Study on the Relationship between Different Wastewater Treatment Technologies and Effluent Standards in Jilin Liaohe River Basin Based on the Coupled Model of AHP and Fuzzy TOPSIS Method

Abstract

Wastewater treatment plants play a critical role in reducing point source pollution in watersheds; however, in taking on the task of reducing pollutants such as COD and NH₃-N, they also consume energy intensively, which can result in additional greenhouse gas (GHG) emissions and operating (OAM) costs. Therefore, appropriate effluent targets should be implemented to achieve a balance between pollution load reduction, greenhouse gas emissions and operating costs, depending on the field conditions of the wastewater plant. In this study, four different wastewater treatment technologies, namely, A²/O, CASS, MBR and A/O-MBR, which are mainly operated in the Liaohe River Basin of Jilin Province, were selected using the coupled AHP and fuzzy TOPSIS models, and the appropriate effluent standards were preferred according to the characteristics of these technologies. Firstly, the AHP model is used to determine the criteria layer (environmental, economic and social benefits) of the four treatment technologies and the weights of each indicator, and then the fuzzy TOPSIS model is used to determine the indicator values of the different alternatives (effluent standards) of the four treatment technologies, and finally the optimal value of the effluent standard is determined, in terms of COD, 30 mg/L for A²/O and CASS process and 50 mg/L for MBR and A/O-MBR.

1. Introduction

1.1. Background

Wastewater treatment plants are not only the main bearers of the task of reducing COD and NH₃-N pollution load, but also one of the most energy-intensive and high-energy industries [1], Although a higher effluent standard will effectively reduce the pollution load, it will also consume more energy and require higher operation and maintenance (OAM) costs, and produce more indirect and direct carbon emissions than a lower effluent standard, so the pursuit of a higher effluent standard will have a negative effect on carbon reduction and economic efficiency [2,3].

As the economies and populations of countries continue to develop, climate change and environmental damage are becoming increasingly prominent, and there is now a global consensus on low-carbon, green and sustainable development. From the Kyoto Protocol in 1997 to the Paris Agreement in 2015, the number of parties has grown to 178, and the Paris Agreement for the first time introduced a “carbon-neutral” emissions reduction target. To meet the challenges that come with it, nearly 200 countries are discussing how to uphold the Paris Agreement during the UN Climate Change Conference in 2021. Under the Paris Agreement, the European Council approved a binding target to reduce economy-wide greenhouse gas emissions by at least 40% by 2030 compared with 1990 (Council Decision (Eu) 2016/1841). In September 2020, China solemnly announced to the world its “double carbon” target, i.e., to reach peak CO₂ emissions by 2030 and work towards carbon neutrality by 2060.

The main cause of climate change is influenced by the large increase in emissions of greenhouse gases, which refer to gases in the atmosphere that have the ability to absorb energy from atmospheric long-wave radiation, and in global warming, CO₂ contributes 49%, CH₄ contributes 15–20%, and N₂O contributes 6%, which is about twice as much as other greenhouse gases [4]. The GHG emissions from the operation of wastewater treatment plants, on the other hand, are dominated by these three gases.

The total GHG emissions from wastewater treatment plants are not significantly influenced by the mode of wastewater treatment [5]. The Greenhouse Gas Accounting System (GHG Protocol), jointly established by the environmental NGO World Resources Institute (WRI) and the World Business Council for Sustainable Development (WBCSD) since 1998, stipulates that the total emissions of an enterprise are divided into direct and indirect emissions. Direct emissions are called Scope 1 and indirect emissions are divided into Scope 2 and Scope 3, where Scope 1 is emissions from sources directly controlled or owned by the company, Scope 2 is indirect emissions from electricity, steam, heating and cooling used by the company, and Scope 3 is all indirect emissions except Scope 2. The greenhouse gases generated from the biochemical wastewater treatment link is an important source of direct greenhouse gas emissions [6,7], and is a direct carbon emission, range 1, in which the removal rate of COD is positively correlated with the emission of CO₂ [8]. Indirect carbon emissions in Scope 2 are emissions from energy consumption, including energy consumption from maintaining the operation of wastewater plants, and energy consumption is an important factor affecting the carbon emissions of urban wastewater treatment systems, of which electricity accounts for about 55% of the total energy [9,10,11,12]. Scope 3 is the energy consumption generated from the treatment of activated sludge, which is a by-product of wastewater treatment [13,14,15], sludge contains a variety of toxic substances that can cause serious environmental pollution [16,17,18], and sludge disposal and recycling are of great concern because of its serious negative ecological impact [18,19], and studies have shown that activated sludge production and COD removal show a positive correlation [20,21,22]. Therefore, the effluent standard of wastewater plants should be based on the actual situation and a balanced analysis of the economic and social benefits of the operation process, so as to obtain a suitable effluent standard, and at present, there is a great potential to reduce the operating costs of wastewater plants through technical and economic methods [23].

