HTTPS://JOURNALS.
UMS.
AC.
ID/INDEX.
PHP/FG/
ISSN: 0852-0682 | E-ISSN: 2460-3945
Optimising Household Solid Waste Collection Facility in Autonomous Regions of Developing Countries: A Case Study on Karanganyar Regency.
Indonesia Eko Setiawan1,2*.
Juang Victorio Kusuma2.
Ganang Adi Sulistyawan2.
Septin Puji Astuti3 Department of Industrial Engineering.
Faculty of Engineering.
Muhammadiyah University of Surakarta.
Sukoharjo 57162.
Indonesia Pusat Studi Logistik dan Optimisasi Industri (Puslogi.
Muhammadiyah University of Surakarta.
Sukoharjo 57162 Indonesia Department of Environmental Sciences.
Universitas Islam Negeri RM Said Surakarta.
Sukoharjo 57168.
Indonesia Citation:
Setiawan.
Kusuma.
Sulistyawan, , & Astuti.
Optimising Household Solid Waste Collection Facility in Autonomous Re-gions of Developing Countries: A Case Study on Karanganyar Regency.
Indonesia.
Forum Geografi.
, 317-328.
Article history:
Received: 12 February 2024 Revised: 30 Agust 2024 Accepted: 31 Agust 2024 Published: 26 September 2024 Correspondence: eko.
setiawan@ums.
Abstract In many developing countries, autonomous regencies have significant rights to establish and manage waste collection facilities that are critical in conveying waste to final disposal sites.
However, limited budgets often restrict these rights, particularly at the community level, where waste management needs are more acute.
Given the growing importance and emergence of waste-related issues over the years, this study addresses the urgent challenge of siting household solid-waste collection facilities within the context of an autonomous We propose several location-allocation models namely a waste-weighted P-median, a pure P-median, a P-centre, a P-dispersion, and a Audistance gapAy model- to optimise the siting of these facilities.
Utilising data from Karanganyar Regency in Indonesia, we demonstrate that the optimal model for facility siting is contingent on the specific objectives of the initiative, such as minimising transportation costs or maximising service coverage.
Our findings underscore the need for enhanced planning around high-capacity waste collection facilities, emphasising their pivotal role in addressing the future demand for household waste management in autonomous regions.
This study provides a framework for policymakers to make informed decisions about siting waste facilities and promote sustainable waste management practices in resource-constrained environments.
Including more autonomous regions, a variety of scenarios on population growth and waste generation, additional perspectives on waste management, environmental and social considerations, and the investigation of emerging technologies in waste management are suggested as areas for future Keywords: Autonomous region.
Facility siting.
Household solid waste.
Location-allocation model.
Introduction In many developing countries, regencies have a relatively high degree of autonomy, allowing a wide spectrum of rights to govern their jurisdictions.
This includes the authority to establish waste collection facilities from which waste is conveyed to the final disposal sites.
In some areas, rights include the management of waste at the most basic level, that is, waste generated by individual producers (Banerjee & Sarkhel, 2.
However, in other areas, limited budgets mean that waste management at this level has not yet been addressed (Yukalang et al.
, 2.
Household solid wastes are no exception.
In the first scenario, household solid waste is collected and transported to waste collection facilities by a designated agency (Arantes et al.
, 2020.
Brotosusilo et al.
, 2020.
Dan et al.
, 2.
In the second scenario, household waste producers must transport their waste to the collection facilities provided by the authorities (Blanchard et al.
, 2.
The importance and emergence of waste-related issues have grown over time (Kennes & Thalasso, 1998.
Krook et al.
, 2012.
McCunney, 1986.
Wang et al.
, 2016.
Yu & Solvang, 2017.
Yuan & Shen, 2011.
Zaman, 2.
Waste creates a variety of risks for people living in surrounding areas (Finkelman, 2004.
Owusu, 2010.
Ziraba et al.
, 2.
otherwise, it is often perceived as dangerous to neighbouring residents (Litmanen, 1999.
