, an expert
forms several sets of rules on a given
problem. These sets of rules reflect the skill
of the expert on the given problem. For this
reason, a set of rules may be applied to the
given task when needed, while on the other
hand the same set of rules is inapplicable to
other problems. Thus, in the case of separate
categories of EIA, a modular structure is
implemented in the system.
6.1 System Strategy and Reasoning
Human solves problems by combining facts
with knowledge. Experts first obtain facts
about the problem and store them in the

short-term memory. Experts then combine
facts with the long-term memory knowledge
and use specific rules of heuristic to derive
logical conclusions.

Fig. 3: Selecting EIA Category
A cooperative expert systems approach
requires mechanisms of cooperative
interactions that permit multiple experts to
work together on the solution of a common
problem. By understanding how human
reasons from their wisdom and expectations,
the system was developed in a manner to
respond to the direction of the data captured
during the acquisition stage. The system
contains over 700 nested rules of type IF
Condition(s)-THEN Action(s) rule structure.
For example, the following rule reports to
the Noise index and the Air pollutant index
for projects under the industrial category.
IF
(?Project_Type (Is_Industries))
AND (eq?Project_Loc (Within_Industrial
_Area))
THEN (Get_NoiseIndex)
(Get_AirPollutantIndex)
(Report_BB)
Normally in solving a problem, an expert
will use knowledge that directs to the
problem-solving. The system implements
this type of knowledge base called
metaknowledge, or in other words,

knowledge about the use and control of
domain knowledge. This knowledge is used
in searching the existing domain knowledge
to determine the problem solution and infer
new knowledge. An advantage of the
modular structure is that it provides a natural
ordering of the domain’s rule sets by a topdown approach, where a specific module is
used when appropriate. Through the enduser interface facility the knowledge base
inference allows inspection of individual
rules, and a recursive explanation function
“WHY” traces the reasoning process of the
system.
7. CONCLUSION
The main objective of EIA is to ensure the
potential problems are foreseen and
addressed at an early stage in project
planning and design. The task of EIA
requires a qualified personal and expertise to
prepare EIA reports and involves the
collection, analysis, interpretation and
review of an extensive amount of
information. A Rule-based expert system
was developed to produce EIA reports in a
significant time. The system has an
advantage over traditional solution to EIA
where these solutions are derived from the
application of a substantial body of
knowledge rather than applying an
imperative algorithm.
References
[1] J. Glasson, R. Therival, A. Chadwick.
“Introduction to Environmental Impact
Assessment: Principles and Procedures,
Process, Practice and Prospects”. UCL
Press Ltd., 1994.
[2] M. Masera and A. G. Colombo.
“Contents and Phases of an EIA Study”.
Environmental Impact Assessment.
Kulwer
Academic
Publishers,
Netherlands. 1992, Vol. 1, pp. 53-77.

[3] R. E. Munn. “Environmental Impact
Assessment:
Principles
and
Procedures”.
United
Nations
Environmental Programme (UNEP),
Environment Canada and UNESCO.
SCOPE. Report 5, Toronto, Canada,
ICSU-SCOPE, 1995.
[4] Y. Samiullah. “Environmental Impact
Assessment: Principles and Practice”.
National Workshop on Environmental
Monitoring and Assessment in Tropical
Urban Applications, 24-29 November
1986, UKM, pp. 323-353.
[5] Hayes-Roth, F., D.A. Watchman, and
D.B. Lenat, eds. (1983), Building Expert
Systems, Addison-Wesley, Reading,
Mass.
[6] Hyde, Andrew Dean (Sept 28, 2010),
“The future of Artificial Intelligence”.
[7] Jackson, Peter (1998), Introduction to
Expert Systems (3rd Ed.), Addison
Wesley, New-Delhi.
[8] Malaysia, Department of Environment.
“Environmental Impact Assessment:
Procedures and Requirements in
Malaysia”. Kuala Lumpur, Department
of Environment, Ministry of Science
Technology and the Environment, 1994.
[9] Nikolopoulos, Chris (1997), “Expert
Systems: Introduction To First And
Second
Generation
And
Hybrid
Knowledge Based Systems”, Mercell
Dekker INC.

Design a rule based expert system for eia

More Related Content

What's hot (20)

Similar to Design a rule based expert system for eia (20)

More from Er. rahul abhishek (12)

Recently uploaded (20)

Design a rule based expert system for eia