1.2. Development of Coupled Multi-Criteria Decision Models (MCDM)

In recent years, coupled multicriteria decision models have received increasing attention, and after Churhman et al. first used the simple weighting method to solve the problem of choosing a business investment strategy, MCDM methods have been continuously developed and expanded in various fields such as economics, management, military, and environmental science, and are now maturing [24].The study of MCDM models has basically used two methods: the AHP and TOPSIS method, and with the development of the times, MCDM has chosen to incorporate more different models and methods to solve decision problems.

Bognár F. et al. [25] combined AHP–PRISM with the TOPSIS method and used it to test it on data related to strategic event sets for incoming logistics business processes at a nuclear power plant, and the results demonstrated that the novel method supports ideal best and worst analysis of alternatives without losing its ability to focus on identifying hidden risks. Mathew, M. et al. [26] adopted a new IT2F multi-criteria decision-making approach based on AHP and TOPSIS that can effectively handle the degree of uncertainty in group decision making and assist in selecting the most feasible and optimal industrial asset maintenance strategy. The results demonstrate the additional advantage of IT2F–AHP–TOPSIS over the traditional AHP–TOPSIS and type 1 fuzzy AHP in that it allows decision makers to define membership functions more flexibly and is able to handle uncertainty in the decision-making process. Du, Y.M. et al. [27] applied fuzzy AHP and AHP–TOPSIS to determine damage assessment levels for 18 soil sites selected from Qinghai Province. The results showed that AHP–TOPSIS was closer to the real damage situation than the single fuzzy AHP method, and the method proved to be more suitable for damage assessment of earthen sites in Qinghai Province. Liu B. et al. [28] combined FST, AHP and TOPSIS to form the MCDM model to optimize wastewater treatment technologies in urban areas of the Liaohe River Basin, and later, Liu B. et al. [29] combined GIS, AHP and remote sensing (RS) technologies to perform site suitability analysis of wastewater treatment plants and applied them for the first time in the Liaohe River Basin of Jilin Province, China. Dai C. et al. [30] verified the rationality of the indicator system of the AHP–TOPSIS model by taking the evaluation of water resources development and utilization level in Jingyu County from 2009 to 2018 as an example, and filtered out the indicators that have a greater impact on the evaluation results by changing the sensitivity analysis method of indicator weights to reduce the uncertainty of too many indicators and low correlation.

Assessing suitable water discharge criteria involves a large number of indicators, and the fuzzy TOPSIS method in the MDCM model, as an important method for multi-attribute decision making, has been applied by more scholars to decision making in the environmental field, and has achieved good research results [31,32,33]. Due to the complexity of the evaluation index system in this study, it is more difficult to solve the index weights, and the TOPSIS method can only rank the water discharge criteria scheme, but not the index weights. To solve the indicator weight ranking problem, this paper introduces the AHP method coupled with the fuzzy TOPSIS method to form the AHP–TOPSIS coupling model, and uses the AHP method to solve the weights and rank the indicators according to their weight sizes. The method can, to a certain extent, solve the problem that the weights are not reasonable enough due to the lack of subjective judgment ability of evaluation experts or differences in experience.

1.3. Main Contributions of This Paper

Firstly, in this study, the authors coupled AHP with the fuzzy TOPSIS method to constitute a scientific and effective decision model. After that, considering the economic, social and environmental needs of the study area, the coupled model was run to calculate the preferred effluent standards for all wastewater plants in the area, which provides a decision basis and reference for the local environmental protection authorities.