Murdock et al.
, 1.
Copyright: A 2024 by the authors.
Submitted for possible open access publication under the terms and conditions of the Creative Commons Attribution (CC BY) license .
ttps://creativecommons.
org/licenses/by/4.
0/).
Setiawan et al.
Landslide (Defu et al.
, 2.
, disturbance to micro hydropower stations (Mateos et al.
, 2.
, and negative impacts on land resources and environment (Lestari & Trihadiningrum, 2019.
Manzoor & Sharma, 2019.
Vaverkovy, 2.
, to name a few, are examples of serious problems resulting from poor waste management.
Poor management of household solid waste leads to a variety of mishaps (Giusti, 2009.
Laurent et al.
, 2.
These issues are critical in developing countries (Abalansa et al.
, 2021.
Agamuthu, 2013.
Chisholm et al.
, 2021.
Mantzaras et al.
, 2.
Unfortunately, studies on waste management practices in developing countries are rare (Laurent et al.
In response to the presence of waste, one available option is the implementation of waste treatment facilities (Treacy, 2.
The establishment of household solid waste collection facilities can be seen as part of this response.
This is especially important considering the drastically growing Page 317 Forum Geografi, 38.
, 2024.
DOI: 10.
23917/forgeo.
production of household solid waste, a situation that occurs in many places worldwide (UNEP.
People concerned with waste-related problems are already familiar with operations research techniques and methods, as well as multi-criteria decision-making approaches to aid waste management (Banias et al.
, 2010.
Berglund & Kwon, 2014.
Cagliano et al.
, 2014.
Chauhan & Singh.
Erkut et al.
, 2008.
Karak et al.
, 2.
In particular, the use of location models in the context of waste management is abundant, including P-centre model (Maharani, 2.
P-dispersion model (Brylian, 2.
, set covering models (Ghiani et al.
, 2012.
Setiawan et al.
, 2018.
Susy Susanty.
Yuni Triani, 2.
P-median models (Aremu et al.
, 2012.
Putra, 2.
, and a combination of these (Setiawan, 2016.
Setiawan et al.
Presenting a combination of location models applied to a given context of the waste facility siting problem and contrasting the performance of the models has not been found in the literature.
However, regional autonomy is indicative of waste facility siting (Al-Khatib et al.
, 2010.
Brylian, 2018.
Maharani, 2018.
Putra, 2017.
Setiawan et al.
, 2.
This study addresses the problem of placing household solid waste collection facilities in the Karanganyar Regency, an autonomous region located in Central Java.
Republic of Indonesia.
A Pcentre model, two P-median models, a P-dispersion model, and a modified P-centre and P-dispersion model were used in this study.
The siting configurations resulting from each model are subsequently presented and discussed.
We hope that this study provides general insights applicable to similar autonomous regions in any developing country.
The remainder of the paper is organised as follows: Section 2 presents the research methods, consisting of a brief overview of the problem context and the proposed mathematical models for the problem.
The results of the model implementation in the problem context are discussed in Section 3.
The paper concludes with the findings presented in Section 4.
Research Methods Problem context The Karanganyar Regency is autonomous in Central Java.
Indonesia.
Located between 70A28Ao and 70A46Ao south latitude and 110A40Ao and 110A70Ao east longitude (BPS Karanganyar, 2.
, the regency consists of 17 sub-regencies, 162 villages, 15 urban-villages, 1,117 sub-village and 2,323 hamlet (BPS Karanganyar 2.
With a total area of 773.
79 km2, about half the size of Greater London, the regency was expected to be inhabited by 871,596 residents in 2017 (BPS Karanganyar, 2.
In terms of waste management, the Ministry of the Environment Agency in the Karanganyar Regency is responsible for household solid waste.
According to the agency (Kusuma, 2017.
Sulistyawan, 2.
, it manages household solid waste generated by the sub-regencies of Tawangmangu.
Karanganyar.
