Sie sind auf Seite 1von 9

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/328890262

ESTIMATION OF MUNICIPAL SOLID WASTE GENERATION AND LANDFILL


VOLUME USING ARTIFICIAL NEURAL NETWORKS

Article  in  International Journal of Civil Engineering and Technology · October 2018

CITATIONS READS

0 107

1 author:

Sura Kareem Ali


University of Baghdad
14 PUBLICATIONS   14 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

WATER QUALITY View project

Pavement Maintenance Management View project

All content following this page was uploaded by Sura Kareem Ali on 12 November 2018.

The user has requested enhancement of the downloaded file.


International Journal of Civil Engineering and Technology (IJCIET)
Volume 9, Issue 10, October 2018, pp. 2123–2130, Article ID: IJCIET_09_10_209
Available online at http://www.iaeme.com/ijciet/issues.asp?JType=IJCIET&VType=9&IType=10
ISSN Print: 0976-6308 and ISSN Online: 0976-6316

© IAEME Publication Scopus Indexed

ESTIMATION OF MUNICIPAL SOLID WASTE


GENERATION AND LANDFILL VOLUME
USING ARTIFICIAL NEURAL NETWORKS
Sura K. Ali, Duaa T. Jassim, and Abbas M. Burhan
Civil Engineering Department, College of Engineering, University of Baghdad, Baghdad, Iraq

ABSTRACT
Municipal solid waste is increasing annually, primarily because of the rapid
population growth and lifestyle developments. This results in millions of tons of
municipal solid waste being incorrectly dumped every year. To reduce the
environmental and health risks that can arise from these wastes, they must be contained
and properly disposed of in landfills to ensure a healthy and disease-free environment.
The current study focused on the city of Baghdad, the capital of Iraq, as it is ranks
among Iraq's most densely populated cities. An estimation was done in terms of the
population of Baghdad for the next 20 years and the quantity of municipal solid waste
generated annually was calculated based on the population increase; also, the volume
of municipal solid waste landfill was predicted using artificial neural network. From
the results it was evident that the quantity of municipal solid waste probably generated
by 2035 will be about 5,176,520 tons per year, and the ANN model used to predict the
landfill volume has a coefficient of determination R2 equal to 0.996 with mean square
error of about 3.6*10-5.
Key words: Municipal Solid Waste, Landfill Site, Artificial Neural Network (ANN),
Baghdad city.
Cite this Article Sura K. Ali, Duaa T. Jassim, and Abbas M. Burhan, Estimation of
Municipal Solid Waste Generation and Landfill Volume Using Artificial Neural
Networks, International Journal of Civil Engineering and Technology, 9(10), 2018,
pp. (2123)-(2130).
http://www.iaeme.com/IJCIET/issues.asp?JTypeIJCIET&VType=6&IType=7

1. INTRODUCTION
In the Asian developing countries, a substantial amount of the municipal solid waste is
irresponsibly disposed on the ground as a final dumping method in an uncontrolled manner.
Human and animal activities are chiefly responsible for the generation of municipal solid waste
in the disposal of undesirable or useless materials. The sources from which municipal solid
waste is generated are residential, commercial, industrial, institutional, agricultural,
construction and demolition [1]. Municipal solid waste includes rubbish, garbage, construction