1.4. The Next Main Structure of This Article

The rest of the paper is organized as follows. Section 2, Materials and Methods, presents the arithmetic process of the AHP–TOPSIS model in this paper. In Section 3, a case study is presented, starting with an overview of the study area and a complete description of the operation of the coupled model. Section 4 concludes the paper with a summary of the study.

2. Materials and Methods

2.1. AHP and Fuzzy TOPSIS Coupling Model

In this study, the coupled AHP and fuzzy TOPSIS model was used to analyze the decision of suitable effluent indexes of each wastewater plant in the study area. The AHP method was selected to determine the weights of each index, and the fuzzy TOPSIS method was used to quantitatively analyze the distance between each alternative and the optimal solution. Firstly, the most suitable water discharge index is divided into the criterion layer and the indicator layer using the AHP method, and the weights of each layer are found out, then combined with fuzzy TOPSIS method, the weighted standardized decision matrix and comprehensive superiority of each index under the criterion layer are found out, and then the optimal ranking of each alternative is derived. The model operation and calculation process are shown in Figure 1.

Figure 1. Comprehensive evaluation model of the most suitable water discharge criteria.

2.1.1. AHP Calculation Procedure

Analytic Hierarchy Process (AHP) is a hierarchical weighting decision analysis method proposed by American operations researcher Saaty in the early 1970s when he applied the network system theory and multi-objective comprehensive evaluation methods to the study of “power allocation based on the contribution of each industrial sector to national welfare” for the U.S. Department of Defense [34,35].

AHP is a combined qualitative and quantitative decision analysis method for solving complex problems with multiple objectives. The method combines quantitative analysis with qualitative analysis, uses the experience of decision makers to judge the relative importance between the criteria that measure whether the objectives can be achieved among them, and reasonably gives the weight of each criterion for each decision option, and uses the weights to find out the order of advantages and disadvantages of each option, which is more effectively applied to those topics that are difficult to be solved by quantitative methods, and this subject uses post-AHP to obtain the weights of each The weights of the indicators, in this study AHP is calculated in the following steps.

(1)

Build the hierarchical model

A three-level criterion decision model is built according to the relationship between decision objective, criterion and alternative, where the decision objective is the highest level, the criterion is placed in the second level as an intermediate element, and the alternative is placed in the third level.

(2)

Construction of judgment matrix

The method of constructing judgment matrix in AHP is the consistent matrix method, according to the opinions of different experts under the scenario of the research content, constructing a two-by-two comparison matrix, and the comparison matrix after establishment is as in Equation (1), where $a_{ij}$ is the result of comparison between element i and j, and the specific rules of taking values are as in Table 1.

$${\left[a_{ij}\right]}_{n\times n}=\begin{array}{ccccc}& I_1& I_2& \cdots & I_n\\ I_1& 1& a_{12}& \cdots & a_{1n}\\ I_2& 1/a_{12}& 1& \cdots & a_{2n}\\ ⋮& ⋮& ⋮& \ddots & ⋮\\ I_n& 1/a_{1n}& 1/a_{2n}& \cdots & 1\end{array}$$

(1)

Table 1. Significance of judgment matrix scale.

Scale	Meaning
1	Two elements are of equal importance
3	Two elements are slightly more important than one another
5	Two elements are more important than one another
7	Two elements are very important compared to one another
9	Two elements are absolutely more important than one another
2, 4, 6, 8	Adjacent intermediate numbers
reciprocal	If the ratio of the importance of element i to j is $a_{ij}$ then the ratio of the importance of j to i is 1 $/a_{ij}$

(3)

Hierarchical single ranking and its consistency test

The feature vector corresponding to the maximum feature root λ_max of the judgment matrix is normalized (so that the sum of the elements in the vector is 1) and then recorded as w. The elements of w are the ranking weights of the same hierarchical element with respect to the relative importance of a factor in the upper level, and this process is called hierarchical single ranking. The consistency metric can be defined as $C_I=\frac{\lambda -n}{n-1}$, where $C_I=0$ indicates perfect agreement; additionally, the closer the $C_I$ value is to 0 the better the consistency is, and the further it is from 0 the worse the consistency is. In order to evaluate $C_I$, AHP introduces the random consistency index $R_I$, defining the consistency ratio: $CR=\frac{C_R}{C_I}$, when the value of CR does not exceed 0.1, the consistency of the matrix is considered to be in the allowed interval and passes the consistency test, otherwise the value of $a_{ij}$ needs to be adjusted, and the value of $R_I$ is taken as shown in Table 2.