Tasikmadu.
Jaten.
Colomadu.
Gondangrejo.
Karangpandan and Kebakkramat.
The remaining subregencies are considered capable of handling the household solid waste they produce.
therefore, they require agency operations in their areas.
From the 2016 secondary data obtained by Sulistyawan .
and Kusuma .
, 56 household waste collection facilities existed in the eight sub-regencies of the regency.
Among these 56 facilities, fieldwork carried out by Ifan et al.
and Sulistyawan .
revealed that some facilities did not exist, and 10 facilities were exclusively devoted to certain communities.
Therefore, the facilities chosen as alternatives in the current study were reduced to 36 and referred to as alternatives for household solid waste collection facilities (WCF.
In 2016, the agency managed household solid waste produced by 39 villages and urban villages, as well as the aforementioned 10 community-devoted waste facilities in eight sub-regencies (Kusuma, 2017.
Sulistyawan, 2.
These 39 villages and villages, and 10 community-devoted waste facilities were used as units of household solid waste producers (WP.
in this study.
Table 1 provides data on the WPs for 2016 and the projected year 2026, rounded to two decimal places, while Figure 1 shows a map of the waste volume estimate in 2016 and its projection in Data for the WPs in 2016 were obtained by multiplying the number of inhabitants at each WP by 1.
45 litres of waste produced per person per day.
The 1.
45-liter figure was derived from the ratio of the total waste produced in July 2016 .
easured in m.
to the total population of Karanganyar Regency .
n individual.
for the same month.
Setiawan et al.
Page 318 Forum Geografi, 38.
, 2024.
DOI: 10.
23917/forgeo.
The data for 2026 were obtained by first forecasting the total waste production for 2026 using the total waste data from 2010 to 2016.
The estimated waste production for each WP in 2026 is calculated by multiplying the total projected waste production for 2026 by the population proportion of each WP.
The proportion was determined by dividing the population of each WP in 2016 by the total estimated population for the same year.
For WPs that are WCFs, the population estimates assume that each WCF produces 6.
00 m3 of solid household waste.
Under this assumption, the estimated population was calculated as 6.
00 m3 divided by .
0/1.
individuals per m3, which equals 4,138 individuals.
Table 1.
Waste Volume Estimate in 2016 and Its Projection in 2026.
Alternative Location Sub-regency/ Community Tawangmangu Karanganyar Tasikmadu Jaten Colomadu Gondangrejo Kebakkramat Setiawan et al.
Community in Karangpandan Village/ Urban Village Sepanjang Tawangmangu
Kalisoro
Blumbang
Nglebak
Lalung
Tegalgede Jungke Cangakan Karanganyar
Bejen
Popongan Buran Papahan Ngijo
Gaum
Pandeyan Jati
Jaten
Sroyo
Brujul
Ngasem
Bolon
Malangjiwan Paulan Gajahan Blulukan Gawanan Gedongan Tohudan Baturan Klodran
Wonorejo
Plesungan Selokaton
Dayu
Tuban
Kemiri Nangsri
AURI
RSUD
Garmindo
RSU Jati Husada
Pondok Bukhori Bukit Hermon
Putri Duyung
El Bethel
Rusunawa
Palur Plasa
Total
3,684
8,675
4,056
3,767
4,883
8,014
9,392
5,789
6,447
4,458
10,282
7,514
4,989
7,161
6,969
5,822
4,957
6,915
15,329
9,780
5,963
5,567
6,709
11,755
3,221
2,149
7,282
6,185
8,711
5,877
10,442
5,555
14,314
9,783
9,085
3,073
7,077
9,214
6,318
4,138
4,138
4,138
2016 Waste Volume .
n m.
2026 Waste Volume .
n m.
4,138 4,138 4,138 4,138 4,138 4,138 4,138 318,543 2016 Population Page 319 Forum Geografi, 38.
, 2024.
DOI: 10.
23917/forgeo.
Figure 1.