http://www.iaeme.com/IJCIET/index.asp 2123 editor@iaeme.com


Sura K. Ali, Duaa T. Jassim, and Abbas M. Burhan

and demolition debris, street scavenging, sewage waste, trade waste, treated medical solid
waste, etc. [2]. As it is crucial to determine the generation rate for municipal solid waste a
description of its components becomes essential for the planning and design of an effective
management system in any region [3]. The population rise and rapid urbanization results in an
increase in the quantity of municipal solid waste [2; 4]. Among the most important factors
affecting the production of municipal solid waste are the population density, the prevailing
socio-economic conditions, development and urbanization, degree of education of the
individual, average consumption of manufactured materials and people's preferences, height of
manufacturing and industrial production, customs and traditions. Other influential factors
include societal norms, per capita income and the degree of community wealth, where the low-
income areas generate less waste, climate factors, geographical and hydrological characteristics
of the area, collection periods of the waste, waste collection systems, programs and frequency
as the escalation in the collection rates induces a rise in the amount collected annually, and the
presence of household crushing machines for garbage to reduce the food remnants [5].
Generally, the term of kg/capita/day is used to express the rate of generation of municipal solid
waste. For the total solid waste produced, it can be calculated by multiplying the total
population by the generation rate of daily waste per capita [6].
In Kolkata, Das and Bhattacharyya (2014) reported that 32* 105 metric tonnes/year was the
resultant amount of municipal solid waste for Kolkata city, by 2035 [7]. Also, in their 2015
study Chabuk et al., estimated the quantity of solid waste for the Babylon (Babil) governorate,
Iraq and calculated that the total waste by 2030 would be 761,104 tons and 94,305 tons, while
the cumulative amount of solid waste expected until 2030 is 12,494,521 tons and 14,095,437
tons according to the first and second methods, respectively [8]. In their study for the Najaf
governorate, Al-Anbari et al., (2016) estimated the amount of domestic solid waste to be
5,914,415 tons and the required landfill volume to be 11,828,829 m3 [9]. Mustafa and others
performed a study in 2017 cited that by 2030 the resulting municipal solid waste would reach
6427,773 tons for the city of Baghdad [10]. Nafae et al., (2017), found that an area
approximately 1603 m2 will be required for sanitary landfill over the next 20 years for municipal
solid waste disposal for Baghdad city. Appropriate design is essential for the successful
operation of the landfill disposal facility at the most convenient site. The design must include a
landfill with the capacity of receiving the solid waste meant for disposal [11].
Modern methods for predicting include evolutionary programming, fuzzy system, expert
system, artificial neural network and different combination of these tools. The artificial neural
network received a lot of attention because of its obvious model, easy execution and good
performance [12]. The calibration of models based on artificial neural networks is easier than
white box models where fewer parameters are used to develop the model. For this reason,
artificial intelligence techniques using artificial neural networks have recently become very
attractive and popular mathematical tools for both modeling and controlling many complex
environmental processes [13]. A study conducted by Zade and noori, 2008 in Mashhad, Iran
for investigating the different structures of feed forward artificial neural networks models to
Forecast the weight of solid waste generation. They concluded that ANN has more advantages
than conventional methods of forecasting municipal solid waste generation [14]. Also,
Jahandideh et al. (2009) applied artificial neural networks and multiple linear regression models
to predict the generation rate of medical waste generation. The results appeared that artificial
neural networks can predict solid waste generation rates well, while the prediction by multiple
linear regression model was not sufficiently accurate [15]. Arabgol and Ko, 2014 Conducted a
study to develop a model for predicting the quantity of health care waste in Iran. The results
obtained showed that the genetic algorithm has a significant effect on improving primary
weights and improving the performance of the artificial neural network [16]. Azadi and Karimi-
Jashni, 2016 used two models are artificial neural networks and multiple linear regression to

http://www.iaeme.com/IJMET/index.asp 2124 editor@iaeme.com


Estimation of Municipal Solid Waste Generation and Landfill Volume Using Artificial Neural
Networks

predict average generation of municipal solid waste. The result was that artificial neural
networks can be used to predict solid waste in future because multiple linear regression model
showed poor prediction performance [17]. The objectives of this study include the estimation
of the municipal solid waste produced during next 20 years for Baghdad city and using artificial
neural networks to predict the landfill volume based on the population and the amount of
generated solid waste.

2. STUDY AREA
Baghdad is approximately 800 km2 in area and is located between 33°14' to 33°25' N latitudes
and 44°31' to 44°17' E longitudes [18]. Baghdad can be distinguished into two sections, namely
Rusafa and Karkh, due to the Tigris River passing through, as shown in Figure 1. Rusafa is
located on the eastern side of Baghdad, while Karkh is on the western side. Baghdad includes
14 municipalities and 486 localities with a population of nearly 8,258,864 on average. Rusafa
contains the municipalities of New Baghdad, Adhamiya, Al-Ghadeer, Shaab, Sadr 1, Sadr 2,
Rusafa and Karada, while the municipalities of Al-Karkh include Rasheed, Al-Mansour, Karkh,
Kadhimiya, Al-Doura, and Shula. Currently, two sites are available for landfill, viz., Al
Nahrawan extending across around 300 acres are used for the municipalities of Rusafa and Abu
Ghraib a site of nearly 100 acres are utilized for the municipalities of Al Karkh. Both sites are
outside the boundaries of Baghdad city and reveal distinct “open dump” features [19].