Table 2. Random consistency index RI values.

n	1	2	3	4	5	6	7	8	9
RI	0.00	0.00	0.58	0.90	1.12	1.24	1.32	1.41	1.45

2.1.2. Calculation Process of Fuzzy TOPSIS Method

TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) method was first proposed by C.L. Hwang and K. Yoon in 1981 and is also known as the superior-inferior solution distance method [36]. This method is a multi-attribute decision making method that evaluates and ranks alternatives based on proximity to the ideal solution and distance from the negative ideal solution by calculating the proximity between several alternatives and the ideal solution. However, the actual problem is continuously complicated, and many times the attribute values of the alternatives cannot be obtained, and the traditional TOPSIS can no longer meet the needs of solving the decision problem, so the triangular fuzzy numbers are introduced into the role model of this study to solve the complex decision problem [37].

1.

Construct a decision matrix

First, the standard decision matrix is constructed, and the matrix $\tilde{A}$ is composed of alternatives AO = {AO₁, AO_2, …, AO_n} and indicators I = {I₁, I_2, …, I_m}, where AO_n is the fuzzy triangular number, AO_n = {${\tilde{a}}_{i1},{\tilde{a}}_{i2}$…,${\tilde{a}}_{in}$}.

$$\tilde{A}=\left[\begin{array}{ccc}\begin{array}{c}{\tilde{a}}_{11}\\ {\tilde{a}}_{12}\end{array} \begin{array}{c}{\tilde{a}}_{21}\\ {\tilde{a}}_{22}\end{array}& \cdots & \begin{array}{c}{\tilde{a}}_{m1}\\ {\tilde{a}}_{m2}\end{array}\\ ⋮& \ddots & ⋮\\ \begin{array}{cc}{\tilde{a}}_{1n}& {\tilde{a}}_{2n}\end{array}& \cdots & {\tilde{a}}_{mn}\end{array}\right]$$

where ${\tilde{a}}_{mn}$ is the triangular fuzzy number of the m scheme at the n index, denoted as ${\tilde{a}}_{mn}=[a_{mn}^L,a_{mn}^U,a_{mn}^V$].

2.

Calculate the weighted normalization matrix

According to the weights w_j of each index determined by the previous AHP, the above decision matrix is weighted and the corresponding matrix is established, and the established matrix is expressed as:

$$\tilde{Z}=\{w_i\times {\tilde{a}}_{mn}\}=\{w_i\times a_{mn}^L,w_j\times a_{mn}^U,w_j\times a_{mn}^V\}$$

3.

Determine the fuzzy PIS and NIS of each index

The optimal value vector ${\tilde{Z}}^{+}$ and the worst value vector ${\tilde{Z}}^{-}$ are obtained according to the $\tilde{Z}$ array, i.e., the optimal and worst solutions in a finite scheme are:

$${\tilde{Z}}^{+}=\left({\tilde{z}}_1^{+},z_2^{+},\cdots ,{\tilde{z}}_n^{+}\right), {\tilde{z}}_n^{+}=\left(z_{mn}^{L+},z_{mn}^{U+},z_{mn}^{V+}\right)$$

$${\tilde{Z}}^{-}=\left({\tilde{z}}_1^{-},z_2^{-},\cdots ,{\tilde{z}}_n^{-}\right),{\tilde{z}}_n^{-}=\left(z_{mn}^{L-},z_{mn}^{U-},z_{mn}^{V-}\right)$$

Among them, $z_{mn}^{L+}= \max \left\{z_{mn}^L\right\}$,$z_{mn}^{U+}= \max \left\{z_{mn}^U\right\}$,$z_{mn}^{V+}= \max \left\{z_{mn}^V\right\}$

$$z_{mn}^{L-}=\max \left\{z_{mn}^L\right\},z_{mn}^{U-}=\max \left\{z_{mn}^U\right\},z_{mn}^{V-}= \max \left\{z_{mn}^V\right\}$$

4.

