PRES22_0231.docx


CHEMICAL ENGINEERING TRANSACTIONS 

VOL. 94, 2022 

A publication of 

The Italian Association 
of Chemical Engineering 
Online at www.cetjournal.it 

Guest Editors: Petar S. Varbanov, Yee Van Fan, Jiří J. Klemeš, Sandro Nižetić 
Copyright © 2022, AIDIC Servizi S.r.l. 
ISBN 978-88-95608-93-8; ISSN 2283-9216 

Supply Chain Design and Optimization of the Municipal Solid 
Waste Considering Waste Classification 

Chaoliang Xie a,b, Yinghua Jiang c, Le Wu a,b,* 
aSchool of Chemical Engineering, Northwest University, Xi’an 710069, China 
bShanghai Engineering Research Center of Solid Waste Treatment and Resource Recovery, Shanghai Jiao Tong 
 University, Shanghai 200240, China 
cGuangxi Key Laboratory of Petrochemical Resource Processing and Process Intensification Technology, School of 
 Chemistry and Chemical Engineering, Guangxi University, Nanning 530004, China 
 lewu@nwu.edu.cn 

Waste supply chain is one of the key methods to lower the negative effects of municipal solid waste (MSW) due 
to its larger production. The supply chain of MSW mainly contains four processes: collection, classification, 
transportation and treatment. MSW can be divided into four categories: kitchen waste, recyclable waste, harmful 
waste and other waste. A mathematical model for the design and optimization of the MSW supply chain is built 
to obtain the optimal supply chain network, including the optimal MSW collection and classification site, the 
classified MSW collection site location and its capability as well as the classified MSW transportation way. The 
total annual cost (TAC) is 5.73×108 CNY·y-1, including the total transportation cost is 4.42×1010 CNY·y-1 and the 
total treatment cost is -4.36×1010 CNY·y-1. The transportation cost takes a large proportion of the TAC. The most 
efficient way to lower the TAC is to decrease the transportation cost. The proposed model can be used for the 
waste supply chain design considering the further treatment and utilization of different waste categories. 

1. Introduction
The acceleration of urbanization has led to a sharp increase in municipal solid waste (MSW) production. At 
present, a large amount of MSW is produced every day. If it is not treated in time, it will affect the normal life of 
residents. Therefore, how to effectively deal with MSW has become a major problem that must be solved. Waste 
classification can greatly improve the efficiency of waste treatment and ensure the healthy development of cities. 
The proposal of waste classification provides the ideas for the effective treatment of MSW. Liu et al. (2022) 
studies the factors affecting waste classification and puts forward some suggestions on waste management. 
Residents' awareness of classification is an important factor affecting waste classification and treatment and a 
new fashionable social atmosphere for waste sorting needs to be formed. Li (2021) analyzed the urban garbage 
recycling system and existing problems in China, and gave relevant suggestions for further improvement of 
China's current policies, and suggested to guide third-party companies to integrate the waste sorting industry 
chain, so that waste classification can truly form a complete closed-loop system of the entire industry chain. 
Zhang et al. (2022) found transport link of municipal solid waste is an important part of the waste treatment 
system. In the cost of waste treatment, the cost of the collection and transportation accounts for a considerable 
proportion. So transportation routes can be optimized. Yang et al. (2021) had classified MSW into two 
categories: kitchen waste and recyclable waste. Considering the infrastructure, the supply chain of MSW had 
been redesigned. The specific method of redesign is to change mixed transportation to the exclusive 
transportation. Zhang et al. (2021) had pointed out that a more refined classification can effectively improve 
many aspects of MSW management, especially regarding economic and environmental benefits. Effective MSW 
management can reduce the cost of waste disposal by 69.4% and greenhouse gas and acidic substance 
emissions and increase the energy utilisation rate four fold. 
However, previous studies of the MSW supply chain ignored the delicacy management or treatment of MSW. 
Despite that Yang et al. (2021) divided waste into two categories, which are different from the current waste 
classification. It is urgent to propose a new supply chain model that considers the current waste classification. 

