You are on page 1of 8

Waste Management 26 (2006) 299–306

www.elsevier.com/locate/wasman

Municipal solid waste landfill siting using intelligent system


Omar Al-Jarrah a, Hani Abu-Qdais b,*
a
Department of Computer Engineering, Jordan University of Science and Technology, Irbid 22110, Jordan
b
Department of Civil Engineering, Jordan University of Science and Technology, P.O. Box 3030, Irbid 22110, Jordan
Accepted 26 January 2005
Available online 12 July 2005

Abstract

Historically, landfills have been the dominant alternative for the ultimate disposal of municipal solid waste. This paper addresses the
problem of siting a new landfill using an intelligent system based on fuzzy inference. The proposed system can accommodate new
information on the landfill site selection by updating its knowledge base. Several factors are considered in the siting process includ-ing
topography and geology, natural resources, socio-cultural aspects, and economy and safety. The system will rank sites on a scale of 0–
100%, with 100% being the most appropriate one. A weighting system is used for all of the considered factors. The results from testing the
system using different sites show the effectiveness of the system in the selection process.
2005 Elsevier Ltd. All rights reserved.

1. Introduction The contamination of groundwater by landfill leachate


has been reported by several researchers (Nixon et al.,
Landfilling has been used for many years as the most 1997; Scultz and Kjeldsen, 1986; Sawney and Kozoloski,
common method for the disposal of solid waste generated 1984). Lee and Jones (1994) reported the potential adverse
by different communities (Komilis et al., 1999). Despite impacts of municipal solid waste (MSW) landfills on those
the intensive efforts that are directed to the recycling and who own or use properties near such facilities. Hirsshfeld
recovery of solid wastes, landfills remain and will re-main et al. (1992) reported that the property values near MSW
an integral part of most solid waste management plans. landfills are adversely impacted by the landfill for dis-
Solid waste disposed in a landfill usually is sub-jected to a tances of a mile or more from the area where waste depo-
series of complex biochemical and physical processes, sition occurs. Adverse environmental impacts, public
which lead to the production of both liquid and gaseous health and socio-economic issues associated with MSW
emissions. As water percolates through the solid waste landfills have led to the issuance of stricter regulations and
matrix, leachate is produced which contains soluble increased public opposition to the siting of such facil-ities
components and degradation products from the refuse. (not in my back yard syndrome) (Ham, 1993). As a result,
Greenhouse gases such as methane and carbon dioxide are the siting of a new landfill has become one of the most
generated during the stabilization of the or-ganic fraction difficult tasks faced by communities involved in MSW
of solid waste. Volatile components of the solid waste tend management (Tchobanoglous et al., 1993).
to be emitted into the atmosphere with the landfill gases. Siting a sanitary landfill requires an extensive evalua-
tion process in order to identify the best available disposal
location. This location must comply with the require-ments
of governmental regulations and at the same time must
* Corresponding author. Tel.: +962 2 7201000; fax: +962 2 7095018. minimize economic, environmental, health, and so-cial
E-mail addresses: aljarrah@just.edu.jo (O. Al-Jarrah), hqdais@ costs (Siddiqui et al., 1996). The site selection proce-dure,
just.edu.jo (H. Abu-Qdais). however, should make maximum use of the

0956-053X/$ - see front matter 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.wasman.2005.01.026
300 O. Al-Jarrah, H. Abu-Qdais / Waste Management 26 (2006) 299–306