The Waste Volume Estimate .
in 2016 .
and the Waste Volume Projection .
in 2026 .
in WP alternatives.
Figure 2.
Number of Villages based on Waste Capacity.
Figure 2 summarises the number of villages based on waste volume.
As the population increases, waste volume increases correspondingly, leading to an increase in the number of villages with high waste volumes.
Setiawan et al.
Page 320 Forum Geografi, 38.
, 2024.
DOI: 10.
23917/forgeo.
Table 2 presents the data for the WCFs.
The geographical coordinates of each WCFs are identified using Google Maps.
Along with the geographical coordinates for each WP, these coordinates were used to calculate travel time distances .
enceforth referred to as AudistancesA.
between each WP and each solid waste collection facility (WCF).
The capacity of each alternative WCF, along with the coordinates and AudistancesAy, were obtained from fieldwork carried out by Sulistyawan .
and Kusuma .
Table 2.
Data on WCFs.
WCF
Sub-regency Total
Colomadu
Karanganyar
Jaten
Tasikmadu
Gondangrejo
Tawangmangu
Location Coordinate Fajar Indah Timur
Klodran Utara
Klodran Selatan Tohudan Pilangan Bolon
Klegen
Blulukan Fajar Indah Barat
Ngasem
Sub-district Office of Colomadu
Jungke Jengglong
Pandes Tegalwinangun
Perum WU
Siwaluh
Perum MA
Perum RSS
Edu Park
Bulu
Perum BGI
Jumok
Perum DA
Getas GPI Papahan Papahan Wonorejo
Plesungan Tuban
Grojogan Sewu
Balaikambang
BPTO
Beji Sepanjang Blumbang (-7.
549698,110.
(-7.
536847,110.
(-7.
540157,110.
(-7.
532492,110.
(-7.
538419,110.
(-7.
537489,110.
(-7.
539610,110.
(-7.
538641,110.
(-7.
546547,110.
(-7.
531412,110.
(-7.
531246,110.
(-7.
601020,110.
(-7.
592744,110.
(-7.
590650,110.
(-7.
602244,110.
(-7.
598950,110.
(-7.
598606,110.
(-7.
606175,110.
(-7.
600323,110.
(-7.
588570,110.
(-7.
571835,110.
(-7.
572403,110.
(-7.
587793,110.
(-7.
573431,110.
(-7.
576311,110.
(-7.
573595,110.
(-7.
582809,110.
(-7.
526267,110.
(-7.
527589,110.
(-7.
472977,110.
(-7.
663518,111.
(-7.
662031,111.
(-7.
663247,111.
(-7.
661876,111.
(-7.
673762,111.
(-7.
664157,111.
Capacity .
n m.
Mathematical Models To address the problem, five mathematical models are proposed: a waste-weighted P-median model, a pure P-median model, a P-centre model, a P-dispersion model and a model aimed at minimising the difference between the maximum and minimum AudistancesAy .
eferred as a Audistance gapAy mode.
The full waste-weighted P-median model is expressed as follows:
Objective function:
ycuycnycu Oc Oc ycOycn ycycnyc ycUycnyc ycn .
yc Constraints:
Oc ycUyc O ycE ycUycnyc Oe ycUyc O 0.
OAycn yun ya, yc yun ya .
Ocyc ycUycnyc = 1.
OAycn yun ya .
yc Setiawan et al.
Page 321 Forum Geografi, 38.
, 2024.
DOI: 10.
23917/forgeo.
Ocycn.
cOycn ycUycnyc Oe yayc ycUyc ) O 0.
OAyc yun ya .
ycUyc , ycUycnyc yun .
, .
OAycn ya, yc yun ya .
Objective function .
dictates that the objective of the model is to minimise the total wasteweighted AudistanceAy.
Constraint .
requires that the total number of WCFs to be opened is equal to a certain value.
Constraint .
represents the requirement that a particular WP be served only by an open WCF, whereas constraint .
ensures that exactly one open WCF serves each WP.