Calculate the distance between the evaluation object and the optimal and inferior solutions

Based on the previously obtained optimal vector ${\tilde{Z}}^{+}$ and the worst vector ${\tilde{Z}}^{-}$, the distances $D_n^{+}$ and $D_n^{-}$ are calculated between them and each indicator of all alternatives.

$$D_n^{+}=\sqrt{\frac{1}{3}\times \left[{\left(z_{mn}^L-z_n^{L+}\right)}^2+{\left(z_{mn}^U-z_n^{U+}\right)}^2+{\left(z_{mn}^V-z_n^{V+}\right)}^2\right]}$$

$$D_n^{-}=\sqrt{\frac{1}{3}\times \left[{\left(z_{mn}^L-z_n^{L-}\right)}^2+{\left(z_{mn}^U-z_n^{U-}\right)}^2+{\left(z_{mn}^V-z_n^{V-}\right)}^2\right]}$$

where $D_n^{+}$ and $D_n^{-}$ denote the distance of the n evaluation object from the optimal solution and the worst solution, respectively.

5.

Calculate the proximity of each evaluation object to the optimal solution

$$C_n=\frac{D_n^{-}}{D_n^{+}+{\mathrm{D}}_n^{-}}$$

$C_n$ takes a value between 0 and 1. the value of $C_n$ close to 1 means that the evaluation object is closer to the optimal level; conversely, the closer to 0 means that the evaluation object is closer to the worst level.

6.

Rank each evaluation object according to the degree of proximity and determine the evaluation effect.

The evaluation objects are ranked according to the size of $C_n$, and the larger the value of $C_n$, the better the overall benefit.

2.2. GHG Generation Calculation

In this study, the amount of activated sludge generated, GHG emissions and sludge treatment cost from the treatment of activated sludge, electricity consumption, electricity cost and CO₂ emissions from the consumption of electricity are calculated for wastewater treatment plants according to different emission targets, where the amount of sludge generated and sludge treatment cost are calculated with reference to the empirical data obtained during field research, and the electricity cost is calculated according to the existing electricity tariff in Jilin Province. The electricity cost is calculated according to the existing electricity price in Jilin Province. Some of the formulas are as follows.

(1)

Amount of methane generated from COD removal

$$E_1=[(R_{COD}-SG\times {\rho }_S)·E{F}_{C{H}_4}]\times GW{P}_{C{H}_4}$$

E₁—Annual emissions of CH₄ converted to CO₂ equivalent from COD removal in wastewater, tCO₂eq/a

${\mathrm{R}}_{\mathrm{COD}}$—Annual COD removal from wastewater treatment plant, tCOD/a

SG—Annual generation of dry matter of sludge from wastewater treatment plant, t/a

${\mathsf{\rho }}_{\mathrm{S}}$—Organic matter content in dry matter of sewage treatment plant, tCOD/a

${\mathrm{EF}}_{{\mathrm{CH}}_4}$—global warming potential value, take 21

Among them ${\mathrm{EF}}_{{\mathrm{CH}}_4}={ \mathrm{B}}_0\times \mathrm{MCF}$

MCF—CH₄ correction factor, take 1

B0—Maximum CH₄ GWP value, taken as 0.25t CH4/t COD

(2)

GHG emissions from sludge treatment

$$E_2=SR\times {\beta }_S{}_{COD}\times DO{C}_f\times MCF\times F\times C_{C{H}_{4/C}}\times GW{P}_{C{H}_4}$$

E₂—Annual emissions of CH₄ converted to CO₂ equivalent from sludge removal, tCO₂ eq/a

SR—sludge dry matter removal, t/a

${\mathsf{\beta }}_{\mathrm{S}}$—Organic matter content in sludge dry matter of wastewater treatment plant, obtained by actual measurement method, tC/a