Figure 1 Map of the city of Baghdad [20]


3. METHODOLOGY
3.1 Data collection
To compute the average rate of generation of municipal solid waste in Baghdad city the data
were collected from different government references, like the Statistics Division of the
Department of Planning and follow-up, the Municipal Administrative Office and the
Department of Statistics in the Municipality of Baghdad. These data include the quantities of
municipal solid waste at each municipality from 2008 to 2016, as well as their population. For
analysis these data have been classified in Microsoft Office Excel 2016 and IBM SPSS 23
software.

http://www.iaeme.com/IJMET/index.asp 2125 editor@iaeme.com


Sura K. Ali, Duaa T. Jassim, and Abbas M. Burhan

3.2 Estimation the generation rate of municipal solid waste


The mathematical calculation is a super-operation that starts with estimating the future
population based on the current trend and then estimating the amount of solid waste generation
in the future for the countries investigated [1].
= 0∗ 1+ 1
Where:
Pi is the population in predicted year
P0 is the population in present year
r is the growth rate
n: is the number of years [10].
∗ ∗ !
= "!!!
(2)
Where:
MSWGR: the municipal solid waste generated in future in (ton/year) [1; 7].
& '( %) & ) * ,
+
# $ % % = (3)
$ ) + %) & ) * ,
-

In this paper the compacting density of solid waste which was used is equal to 0.45 ton/m3
or 450 kg/m3, and assume the soil cover equal to 15% of the total volume [9].

4. ARTIFICIAL NEURAL NETWORKS (ANNS)


Artificial neural networks are essentially a computer approach consisting mostly of a three-
layer structure, the input layer, the hidden layer, and the output layer where it simulates the way
in which biological neurons receive, transmit, and output signals [21]. Figure 2 illustrates a
typical neural network [22]. Typically, most artificial neural network prediction models have a
clearly defined structure with a specified number of variables entered into the input layer and
the corresponding number of expected outputs in the output layer. based on the problem to be
modeled, many input variables can be selected for a given number of expected outputs. Thus,
the selection of any single model, multi-purpose under any prevailing situations is unrealistic.
Structurally, such a multi-purpose model requires more complicated algorithms capable of
handling multiple calculations at once for some of the required numbers of input variables and
then comes with optimal projections [21]. Artificial neural networks include interconnection
between several neurons. In this work, one type of network was discussed, which is called
multilayer perception (MLP). This network is capable of approximating every measurable
function [23].

Figure 2 A typical neural network

http://www.iaeme.com/IJMET/index.asp 2126 editor@iaeme.com


Estimation of Municipal Solid Waste Generation and Landfill Volume Using Artificial Neural
Networks

5. RESULTS AND DISCUSSION


The population of Baghdad was about 8,258,864 in 2015 with an estimated population growth
rate of 3.5% [19]; Equation 1 was used to estimate the number of Baghdad’s population in the
next 20 years, while Equation 2 was used to estimate the amount of solid waste generated by
the municipality by adopting the average daily generation rate for the individual as 0.875 kg
/capita/day. Table 1 illustrates the results drawn from the two equations cited above, note that
the amount of solid waste generated in the city of Baghdad by 2035 will be 5176520 tonnes,
due to the growth rate of Baghdad's population and consumption of goods and products, as the
increase in consumption is accompanied by an increase in the generation of solid waste.