Constraint .
requires that the service provided by WCF does not exceed its capacity.
Finally, the decision variables must be binary, as reflected in constraint .
In the meantime, the objective of the pure P-median .
ee constraint .
) is to minimise the total AudistanceAy given the existence of constraints .
Ae .
min Ocycn Ocyc ycycnyc ycUycnyc .
The P-centre model, in contrast, is defined by constraints .
Ae .
and constraints .
, with the objective function appearing in constraint .
The complete model is as follows:
Objective function:
min ycOycoycaycu .
Constraints .
Ae .
Ocyc ycycnyc ycUycnyc Oe ycOycoycaycu O 0.
OAycn yun ya .
The model aims to minimise a maximum AudistanceAy Ae as reflected by objective function .
-, given the existence of constraints .
Ae .
and any possible values for the maximum AudistanceAy .
ee constraints .
In contrast to the abovementioned P-centre model, the proposed P-dispersion model aims to maximise a minimum AudistanceAy Ae as reflected by objective function .
-, given the existence of constraints .
Ae .
and any possible values for the minimum AudistanceAy .
ee constraints .
min ycOycoycnycu .
ycOycoycnycu Oe Ocyc ycycnyc ycUycnyc O 0.
OAycn yun ya .
Finally, the proposed Audistance gapAy model minimises the gap between the maximum AudistanceAy .
ee the P-centre mode.
and the minimum AudistanceAy .
ee the P-dispersion mode.
, as it is represented by objective function .
The model is defined by constraints Equation .
Ae .
min ycOycoycaycu Oe ycOycoycnycu .
Sets:
ya = set of WPs.
ya = set of alternatives for WCFs Parameters:
ycE = maximum number of WCFs required for establishment ycycnyc = AudistanceAy from WP i, i = i, 2, .
I to alternative site for WCFs j, j = 1, 2, .
ycOycn = waste volume of WP i.
yayc = capacity of WCF alternative j, j = 1, 2, .
Decision variables:
ycUyc = { 1, if alternative j is selected as WCF 0, otherwise ycUycnyc = { 1, if WP i is served by WCF alternative j 0, otherwise ycOycoycnycu = total AudistancesAy to be minimised.
ycOycoycaycu = total AudistancesAy to be maximised.
Setiawan et al.
Page 322 Forum Geografi, 38.
, 2024.
DOI: 10.
23917/forgeo.
Results and Discussion Subsequently, all five models were applied to the available data.
In this case, the maximum number of established WCFs is set to 36.
Model implementation was carried out using the Lingo 11 Table 3 summarises the results of model implementation of the data.
The results of the 2016 implementation are presented in Table 4.
Table 5 summarises the results in association with Table 3.
Summary of the Results Indicator Year Waste-weighted P-median 1, 2, 3, 4, 6, 7, 9, 10, 11, 12, 13, 14, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 31, 32, 33, 34, 1, 2.
3, 4, 6, 7, 8,
9, 10, 12, 13, 14,
15, 18, 19, 21, 22,
23, 24, 26, 27, 32,
34, 35
28 facilities 1, 2, 3, 4, 6, 7, 9, 10, 11, 12, 13, 14, 18, 19, 20, 21, 22, 23, 24, 26, 27, 28, 31, 32, 33, 34, 1, 2, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 15, 18, 19, 21, 22, 23, 24, 26, 27, 28, 32, 34, 35 27 facilities 25 facilities 1,328.
00 m3 Selected WCFs Total # of WCF alternatives selected Total capacity of the selected WCFs Unused capacity of the selected WCFs Largest AudistanceAy Smallest AudistanceAy Gap of AudistanceAy Total wasteweighted AudistanceAy Total pure AudistanceAy Total iteration Pure P-median P-centre 1, 2, 3, 4, 5, 6, 7, 9, 11, 12, 13, 15, 17, 18, 19, 21, 23, 24, 25, 28, 29, 34.