${\mathrm{DOC}}_{\mathrm{f}}$—Degradable organic carbon ratio in sludge dry matter, taken as 50%

F—Ratio of degradable organic carbon that can produce C in CH₄, taken as 50%

$C_{C{H}_{4/C}}$—ratio of molecular weight, taken as 16/12

(3)

CO₂ emissions from electricity consumption in wastewater treatment plants

$$E_3=EH\times E{F}_{CO}{}_2\times GW{P}_{C{O}_2}$$

E₃—CO₂ equivalent from the annual electricity consumption of the wastewater treatment plant equipment operation, tCO₂eq/a

EH—Electricity consumption from the operation of wastewater treatment plant equipment per year, MWh/a

${\mathrm{EF}}_{\mathrm{CO}}{}_2$—Power CO₂ emission factor, tCO₂/MWh

${\mathrm{GWP}}_{{\mathrm{CO}}_2}$—CO₂ global warming potential (GWP) value, taken as 1

(4)

Electricity calculation for wastewater plants

$$P_{COD}=58.503CO{D}_{in}^{-0.8431}+0.01\ast \left(50-CO{D}_{eff}\right)/10$$

P_COD—Power consumption per unit of COD reduction, kW·h

COD_in—Feedwater COD concentration, mg/L

COD_eff—Effluent COD concentration, mg/L.

3. Case Study

3.1. Study Area Overview

The Liaohe River Basin of Jilin Province (later referred to as the Liaohe River Basin) is located in the southwestern part of Jilin Province, China. The geographical location is in 123°42′~125°31′ E, 42°34′~44°08′ N. The area of the region is 11,293 km², accounting for 6.55% of the total land area of Jilin Province, and the geographical location is shown in Figure 2.

Figure 2. Overview of study area.

The main rivers in the territory include three major trunk (tributaries) rivers, namely Dongliao River, Zhaosutai River and Tiaozi River. The Dongliao River is 321 km long in Jilin, accounting for more than 80% of the total river length, controlling a watershed area of 10,136 km². The Zhaosutai River is a tributary of the Liao River, originating in Tumenling of Wangxiangtun in Lishu County and flowing westward through Lishu County to the mouth of Tongjiang River in Changtu County, Liaoning Province. Zhaosutai River is 103 km long in Jilin, with a controlled watershed area of 1147 km². The Jiaozi River originates in Lanjiagou, Shiling Town, Lishu County, and flows westward to Qujiadian Township, Changtu County, Liaoning Province, where it joins the Zhaosutai River. The Jiaozi River in Jilin is 58.3 km long and controls a watershed area of 463 km².

Water functional zones (WFZ) are unique to China and are artificially divided into different areas according to different functions. It is based on the economic development plan and comprehensive utilization plan of water resources in the basin, combined with the current situation of regional water resources development and utilization and social needs, in the corresponding watershed delineated with specific functions, to meet the requirements of rational development and utilization of water resources and protection and can play the best benefit of the zone (i.e., WFZ). China’s WFZ are divided into a two-tier system: the first-tier zone is water source protection zone, buffer zone, development and utilization zone and its reservation zone; the second-tier zone is drinking water source zone, industrial water use zone, agricultural water use zone, fishery water use zone, landscape recreation water use zone, transition zone and sewage control zone. Each WFZ has different water quality requirements according to the previous functional division. There are 38 primary WFZ and 34 secondary WFZ in the watershed of Liaohe River in Jilin Province. According to the different functions, there are different requirements on the water quality discharged into the water bodies by sewage plants. All sewage treatment plants in the Liaohe River Basin are mainly located in three cities and counties, with 10 urban sewage treatment plants in Gongzhuling District of Changchun City, 32 in Siping City, and 11 in Liaoyuan City. Most wastewater treatment plants currently running with four processes: A²/O, CASS, MBR and A/O-MBR. China currently has a variety of effluent standards, the most widely used standard is GB 18918-2002 “urban wastewater treatment plant pollutant discharge standards” Class A discharge standards, but there is also the implementation of the more stringent DB11/890-2012 “Beijing local standards” Class A discharge standards. At present, most of the wastewater plants in Liaohe River Basin implement the standard GB 18918-2002 Class A, and a few implement the latter.