 DOI: 10.3303/CET2294188 

Paper Received: 20  April  2022; Revised: 01  July  2022; Accepted: 05  July  2022 
Please cite this article as: Xie C., Jiang Y., Wu L., 2022, Supply Chain Design and Optimization of the Municipal Solid Waste Considering 
Waste Classification, Chemical Engineering Transactions, 94, 1129-1134  DOI:10.3303/CET2294188 

1129



In this work, on the basis of considering waste classification, the MSW is divided into four categories: kitchen 
waste, recyclable waste, harmful waste and other waste. A mathematical model of waste classification, 
transportation and treatment is proposed by using the existing waste collection sites, transfer centres and 
treatment plants to obtain the optimal supply chain network including the MSW transportation style and MSW 
treatment technology. 

2. Problem statement
The MSW treatment divided into three parts: collection, transportation, and treatment processes. The supply 
chain mainly contains waste collection, waste classification, classified waste transportation from collection sites 
to transfer centers, classified waste transportation from transfer centers to treatment plants. The optimal 
transportation route and style for classified treatment of MSW can be obtained, if the following parameters are 
known: the classified waste supplies around the collection sites, the distances among the collection sites, 
transfer centers and treatment plants, the capacities of transfer centers and unit transportation prices of different 
transportation styles. 

Figure 1: Supply chain of waste classification, transportation and treatment 

3. Mathematical model
3.1 Objective function

The total annual cost (TAC) is chosen as the objective function in the design and optimization of the supply 
chain. It contains the operating cost and the treatment cost. The operating cost is mainly the transportation cost 
of MSW while the treatment cost consists of the MSW treatment cost and the residents’ sanitation cost. The 
calculation formula is shown in Eq(1). 

(1) 
where TAC denotes the total annual cost, in CNY·y-1. CNY is the abbreviation of China Yuan. TOC represents 
the annualized operating cost, in CNY·y-1; TTC is the annualized treatment cost, in CNY·y-1. 
Transportation is mainly considered in calculating the TOC. The transportation cost consists of the transportation 
from collection sites to the transfer centers and classified waste transportation from transfer centers to treatment 
plants. The transportation cost from the collection site to the transfer center is calculated by Eq(2). 

(2) 

where C denotes the cost, in CNY·day-1; A denotes the MSW amount, in t·day-1; d is the distance, in km; p 
represents the prices of different transportation styles, in CNY·t-1·km-1; Superscripts tc and ab are transportation 
cost and collection site to transfer center; Subscripts a and b are the sets for the collection sites, transfer center 
respectively, and te means the transportation style of electric truck. 
The transportation cost from the transfer center to the treatment plant is calculated by Eq(3). 

(3) 

where superscript bc denotes the connection between transfer center and treatment plant; Subscripts th is the 
transportation style of heavy truck. 
The total transportation cost can be obtained by the following Eq(4) and Eq(5). 

1130



 
(4) 

Where tcC  is the total transportation cost, in CNY·day-1; z is the binary variable to determine the existence of 
the transportation route, 0 or 1. 

 
(5) 

where AOT is the annual operating time, in day·y-1. 
The treatment cost consists of the kitchen waste treatment costs, incineration cost, landfill cost, recovery cost 
and resident’ sanitation cost. These costs can be calculated by the following Eq(6). 

 
(6) 

where superscript i is the sets for the treatment way. 
The total treatment cost can be obtained by the following Eq. (7). 

 
(7) 

where superscript kit, inc, lan and rec are the treatment ways of kitchen waste, other waste, recyclable waste 
and harmful waste, respectively; Superscript res denotes resident’ sanitation cost. 

3.2 Mass balance 

The mass balances of the collection sites, transfer centers and treatment plants are shown in Eq(8) ~ Eq(10). 