available information and ensure that the outcome of the from highway. For example, the distance from highway
process is acceptable by most stakeholders. Therefore, can range from few meters to kilometers and the land slope
landfill siting generally requires processing a variety of can vary across the site from flat to steep. This makes
spatial data. Processing the data by conventional methods fuzzy logic a more natural approach to this prob-lem
using drawing and calculation tools is generally time and (Charnpratheep et al., 1997). Fuzzy logic was intro-duced
resources consuming. The utilization of artificial intelli- by (Zadeh, 1965) to describe and to deal with such
gence technology, such as expert systems, will help in solid uncertainties. In this paper, the power of fuzzy logic and
waste planning and management, particularly in the land- fuzzy inference systems were utilized for preliminary
fill siting process (Thomas et al., 1990). landfill site selection. All of the selection factors are rep-
Ramuu and Kennedy (1994) presented a heuristic resented by linguistic variables. The selection criteria are
algorithm for locating a solid waste disposal site. They weighted to represent the importance and contribution of
indicated that there are a number of criteria that can be each factor.
used to determine the suitability of a potential disposal site. The proposed system encodes the expertise of the envi-
In order to minimize the transportation cost, the ap-proach ronmentalists in a fuzzy inference system that produces the
proposed by Ramuu and Kennedy allowed that the landfill degree of preference of the selected site. It can be eas-ily
could be located along any road linkage. Although used by MSW planners to help in the initial selection
transportation cost is an important consider-ation, other process of landfill site. The system is flexible and can be
factors such as the aesthetics of the solid waste disposal adapted for new information on the landfill site that may be
site also should be taken into account. incorporated into the knowledge base. This will allow for
Kao and Lin (1996) proposed a siting model that was interaction of the user with the computer to help in better
explored for use with raster-based GIS. A mixed integer- understanding the landfill siting process. Several factors
programming model was developed to obtain a landfill site were considered in the siting process. These factors cover
with optimal compactness of the site, which refers to the topography and geology, natural re-sources, socio-cultural
ratio of perimeter to site area. aspects, and economy and safety.
Siddiqui et al. (1996) used GIS and the analytical The present paper is organized as follows. The fol-
hierarchy process (AHP) decision-making procedure to aid lowing section describes the landfill selection factors.
in preliminary site selection. The GIS was used to Section 3 presents an overview of fuzzy sets and fuzzy
manipulate and present spatial data, while the AHP was logic. The fuzzy inference system is presented in section 4,
used to rank potential landfill areas based on a wide variety while Section 5 is devoted for presenting and discuss-ing
of criteria, such as hydro-geology, land use and proximity the simulation results. Finally, the paper is con-cluded in
from urban centers. Section 6.
Valentine (1997) presented a system of a multiple-as-
pect alternative evaluation for the selection of a landfill in
the Urbino region in Italy. The system consists of a matrix 2. Landfill siting factors
that provides environmental categories and im-pact
indicators. He stressed the importance of objective In assessing a site as a possible location for solid waste
methodologies to help in making the best decision in the landfilling, many factors could be considered (Sa-vage et
selection of landfill sites. al., 1998). To achieve a successful siting pro-cess, several
Charnpratheep et al. (1997) explored the prospect of significant environmental and political obstacles have to be
coupling fuzzy set theory with GIS for the preliminary overcome. As a result, convincing the decision-makers and
screening of landfill sites in Thailand. Proximity of geo- gaining public acceptance for the proposed landfill site are
graphic objects, slope and elevation were the criteria used prerequisites for a success-ful siting process. There are
for the investigation. However, the authors recom-mended several factors that should be considered when selecting a
carrying out further studies involving other landfill landfill site. These fac-tors may be presented in many
selection criteria for preliminary screening, such as socio- ways;, however, the most useful way is the one that may be
economic and hydro-geologic criteria. Gupta et al. (2003) easily under-stood by the community (Tchobanoglous et
utilized fuzzy logic, which took into account the al., 1993). The following factors are usually considered in
uncertainty during the process of the environmental the landfill siting process:
impacts assessment of landfill siting and considered the
frequency of impacts occurrence.
It is very difficult to come up with a selection criterion Land slope.
that can precisely describe the preference of one site over Soil characteristics.
another, because most of the selection parameters can-not Depth to groundwater.
be given precisely such as land slope, depth to ground Surface water.
water from landfill base, distance to surface water bodies, Environmentally rare or endangered species breeding
distance to residential area, and distance areas.
O. Al-Jarrah, H. Abu-Qdais / Waste Management 26 (2006) 299–306 301