Table 1 Estimation the population and municipal solid waste of Baghdad city for the period from
(2016- 2035)

Years Population Total municipal solid waste generated (ton/year)


2016 8547924 2692596
2017 8847102 2786837
2018 9156750 2884376
2019 9477236 2985329
2020 9808939 3089816
2021 10152253 3197960
2022 10507581 3309888
2023 10875347 3425734
2024 11255984 3545635
2025 11649943 3669732
2026 12057691 3798173
2027 12479711 3931109
2028 12916500 4068698
2029 13368578 4211102
2030 13836478 4358491
2031 14320755 4511038
2032 14821981 4668924
2033 15340751 4832336
2034 15877677 5001468
2035 16433396 5176520
sum
Artificial neural networks were used to predict the volume78747304
of the landfill, depending on the
population and the amount of solid waste generated as independent variables. The results
showed that the coefficient of determination (R2) was 0.996 as shown in Figure 3 with 3.6*10-
5
as mean square error, this indicates that the model obtained is highly accurate and reliable in
the future to find the required size of the sanitary landfill also, it was found that the population
in the city of Baghdad is the most influential factor as shown in Table 2.

http://www.iaeme.com/IJMET/index.asp 2127 editor@iaeme.com


Sura K. Ali, Duaa T. Jassim, and Abbas M. Burhan

Figure 3 The relation between predicted and observed values

Table 2 Independent Variable Importance

Independent Variable Importance Normalized Importance


Population (capita/year) .878 100.0%
Municipal solid waste (ton/year) .122 13.9%
The final equation was found using artificial neural networks as follows:
01!23! "22
/=
. + "! 2 >>!
"+ 45 !.1""7".1"∗89:; <" 7!.= ∗89:; <0 5!.0∗89:; <
Where:
X1= 0.355 + 1.45*10-10* population – 2.85 *10-11* MSW
X2= -0.0131 + 1.92*10-10* population – 1.74 *10-10* MSW
X3= 0.011 – 1.86*10-10* population +1.258 *10-10* MSW
Population is the population density in (capita/ year)
MSW is the municipal solid waste amounts in (ton/year)
?@ is the predicted landfill volume in (m3)

6. CONCLUSION
1. By 2035, the quantity of municipal solid waste generated in the city of Baghdad will
be 517,652 tons/year.
2. The ANN model for landfill volume has a coefficient of determination R2 equal to
0.996 with mean square error of 3.6*10-5.
3. According to the resulting model, the population of the city of Baghdad is the most
influential factor on the volume of the landfill, that is, the relationship between the
volume of landfill and population density is a direct relationship.

http://www.iaeme.com/IJMET/index.asp 2128 editor@iaeme.com


Estimation of Municipal Solid Waste Generation and Landfill Volume Using Artificial Neural
Networks