P-dispersion 1, 2, 3, 5, 6, 8, 9, 10, 12, 13, 14, 16, 18, 21, 22, 23, 24, 26, 28, 29, 35 AuDistance gapAy 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 16, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 31, 32, 33, 34, 35, 36 23 facilities 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 19, 21, 22, 24, 26, 27, 28, 34, 21 facilities 25 facilities 20 facilities 24 facilities 23 facilities 1,322.
00 m3 1,008.
00 m3 1,293.
00 m3 1,436.
00 m3 1,358.
00 m3 1,388.
00 m3 1,234.
00 m3 1,364.
00 m3 1,360.
00 m3 11 m3 11 m3 11 m3 11 m3 11 m3 23 m3 23 m3 23 m3 23 m3 23 m3 00 min 00 min 00 min 00 min 00 min 00 min 3,028.
61 min 00 min 00 min 00 min 00 min 00 min 00 min 3,032.
41 min 00 min 00 min 00 min 00 min 00 min 00 min 8,475.
06 min 00 min 00 min 00 min 00 min 00 min 00 min 22,546.
90 min 00 min 00 min 00 min 00 min 00 min 00 min 17,065.
80 min 5,110.
58 min 5,127.
57 min 11,627.
00 min 34,302.
20 min 28,508.
80 min 00 min 00 min 00 min 00 min 00 min 00 min 2,456.
00 min 2,520.
00 min 1,798.
00 min 2,019.
00 min 558,002 8,225 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 18, 21, 22, 23, 26, 34, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 15, 16, 21, 22, 23, 24, 26, 27, 28, 34, 35 31 facilities The output of the model implementation (Table .
shows that the volume of waste produced is still manageable.
This was indicated by the total waste volume being less than the total capacity of the selected WCFs.
The implementation also shows that an increasing volume of household solid waste leads to an increase in the total capacity of the selected WCFs and a decrease in their unused capacity.
Additionally, while an increase in household solid waste volume leads to an increase in the total waste-weighted AudistanceAy, it does not necessarily correlate with an increase in the total number of WCFs selected.
Model implementation suggests that the objective of site positioning drives the best model.
Siting facilities to minimise the total waste-weighted AudistanceAy can be achieved by using the wasteweighted P-median model.
The pure P-median model is ideal for achieving a minimum total pure AudistanceAu in waste collection facility siting.
These two models are suitable in situations where waste is collected by a single authoritative body (Al-khatib et al.
, 2007.
Al-Khatib et al.
, 2010.
Henry et al.
, 2.
and, simultaneously, the existence of household solid waste collection facilities is welcomed or perceived as not threatening (Johnson & Scicchitano, 2.
The wasteweighted AudistanceAy model is appropriate when waste production quantity varies significantly across the region (Al-khatib et al.
, 2007.
Al-Khatib et al.
, 2010.
Henry et al.
, 2.
, whereas the pure AudistanceAy model suits regions where waste production quantity is relatively uniform.
Sitting facilities to achieve the minimum largest AudistanceAy and fairness are best approached using the Pcentre model.
In contrast, the P-dispersion model best addresses the objective of obtaining a maxiSetiawan et al.
Page 323 Forum Geografi, 38.
, 2024.
DOI: 10.
23917/forgeo.
mum smallest AudistanceAy with fairness.
These two models fit the circumstances in which inhabitants in surrounding areas must transport their household solid waste to collection facilities (Aliu et al.
, 2014.
Djunaidi et al.
, 2.
The P-centre model is suitable when the presence of household solid waste collection facilities is welcomed by communities or when facility alternatives are distant from residences.
In contrast, the P-dispersion model is appropriate in the presence of NIMBY syndrome .
uch as that described in Johnson et al.
, 2.
or, more generally, when an environmental justice issue exists (Kubanza et al.
, 2.
Siting facilities with the main objective of achieving a relatively equal pure AudistanceAy is best accomplished with the Audistance gapAy model, a combination of the P-centre and P-dispersion models.