Distance to residential, religious and archeological Fuzzy sets are sets with imprecise boundaries. A fuzzy
sites. set provides a mechanism to express the degree of mem-
Land use. bership rather than accepting or denying the membership.
Major infrastructure systems (e.g., electrical trans- The wide use and popularity of fuzzy set is related to its
mission lines, gas, water or sewer pipelines). ability to tolerate imprecise and linguistic data. It assigns
Seismic activity. each element in the universe of discourse a value repre-
Land cost. senting its grade of membership in the fuzzy set. This
Distance from highway. number represents the certainty or belief this individual is
Distance from waste generation source. compatible with the concept represented by the fuzzy set.
Site capacity.
Distance from airport runway. A membership function (MF) is a curve that maps each
element in the input space into a membership value called
In addition to the previously mentioned factors, other the degree of membership. The only restriction on the MF
factors may be considered based on local condi-tions and is that it must vary between 0 and 1. The function itself
circumstances. In this study, the factors used for the may take any shape that is defined and specified by the de-
analysis of landfill site suitability were grouped into four signer to suit the nature of the problem from the point of
main categories, including topography and geology, natural view of simplicity, convenience, speed and efficiency.
resources, socio-cultural, and economy and safety. One of the most common classes of MFs is the trian-
Topography and geology factors include land slope (LS) gular MF. A triangular MF can be specified by three
and soil hydraulic conductivity (SHC). Depth to the ground parameters {a, b, c}. It is expressed as
water from landfill base (DGW) and distance to surface 8 0; x 6 a; c 6 x;
water bodies (DSW) comprise the natural resources group.
The socio-cultural effects are measured using the distance lx > x a ; a 6 x 6 b; 1Þ
ðÞ¼< b a ð
: b 6 x 6 c;
to Residential, Reli-gious and Archeological sites (DRRA). > c b;
c x

Four parame-ters are used to characterize the economy and


safety, which are land cost (LC) as a percentage of the where (a < b < c) are the X coordinates of the three cor-
highest price, distance from highway (DH), distance from ners of the underlying MF.
waste generation source (DWGS) measured by the driving Fuzzy sets representation conforms to the objectives
time, and distance from airport runway (DAR). appearing in our daily linguistic usage such as ‘‘small’’,
‘‘medium’’ or ‘‘large’’. These expressions are called lin-
The impact severity of each parameter was deter-mined guistic values and the universe of discourse on which these
as high, moderate, low, and nil. Among the many aspects values are defined on is called a linguistic variable. The
to be considered in the site selection, it is impor-tant to basic operations on a classical set are union, inter-section
consider most of them and with objective weight-ing. The and complement. Corresponding to those basic operations,
degrees of suitability for a specific factor may be different. Zadeh defined similar operations on fuzzy sets (Zadeh,
A scoring system is commonly used to ex-press this 1965).
variability. In the present study, these factors were given Fuzzy IF–THEN rules form the rule base in a fuzzy
weights based on a scoring system ranging from 0 to 1. inference system and they provide a means to encode
These weights were given judgmentally based on a conditional propositions. A single IF–THEN rule takes the
consensus of those professionals with exper-tise in sanitary form
landfill siting and on data from the tech-nical literature.
IF the road is narrow and the rain is heavy THEN . . .
All parts of the antecedent are evaluated simulta-
neously and resolved to a single number using the logi-cal
3. Fuzzy sets and fuzzy logic operators (AND, OR, NOT).
Fuzzy reasoning, known also as approximate reason-
Uncertainty is a major concept in our daily life. Fuzzy ing, is the process of deriving conclusions from a set of IF–
sets and fuzzy logic provide an approach to deal with this THEN fuzzy rules using an inference procedure. By fuzzy
concept. In recent years, fuzzy logic has been successfully reasoning, the truth of the consequent is inferred from the
applied in a variety of disciplines including computer degree of truth of the antecedent.
vision, weather prediction, image processing, nuclear The concept of fuzzy set theory, IF–THEN rules, and
reactor control, control of biomedical processes, automatic fuzzy reasoning together constitute a computing frame-
tuning and many other fields of research (Ya-ger and work usually called fuzzy inference system (FIS). The
Zadeh, 1993; Gzogala and Rawlik, 1989; Al-Jar-rah and structure of a fuzzy inference system consists of three
AL-Rousan, 2001; Al-Jarrah and Halawani, 2001; Al- major parts: a rule base that holds the fuzzy IF–THEN
Jarrah and Bani-Melhem, 2001). rules used in the inference process, a database that
302 O. Al-Jarrah, H. Abu-Qdais / Waste Management 26 (2006) 299–306