 
(8) 

 
(9) 

 
(10) 

where S denotes the waste supply, in t; Cap is the capacity, in t; Tre represents the amount of waste treatment, 
in t; Superscript a, b and c are the sets for the collection site, transfer center and treatment plant. 

3.3 Constraints 

Determination of daily waste supply and limitation on the capacity of the transfer center can be calculated by 
Eq. (11) and Eq. (12). 

 
(11) 

 
(12) 

4. Case study 
4.1 Base parameters 

In this work, taking Xi'an, located in middle of China, as an example, there are 12 waste classification collection 
sites (a1 - a12), 8 waste treatment transfer centers (b1 - b8) and 6 waste treatment plants (c1 – c6). The 
proportion of various types MSW are shown in Table 1(Yang et al., 2021). The prices of treating various types 
MSW and transportation are listed in Table 2 (Deppon Express, 2022). The supply of various types MSW in 
each waste collection site is listed in Table 3. The distances among the collection sites, transfer centers and 
treatment plants are presented in Table 4 and Table 5 (Gaode map, 2022). MSW need pass through collection 
sites, transfer centers and before they enter the treatment plants. 

Table 1: Proportion of waste type 

Type Kitchen waste Other waste Recyclable waste Harmful waste 
Proportion/% 52.2 20.7 17.2 9.9 

1131



Table 2: Price of waste treatment and transportation 

 Waste treatment    Transportationa)  
 
Price/CNY·t-1 

Kitchen 
52.2 

Other 
20.7 

Recyclable 
17.2 

Harmful 
9.9 

Electric truck 
3.7 

Heavy truck 
3 

Note: a) The unit for transportation price is CNY·t-1·km-1. 

Table 3: Waste supply of the 12 collection sites 

Supply/t·day-1 Kitchen waste Other waste Recyclable waste Harmful waste 
a1 
a2 
a3 
a4 
a5 
a6 
a7 
a8 
a9 
a10 
a11 
a12 

68,990.1 
77,235.1 
98,376.1 
86,349.2 
48,060.5 
23,161.1 
10,1758.7 
48,389.4 
98,564 
10,1946.6 
46,181.3 
45,387.7 

27,358.2 
30,627.7 
39,011.2 
34,242 
19,058.5 
9,184.6 
40,352.6 
19,189.9 
39,085.8 
40,427.1 
18,313.3 
17,996.6 

22,732.4 
25,449.1 
32,415.1 
28,452.2 
15,836 
7,631.6 
33,529.7 
15,944.4 
32,477 
33,591.6 
15,216.8 
14,953.7 

13,084.3 
14,648 
18,657.6 
16,376.6 
9,115 
4,392.7 
19,299 
9,177.3 
18,693.2 
19,334.7 
8,758.6 
8,607 