REFERENCES
[1] Khajuria, A., Yamamoto, Y. and Morioka T. Estimation of Municipal Solid Waste
Generation and Landfill Area in Asian Developing Countries. Journal of Environmental
Biology, 31,2010, pp. 649-654.
[2] Pamnani, A., and Srinivasarao M. Municipal Solid Waste Management in India: A Review
and Some New Results. International Journal of Civil Engineering and Technology (Ijciet),
5, 2014, pp. 01-08.
[3] Khan, D., Kumar, A. and Samadder R. S. Impact of Socioeconomic Status on Municipal
Solid Waste Generation Rate. Waste Management, 2016, pp. 1-11.
[4] Al-Ansari, N., Pusch, R., and Knutsson S. Suggested Landfill Sites for Hazardous Waste in
Iraq. Natural Science 5, 2013, pp. 463-477
[5] Abdel-Magid, M. I. Solid Waste Engineering and Management, 1st Edition, Khartoum,
2006, pp. 16- 23.
[6] Hadi, S. N. Estimation of Municipal Solid Waste Generation Rate (Case Study of Hilla
City). Journal of Kerbala University, 12, 2014, pp 72-80.
[7] Das, S. and Bhattacharyya, K. B. Estimation of Municipal Solid Waste Generation and
Future Trends in Greater Metropolitan Regions of Kolkata, India. Journal of Industrial
Engineering and Management Innovation, 1, 2014, pp. 31-38,
[8] Chabuk, A., Al-Ansari, N., Hussain, M. H., Knutsson, S. and Pusch, R. Present Status of
Solid Waste Management at Babylon Governorate, Iraq. Scientific Research Publishing, 7,
2015, pp.408-423.
[9] Alanbari, A. M., Al-Ansari, N., Jasim, K. H. and Knutsson, S. Modeling Landfill Suitability
Based on GIS and Multicriteria Decision Analysis: Cas Study in Al-Mahaweel qadaa.
Natural Science, 6, 2014, pp. 828-851.
[10] Mustafa, S. A., Mohsin, A. A. and Ali, N. L. Management of Municipal Solid Waste in
Baghdad, Iraq. International Journal of Environmental and Ecological Engineering, 11,
2017, pp. 700- 704.
[11] Nafae, M. T., Al-Najar, A. H., Al-Tameemi, H. A. N. and Karkoush, N. H. Design of
Municipal Solid Waste Landfill for Baghdad City. World Scientific News, 90, 2017, pp. 1-
10.
[12] Batinic, B., Vukmirovic, S., Vujic, G., Stanisavljevic, N., Ubravin, D. and Goran, Using
ANN Model to Determine Future Waste Characteristic in Order to Achieve Specific Waste
Management Target- Case Study of Serbia. Journal of Science and Industrial Research, 70,
2011, pp. 513-518.
[13] Karpušenkaitė, A., Denafas, G. and Ruzgas, T. Forecasting Hazardous Waste Generation
Using Short Data Sets: Case Study of Lithuania. Environmental Protection Engineering, 8,
2016, pp. 357–364.
[14] Zade, J. G., M. and Noori, R. Prediction of Municipal Solid Waste Generation by Use of
Artificial Neural Network: A Case Study of Mashhad. Int. J. Environ. Res. 2, 2008, pp. 13-
22.
[15] Jahandideh, S., Jahandideh, S., Asadabadi, E. B., Askarian, M. M., Mohammad, M.,
Hosseini, S. and Jahandideh, M. The Use of Artificial Neural Networks and Multiple Linear
Regression to Predict Rate of Medical Waste Generation. Waste Management, 29, 2009, pp.
2874-2879.
[16] Arabgol, S. and Ko, S. H. Application of Artificial Neural Network and Genetic Algorithm
to Healthcare waste Prediction. JAISCR, 3, 2014, pp. 2 43-250.
[17] Azadi, S. and Karimi-Jashni, A. Verifying the Performance of Artificial Neural Network
and Multiple Linear Regression in Predicting the Mean Seasonal Municipal Solid Waste

http://www.iaeme.com/IJMET/index.asp 2129 editor@iaeme.com


Sura K. Ali, Duaa T. Jassim, and Abbas M. Burhan

Generation Rate: A Case Study of Fars Province, Iran. Waste Management 48, 2016, pp.
14–23.
[18] Khudair, H. B. Assessment of Water Quality Index and Water Suitability of The Tigris
River for Drinking Water Within Baghdad City, Iraq. Journal of Engineering, 19, 2013, pp
764-774.
[19] Baghdad Municipality / Municipal Administrative Office Agent / Municipal Waste
Department / Planning and Follow-Up Department, 2017.
[20] Khudair, H. B., Ali, K. S. and Jassim, T. D. Prediction of Municipal Solid Waste
Generation Models Using Artificial Neural Network in Baghdad City, Iraq. Journal of
Engineering, 24, 2018, pp 113-123.
[21] Lomeling, D. and Kenyi, W. S. Forecasting Solid Waste Generation in Juba Town, South
Sudan Using Artificial Neural Networks (ANNs) and Autoregressive Moving Averages
(ARMA). Journal of Environment and Waste Management, 4, 2017, pp. 211-223.
[22] Al-Shayea, Q. and El-Refea, G. Predicting the Effects of Medical Waste in The
Environment Using Artificial Neural Networks: A Case Study. IJCSI International Journal
of Computer Science Issues, 10, 2013, pp. 258-261.
[23] Abdoli, A. M., Nezhad, F. M., Sede, S. R. and Behboudian, S. Longterm Forecasting of
Solid Waste Generation by The Artificial Neural Networks. Environmental Progress &
Sustainable Energy, 2011, pp 1-9.

http://www.iaeme.com/IJMET/index.asp 2130 editor@iaeme.com

View publication stats

Das könnte Ihnen auch gefallen