contains the membership functions that characterize the The Mamdani model was proposed by Mamdani in
fuzzy sets, and a reasoning mechanism that performs the 1975 to control a steam engine and boiler combination. It
inference procedure and derives conclusions depending on was among the very first control systems built based on the
a set of rules and facts. The fuzzy inference process theory of fuzzy logic (Mamdani and Assilian, 1975). This
consists of five steps including fuzzification, application of model is characterized by representing the consequents of
the fuzzy operators, fuzzy implication, fuzzy aggrega-tion, its rules using fuzzy sets. The aggrega-tion of the outputs
and defuzzification. The last step uses a defuzzifica-tion of all rules yields a single fuzzy set (output). A
method to produce a single crisp number for each output defuzzification process is then performed to extract a crisp
variable (Ros, 1995). value from the output fuzzy set. For example, using the
There are three widely used types of fuzzy inference centroid defuzzification method (Ros, 1995), the crisp
systems: Mamdani, Sugeno, and Tsukamoto (Mamdani and output of the Mamdani fuzzy inference system is given by
Assilian, 1975; Takagi and Sugeno, 1985; Tsukam-oto, R
1979). The basic difference between various models lies in
Def Y lB0 ðyÞy dy
the representation of the consequents of their fuz-zy rules.
Accordingly, the aggregation and defuzzifica-tion
procedures of the three models are different. In this paper, R .
y¼ Y lB0 ðyÞ dy ð2Þ
the Mamdani model will be utilized in devel-oping the
system, as this is the most suitable model when encoding 4. Landfill site selection using fuzzy inference
experts opinion in which their consequences are expressed
as linguistic variables, which is the case in this paper. The proposed system utilizes the power of fuzzy sets in
describing uncertainties in the different factors

W0
LS
SHS W1
DGW
W
DSW 2 Fuzzy Degree of
Fuzzy Defuz.
DRRA Aggregation Suitability
LC Implication W3
DH
W
DWG 4

S
DAR

Fuzzy
Rule Base

Fig. 1. Landfill site selection using fuzzy inference system.

1 Z S M L 1 Z S M L

0.5 0.5

0 0
-14 -10 -5 0 0 20 40 60 80 100
(a) SHC (m/s) (b) LS (%)

Fig. 2. Membership functions on SHC and LS.


O. Al-Jarrah, H. Abu-Qdais / Waste Management 26 (2006) 299–306 303

1 Z S M L 1Z S M L


0.5 0.5

0 0
0 200 400 600 800 1000 1200 0 50 100 150 DGW (m)
(a) DSW (m) (b)

Fig. 3. Membership functions on DSW and DGW.

1 Z S M L 1 Z S M L

0.5 0.5

00 20 40 60 80 100 0 0 1000 2000 3000 4000


(a) LC (%) (b) DRRA (m)

Fig. 4. Membership functions on LC and DRRA.

1 Z S M L 1 Z S M L

0.5 0.5

0 0
0 5 10 15 0 10 20 30 40 50 60
(a) DH (Km) (b) DWGS (min)

Fig. 5. Membership functions on DH and DWGS.

involved in landfill siting. Each factor is represented by a of suitability of a site is expressed as a weighted sum of all
linguistic variable defined over a universe of dis-course factors. These weights are given judgmentally based on a
that represents all permissible values. The expertise of an consensus of those professionals with expertise in sanitary
environmentalist is coded into a fuzzy rule base, which landfill siting and on data from the technical literature. The
drives the inference system as shown in Fig. 1. natural resources are given the highest impact (27%),
topography and geology (25%), and so-cio-cultural effects
The contribution and severity of the various factors in (18%). The impact for economy and safety was chosen
the site selection may be different. A scoring system is such that the land cost is given (14%) and the others
commonly used to express this variability. The degree parameters (16%).
304 O. Al-Jarrah, H. Abu-Qdais / Waste Management 26 (2006) 299–306

If DRRA is Z. then DS is D (0.18)


.
F D C B A .
1 If DRRA is L then DS is A (0.18)

If LC is Z then DS is A (0.14)
.
.
.

0.5 If LC is L then DS is D (0.14)

If DH is Z and DWGS is Z and DAR is Z then DS is B


(0.16) .
.
0 .
0 20 40 60 80 100
If DH is L and DWGS is L and DAR is L then DS is C
DS (%)
(0.16)
Fig. 6. Linguistic variables for the degree of landfill site suitability.
The number between the parentheses represents the
Four linguistic values are defined on each linguistic weight on the rule. The weight is used in the implica-tion
variable, which are zero (Z), small (S), medium (M), and phase to reshape the membership function obtained after
large (L) as shown in Figs. 2–5. The Mamdani mod-el is the application of an implication operator. In this paper, we
used in our system as shown in Fig. 1. Five linguistic have used the minimum operator as the implication method
values are used on the degree of suitability (DS) as shown and the maximum operator for aggregating the outputs
in Fig. 6. These values represent the following grades A, B, fuzzy sets. The final grade is given after defuzzification of
C, D, and F. The rule base consists of 104 rules that the aggre-gated outputs using the centroid method as
describe the various contributions of these factors on the described in Eq. (2).
degree of suitability. According to the scoring system, we
defined five groups of fuzzy rules as follows:

5. Experimental results
If LS is Z and SHC .is Z then DS is A (0.25)
.
. Several experiments were conducted to investigate the
If LS is L and SHC is L then DS is F (0.25) performance of the system. In our experiments, we asked
an expert to analyze different sites for suitability for
possible landfill siting. The relevant information rep-
If DGW is Z and DSW. is Z then DS is F (0.27) resenting the various factors was provided. The recom-
.
. mendation of the expert is then compared with the grade
If DGW is L and DSW is L then DS is A (0.27) given by our system.

Table 1
Landfill siting grades for different sites
Factor Site I Site II Site III Site IV
Topography and geology
Land slope (%) 7.0 11.0 50.0 25.0

Soil hydraulic conductivity (m/s) 10 12 10 6 10 3 10 10


Natural resources
Depth to ground water from landfill base (m) 120 50 8 100
Distance to surface water bodies (m) 1200 800 180 80
Socio-cultural
Distance to residential, religious and archeological sites (m) 3200 2000 450 1000
Economy and safety
Land cost (% of the highest price) 7.0 15.0 60.0 30.0
Distance from highway (km) 1.5 3.0 12.0 5.0
Distance from waste generation source (driving time) (min) 5.0 15.0 35.0 20.0
Distance from airport runway (km) 7.0 4.0 1.0 1.0
Grade (%) 91.19 53.00 23.33 52.09
O. Al-Jarrah, H. Abu-Qdais / Waste Management 26 (2006) 299–306 305

Fig. 7. Location of Al Ghabawi landfill site with respect to the capital city Amman.

In the first experiment, the ability of the system in of a limited number of sites for more detailed evalua-tion.
grading an excellent site was tested. Table 1 shows the The technique should establish criteria to illustrate the
parameters used for this site (Site I) in the experiment and relative suitability of land for sanitary landfilling. In
the corresponding grade. The grade given by the system, addition, the technique should be practical, taking into
which was 91.19%, is consistent with the expert opinion. A account the resources available and should explain clearly
lower quality site was selected in the second experiment and directly the analysis and results in an easily
(Site II). The system gave this site a 53% grade, which was understandable format.
also acceptable in terms of the param-eters of this site as In this paper, an intelligent system approach for select-
shown in Table 1. The third site (Site ing a solid waste landfill site was utilized. The system uses
III) was selected with a very low quality as indicated in a fuzzy inference system that can encode the expertise of
Table 1. The grade given by the system was 23.33%, which an environmentalist and it can adapt by adjusting the
clearly indicates that this site is an unacceptable choice. In knowledge base. Several factors that affect the siting pro-
the last experiment, a site (Site IV) with a mixed set of cess were considered including topography and geology,
parameters in term of acceptance was se-lected as shown in natural resources, socio-cultural aspects, and economy and
Table 1. The system graded this site with 52.09%. The last safety. A scoring system was utilized to represent the
experiment clearly shows the util-ity of the system when severity of these factors. The proposed system ranks sites
presented with conflicting param- using a scale that ranges from 0% to 100%, with 100%
eters. The result for this site is consistent with the expert s representing the optimal choice. Experimental re-sults
preference. To test the ability of the system in grading reveal the effectiveness of the system in terms of its
existing landfill sites, a test was carried out to grade the ranking capabilities. The proposed system can be used as a
newly operated Al Ghabawi landfill site in Amman Jordan tool by planners and decision makers in the process of
(Fig. 7). The site factors were used as in-put to run the initial screening to select a suitable landfill site.
system. The grade given to the site was 81%, which
indicates that the system is capable of grad-ing sites during
the operational phase. References

Al-Jarrah, O., AL-Rousan, M., 2001. Fault detection and accommo-dation


6. Conclusions in dynamic systems using adaptive neuro-fuzzy systems. IEE
Transactions on Control Theory and Applications 148 (4), 238– 290.