Table 4: Distances among 12 collection sites and 8 transfer centers 

Distance/km  Transfer centers 

Collection site 

 
a1 
a2 
a3 
a4 
a5 
a6 
a7 
a8 
a9 
a10 
a11 
a12 

b1 
11.2 
11.7 
20.1 
37.8 
8.3 
36 
8.1 
23.6 
12.2 
14 
12.3 
12.1 

b2 
4.5 
3.7 
11.6 
31.9 
6.5 
30.1 
8 
14.8 
9.2 
4.2 
5.8 
5.6 

b3 
5.2 
4.9 
14.3 
29.4 
8.7 
27.6 
11.4 
17.5 
12.4 
6.3 
8.5 
8.3 

b4 
4.5 
5.3 
15.5 
31.1 
3.3 
29.3 
9.9 
19.4 
11.8 
9.2 
9.4 
9.3 

b5 
3.3 
4.1 
15.4 
29.7 
0.28 
27.9 
11.1 
18.7 
13 
8 
9.7 
9.5 

b6 
14.6 
14 
18.2 
42 
14.2 
40.2 
5.9 
21.4 
9.9 
12.3 
10.2 
10 

b7 
8.9 
9.3 
23.8 
21.9 
11.3 
20 
20.1 
27.1 
22.1 
15.8 
18.1 
17.9 

b8 
6.3 
6.8 
18.4 
29.1 
3.5 
27.3 
14 
21.7 
16.1 
11.1 
14 
12.5 

Table 5: Distances among 8 transfer centers and 6 treatment plants 

Distance/km Treatment plant 

Transfer center 

 
b1 
b2 
b3 
b4 
b5 
b6 
b7 
b8 

c1 
29.3 
16.5 
13.6 
22.3 
20.3 
27.8 
11.5 
21.4 

c2 
46.5 
32.5 
29.9 
40.7 
38.8 
43.3 
30.7 
40.6 

c3 
16.8 
2.7 
3.3 
11 
9.1 
13.5 
11.9 
10.9 

c4 
8.6 
18.8 
21.2 
11.5 
12.8 
12.4 
21.7 
11.8 

c5 
10.2 
6.7 
8 
3.8 
0.6 
13.6 
10.5 
3.9 

c6 
25.1 
14.4 
17.3 
19.9 
18.4 
21.5 
25.2 
21.2 

4.2 Optimal supply chain 

According to the relevant data shown in Table 1 to Table 5, the proposed model is solved in GAMS 24.1.3 with 
solver SCIP. Then the optimal solution is obtained. The optimal waste flowrates among the collection sites, the 
transfer centers and the treatments are presented in Figure 2 and Figure 3. 

1132



a1

a3

a4

a5

a6

a7

a8

a11

a12

b713,084.300

a2 b7

b318,657.600

b716,376.600

b59,115.000

b74,392.700

b419,299.000

b39,177.300

a9 b418,693.200

a10 b319,334.700

b3

b38,607.000

c3

Harmful  Waste

a1

a3

a4

a5

a6

a7

a8

a11

a12

b522,732.400

a2 b5

b432,415.100

b728,452.200

b515,836.000

b77,631.600

b433,529.700

b515,944.400

a9 b432,477.000

a10 b533,591.600

b4

b414,953.700

c5

Recovery  Waste

a1

a3

a4

a5

a6

a7

a8

a11

a12

b727,358.200

a2 b7

b339,011.200

b734,242.000

b519,058.500

b79,184.600

b640,352.600

b319,188.900

a9 b339,085.800

a10 b3

b3

b317,996.600

c2

Other  Waste

Unit: t/day

a1

a3

a4

a5

a6

a7

a8

a11

a12

b4

b5

a2
b4

b5

b198,376.100

b786,349.200

b548,060.500

b723,161.100

b6101,758.700

b248,389.400

a9
b2

b4

a10
b2

b3

b5

b5

46,181.300

c3

Kitchen Waste

 

Figure 2: The optimal supply chain network of kitchen waste, other waste, recovery waste and harmful waste (a 
denotes collection site; b is transfer center; c represents treatment plant.) 

According to Figure 2, there are 12 collection sites, 6 transfer centers and 4 treatment plants. The transportation 
styles from the collection sites to the transfer centers and the transfer centers to the treatment plants are electric 
truck and heavy truck, respectively. It is worth noting that there are 8 transfer stations and 6 treatment plants in 
the original plan. However, when the maximum capacity of waste transfer center is 300,000.015t and the waste 
treatment plant is unlimited, only 6 transfer centers and 4 treatment plants are selected. 

b2 i1300,000.015 b3

i1

i2

i4

174,022.900 b4

i1

i3

i4

128,592.300

b5

i1

i2

i3

i4

b6

i1

i2

b7

i1

i2

i3

i4

300,000.015 300,000.015 279,995.355

300,000.015
142,111.300

295,508.200

  

Figure 3: Amount and type of MSW contained in each transfer center (i denotes the waste type) 

The specific flowrates of various types MSW in 6 waste transfer centers are shown in Figure 3. The b2, b3, b4, 
b5 and b7 are close or even reach to their maximum capacity. In b5 and b7, there are all kinds of MSW, so in 
order to improve the efficiency and total amount of waste treatment, it can be considered to expand their scale. 