In any situation, the site positioning policy should consider the impact of positioning on residents (Zhang & Chen, 2.
Site positioning should be placed within the broader context of waste management to ensure that every stakeholder is in a position of acceptance.
Table 4.
The WPs Served by the WCFs in Each Model Ae Year 2016.
Alternatives for WCF
The WPs served by each of the WCF alternatives resulted from each model
WastePure PP-centre P-dispersion AuDistance weighted Pmedian gapAy 19, 23
23, 34
14, 26, 27, 32
4, 8, 38
9, 25, 38,
40, 46
40, 46
13, 46
5, 25
3, 36
1, 25
4, 10
19, 21
6, 20, 36
6, 22, 36,
27, 34, 42
27, 42
7, 11, 16, 17,
12, 21, 35, 37, 12, 20, 21,
28, 29, 30, 31,
39, 46
24, 37
34, 38, 41, 42
3, 6, 31
3, 6, 31
20, 23, 47
14, 24
14, 30, 32
10, 30
10, 11, 15, 20
10, 11, 15, 20
10, 40
9, 28, 34
5, 44
5, 32
28, 40
8, 17, 28, 37
8, 15, 24, 48
2, 7, 31
11, 23, 31
8, 17, 24, 28,
26, 39
15, 42
12, 21, 32
5, 9, 25, 26, 30,
12, 21, 32
5, 9, 25, 30,
1, 22
19, 40
3, 16, 18, 33, 41
19, 33, 39
2, 18, 28
13, 18, 22, 29,
33, 45
13, 18, 22, 29,
33, 45
6, 9, 18, 33, 45
11, 27, 43
8, 15, 16,
17, 27, 48
For 2016 .
ee Table .
, the model implementation results revealed the following findings.
First, the alternatives selected in the waste-weighted P-median .
and pure P-median .
models were relatively similar, with the only difference being alternative 25.
The allocation of WPs to the selected alternatives was also relatively unchanged.
Second, alternatives 1, 2, 3, 6, 9, 12, 13, 18, 21, 23, 24, 26, 28, 35 are always selected by each model.
Among these, alternatives 1 and 6 seem to be favoured by the models.
Third, alternative 30 was not selected by Setiawan et al.
Page 324 Forum Geografi, 38.
, 2024.
DOI: 10.
23917/forgeo.
any of the five models, making it the least preferred option.
Fourth, alternatives with large capacities .
, 6, 10, 21, 26, 28, and .
served more waste producers in most models.
Alternative 10 is not selected in the P-centre model, possibly due to its AudistancesAy to WPs being too great.
Alternative 1 is always selected by each model despite its relatively small capacity, likely because of its proximity to the WPs.
Table 5.
The WPs Served by the WCFs in Each Model Ae Year 2026.
Alternatives for WCF
The WPs served by each of the WCF alternatives resulted from each model
WasteAuDistance weighted PPure P-median P-centre P-dispersion gapAy 14, 26, 27, 19, 23
23, 34
4, 8, 38
9, 25, 38, 43
40, 46
40, 46
13, 46
5, 25
3, 36
1, 25
4, 10
19, 21
6, 20, 36
6, 22, 36, 42
7, 11, 16, 17,
28, 29, 30,
12, 21, 35, 37, 12, 20, 21, 24,
27, 34, 42
27, 42
31, 34, 38,
39, 46
41, 42
3, 6, 31
3, 6, 31
20, 23, 47
14, 24
14, 30, 32
10, 30
10, 11, 15, 20
10, 11, 15, 20
10, 40
9, 28, 34
5, 44
5, 32
28, 40
8, 17, 28, 37
8, 17, 24, 28, 37
26, 39
8, 15, 24, 48
12, 21, 32
5, 9, 25, 26,
30, 36
12, 21, 32
1, 22
5, 9, 25, 30, 36
13, 18, 22, 29,
33, 45
13, 18, 22, 29,
33, 45
6, 9, 18, 33,
2, 7, 31
15, 42
11, 23, 31
19, 40
19, 33, 39
3, 16, 18, 33, 41
2, 18, 28
8, 15, 16, 17,
27, 48
11, 27, 43
Considering the results of the model implementation for 2026 .
ee Table .