An important element of the landfill siting process is a Al-Jarrah, O., Halawani, A., 2001. Recognition of gestures in arabic sign
technique for evaluating the basic suitability of all language using neuro-fuzzy systems. Artificial Intelligence Journal
available land for sanitary landfill as an aid in selection 133, 117–138.
306 O. Al-Jarrah, H. Abu-Qdais / Waste Management 26 (2006) 299–306

Al-Jarrah, O., Bani-Melhem, O., 2001. Building maps for mobile robot Ros, T.G., 1995. Fuzzy Logic with Engineering Applications.
navigation using fuzzy classification of ultrasonic range data. Journal McGraw-Hill, New York.
of Intelligent and Fuzzy Systems 11 (3), 171–184. Savage, G.M., Diaz, L.F., Golueke, G.C., 1998. Guidance for landfilling
Charnpratheep, K., Zhou, Q., Garner, B., 1997. Preliminary landfill site waste in economically developing countries. EPA 600/ R-98-040, US
screening using fuzzy geographical information systems. Waste Environmental Protection Agency, National Risk Management
Management and Research 15, 197–215. Research Laboratory, Cincinnati, OH.
Gupta, R., Kewalramaniz, M.A., Ralegaonkar, R.V., 2003. Environ- Sawney, B.L., Kozoloski, R.P., 1984. Organic pollutants in leach-ates
mental impact analysis using fuzzy relation for landfill siting. Journal from landfill sites. Journal of Environmental Quality 13, 349–353.
of Urban Planning and Development 129 (3), 121–139.
Gzogala, E., Rawlik, T., 1989. Modelling with a fuzzy controller with Scultz, B., Kjeldsen, P., 1986. Screening of organic matter in leachate
application to the control of biomedical processes. Fuzzy Sets and from sanitary landfills using gas chromotography combined with mass
Systems 31, 13–32. spectrophotometry. Water Research 20, 965–972.
Ham, R.K., 1993. Overview and implications of US sanitary landfill
practice. Air and Waste 43, 178–190. Siddiqui, M.Z., Everett, J.W., Vieux, B.E., 1996. Landfill siting using
Hirsshfeld, S., Vesilind, P.A., Pas, E.I., 1992. Assessing the true cost of geographic information systems: a demonstration. Journal of
landfill. Waste Management and Research 10, 471–484. Environmental Engineering 122, 515–523.
Kao, J.J., Lin, H.U., 1996. Multifactorial spatial analysis of landfill siting. Takagi, T., Sugeno, M., 1985. Fuzzy identification of systems and its
Journal of Environmental Engineering 122, 902–908. applications to modeling and control. IEEE Transactions on Systems,
Komilis, D.P., Ham, R.K., Stegmann, R., 1999. The effect of municipal Man, and Cybernetics, 116–132.
solid waste pretreatment on landfill behavior: a literature review. Tchobanoglous, G., Theisen, H., Vigil, S.A., 1993. Integrated Solid Waste
Waste Management and Research 17, 10–19. Management, Engineering Principles and Management Issues.
Lee, G.F., Jones, A., 1994. Impact of municipal and industrial non- McGraw-Hill, NewYork, USA, p. 377.
hazardous waste landfills on public health and the environment: an Thomas, B., Tamblyn, D., Baetz, B., 1990. Expert systems in municipal
overview. Report of California s Environmental Protection Agency solid waste management planning. Journal of Urban Planning and
Comparative Risk Project. Development 116, 150–155.
Mamdani, E.H., Assilian, S., 1975. An experiment in linguistic synthesis Tsukamoto, Y., 1979. An approach to fuzzy reasoning method.
with a fuzzy logic controller. International Journal of Man–Machine Advances in Fuzzy Set Theory and Applications, 137–149.
Studies, 1–13. Valentine, E., 1997. Environmental impact assessment in the site selection
Nixon, W.B., Murphy, R.J., Stessel, R.I., 1997. An emprical approach to process: a case study. In: Sixth International Landfill Congress,
the performance assessment of solid waste landfills. Waste Sardinia, Italy, pp. 223–231.
Management and Research 15, 607–626. Yager, R.R., Zadeh, L.A., 1993. An Introduction to Fuzzy Logic
Ramuu, N., Kennedy, W., 1994. Heuristic algorithm to locate solid waste Applications in Intelligent System. Kluwer Academic Publishers,
disposal site. Journal of Urban Planning and Development 120, 14–21. Dordrecht.
Zadeh, L.A., 1965. Fuzzy sets. Information and Control 8, 338–353.

You might also like