1133



Table 6: The details of the total annual cost 

Cost 
Transportation from 
collection site to 
transfer center 

Transportation from 
transfer center to 
treatment plant 

Recyclable waste TAC 

Value/CNY·y-1 2.17×1010 2.25×1010 -4.36×1010 5.73×108 

 
According to Table 6, the transportation cost from the collection sites to the transfer centers is 2.17×1010 CNY·y-
1, and transportation cost from the transfer centers to the treatment plants is 2.25×1010 CNY·y-1. The treatment 
cost is -4.36×1010 CNY·y-1. That is to say, there are economic benefits in waste treatment considering 
classification, which is one of the reasons why waste classification is necessary. The TAC is 5.73×108 CNY·y-1. 
The supply chain proposed in this work makes full use of the existing infrastructure and further improves the 
waste treatment effect through more detailed division of waste. Through intuitive data, it is proved that our 
conclusion is the same as Yang et al. (2021)’ conclusion that waste classification improves the treatment effect 
of MSW. Furthermore, more classifications of MSW can enhance the economic performance of the system 
according to the comparison with Yang et al. (2021)’ work. 

5. Conclusion 
A mixed integer programming is proposed to optimize the supply chain of MSW transportation and treatment by 
considering the waste classification. The supply chain can be optimized by solving the proposed model. After 
the waste is classified for treatment, the waste treatment effect is apparently improved. The results show that it 
has benefits in the link of waste classification and treatment. Considering the higher MSW transportation cost 
and relatively lower investments of collection site and transfer center, the most effective way to lower the total 
cost of the supply chain is to build the new waste collection sites and transfer centers in appropriate locations 
based on the waste classifications in order to decrease the distances between transportation routes. Future 
research should focus on the impact of seasonal changes and the uncertainty of supply of various waste 
collection. 

Acknowledgments 

The authors gratefully acknowledge funding by the project (No. 2021GFZX002) sponsored by Open Research 
Fund of Shanghai Engineering Research Center of Solid Waste Treatment and Resource Recovery, the project 
(CXY-2021-130) sponsored by Production and Research Project of Yulin and the project (2021K007) sponsored 
by Opening Project of Guangxi Key Laboratory of Petrochemical Resource Processing and Process 
Intensification Technology. 

References 

Deppon Express, 2022, Transportation price, Deppon Express <deppon.com/newwebsite/mail/price> accessed 
10.03.2022. 

Gaode Map, 2022, Distance, Gaode Map <amap.com> accessed 10.03.22. 
Li K., 2021, Analysis and suggestions on classification and recycling of urban domestic waste. IOP Conference 

Series Earth and Environmental Science, 647, 012177. 
Liu Q., Xu Q., Shen X., Chen B., Esfahani SS., 2022, The mechanism of household waste sorting behaviour—

A study of Jiaxing, China. International Journal of Environmental Research and Public Health, 19(4), 2447. 
Yang X., Guo X., Yang K., 2021. Redesigning the municipal solid waste supply chain considering the classified 

collection and disposal: a case study of incinerable waste in Beijing. Sustainability, 13(17), 9855. 
Zhang, Y., Luo, X., Han, X., Lu, Y., Wei, J., Yu, C., 2022, Optimization of urban waste transportation route based 

on genetic algorithm. Security and Communication Networks, 2022. 
Zhang J., Zhang Z., Zhang J., Fan G., Wu D., 2021, A quantitative study on the benefit of various waste 

classifications. Advances in Civil Engineering, 2021(2), 1-15. 

1134


	PRES22_0391.pdf
	Supply Chain Design and Optimization of the Municipal Solid Waste Considering Waste Classification