, several insights can be provided.
First, the alternatives selected for the waste-weighted P-median .
and pure P-median .
were relatively the same.
The only difference was between Alternatives 3 and 5.
The allocation of WP to selected alternatives was relatively indifferent.
Second, from all the alternatives, alternatives 1, 2, 4, 6, 7, 8, 9, 10, 12, 13, 21, 22, 24, 26, 34, 35 are always selected by each of the models.
Of these, alternatives 1, 6, 7, 21, and 35 appear to be the favourites in all models.
Third, alternatives 17, 20, 25, 29, 30, 31, 33 and 36 were not selected for the models.
Fourth, alternatives with large capacities .
, 6, 7, 10, 21, 26, 28, and .
served more waste producers in most models.
The total number of WPs allocated to alternative 10 is not as high as that allocated to alternative 21, possibly because its AudistancesAy to WPs are far.
Alternative 1 was always selected by each model despite its relatively small capacity.
The total number of WPs allocated to alternative 7 is not as high as that allocated to alternative 21 even though both of them have the capacity of 50 m3, possibly because the alternative.
Klegen, is relatively AudistantAy to WPs compared to alternative 1, i.
Fajar Indah Timur.
Setiawan et al.
Page 325 Forum Geografi, 38.
, 2024.
DOI: 10.
23917/forgeo.
By contrasting the performance of each model using the 2016 and 2026 data, it is evident that more than 40% .
at least 15 out of .
of the WCF alternatives were selected for both datasets.
Moreover, the siting configuration indicated that alternatives with larger capacities were favoured as the waste volume increased.
Conclusion The analysis highlighted that the optimal model for site positioning in autonomous regions was fundamentally determined by clear waste management objectives.
Regions with autonomy similar to the Karanganyar Regency should prioritise defining their main goals when establishing household solid waste facilities.
Site positioning must be integrated into a broader waste management strategy to ensure stakeholder acceptance and effective implementation.
A comparison of model performance using 2016 and 2026 data suggests that over 40% of waste collection facility (WCF) alternatives are consistently selected across both timeframes.
This trend indicates a preference for larger-capacity facilities as waste volumes increase, underscoring the importance of anticipating future waste management needs.
Thus, regions with similar challenges should focus on developing large-capacity facilities to address future demands efficiently.
This study relies on assumptions about population growth and waste generation, which may not accurately reflect real conditions.
Future research could explore various scenarios to assess their impact on outcomes.
The analysis was specific to Karanganyar Regency and may not be applicable to other regions.
The inclusion of diverse areas can improve the generalisability of the findings.
Acknowledgements This paper is financially supported by the Applied Research Program No.
25/A.
3-i/LPPM/V/2.
from Directorate General of Research and Development Strengthening.
Ministry of Research.
Technology and Higher Education.
Republic of Indonesia.
Author Contributions Conceptualisation: Setiawan.
methodology: Setiawan.
investigation: Kusuma.
Sulistyawan, writingAioriginal draft preparation: Setiawan.
writingAireview and editing: Setiawan.
Astuti.
visualisation: Astuti.
All authors have read and agreed to the published version of the manuscript.
Conflict of interest All authors declare that they have no conflicts of interest.
Data availability.
Data is available upon Request.
Funding This research received no external Setiawan et al.
While stakeholder acceptance is emphasised, this study does not fully consider all perspectives of waste management.
Future studies should use interviews or surveys to examine these views.
Current models focus on logistic factors and overlook both environmental and social factors.
Integrating these factors can offer a more comprehensive approach to siting.
Finally, this study did not consider the technological advancements in waste management.
The investigation of emerging technologies can inform future planning.
References