3.2. AHP Construction of Index System

(1)

Environmental benefits

In the more polluted Liaohe River Basin, in order for the wastewater treatment plant to operate stably, its contribution to the discharged water bodies and the surrounding environment should be ensured and the GHG emissions generated during the treatment process should be minimized. In this study, the environmental benefits were measured using three indicators: GHG generation (I₁), WFZ contribution (I₂) and resource recovery (I₃). GHG generation (I₁) includes the amount of methane generated from COD removal (A₁₁), the amount of GHG generated from sludge treatment (A₁₂), and the amount of CO₂ emissions from electricity consumption (A₁₃). Resource recovery (I₃) is mainly reflected in the reuse of wastewater and the recycling of nutrients.

(2)

Economic benefits

The weight of economic benefits is closely related to the economic conditions of the study area. The cities in which the Liaohe River Basin located are economically underdeveloped should have a higher weight, and the normal OAM of the wastewater plant and the cost of electricity need local financial support. In this study, the three indicators of sludge treatment cost (I₄), electricity consumption cost (I₅) and OAM cost (I₆) are used to measure. Among them, the sludge treatment cost is the amount of activated sludge generated per unit weight of COD removed, the cost incurred to treat the corresponding sludge, the cost of electricity consumption is the amount of electricity consumed per unit weight of COD removed and thus the cost incurred, and the OAM cost includes the cost of various resources required for operation and the cost of labor.

(3)

Social benefits

Social benefits need to be considered from the socio-cultural perspective of the decision scenario, measuring the applicability of the process in the local context. In this study, four indicators of public evaluation, exposure to cold climate, stability and risk assessment were used for measurement. Public evaluation for the process in the treatment of malodorous gases and mosquitoes and other situations that will have a negative impact on public health, rural areas have higher requirements for public health compared to urban areas with poor infrastructure. Stability is the level of control to maintain a certain level of removal rate during operation, which is related to the complexity of the process and the average quality of the operator. The cold season performance is considered because the Liaohe River Basin is located in northeast China, and the cold climate in winter will last for 4–5 months, so the river flow is small and the water environment capacity decreases sharply compared with summer, whereas the pollution from living point sources will not decrease, and at the same time, it is difficult to remove pollutants with low incoming water temperature, so the cold season performance of wastewater plants is especially important. Stability is the level of control to maintain a certain level of removal rate during operation, which is related to the complexity of the process and the average quality of the operator. Risk assessment is then based on the amount of incoming water, whether a certain level of removal rate can meet the water quality requirements of the water function area. The results of the establishment of the indicator system are shown in Table 3.

Table 3. Indicators system.

Guidelines	Indicators	Properties
Environmental Benefit C1	GHG generation I1	GHG generation A1	A11 The amount of methane generated from the removal of COD
			A12 Amount of greenhouse gases from sludge treatment
			A13 CO2 emissions from electricity consumption
	WFZ contribution I2	Pollution load A2
	Resource recovery I3	Reclaimed water reuse A3
Economic benefit C2	Sludge treatment cost I4	Sludge treatment cost A4
	Power consumption cost I5	Power consumption cost A5
	OAM cost I6	OAM cost A6
Social Benefit C3	Public evaluation I7	Malodorous gas generation A7
	Influenced by cold climate I8	Affected by cold climate A8
	Stability I9	Stability A9
	Risk assessment I10	Risk assessment A10

3.3. AHP Index System Scenario Analysis

The indicator weights of the AHP model should be analyzed under multiple decision scenarios. In this study, the indicator system is applied to the typical technologies used in most wastewater plants in the Liaohe River Basin, which are the A²/O (DS₁) process, CASS process (DS₂), MBR process (DS₃) and A/O+MBR (DS₄). In this study, the model was applied to the above four different treatment processes to optimize the appropriate effluent index.

This study summarizes the wastewater treatment technologies used in wastewater plants in the basin by reviewing data, field visits and issuing questionnaires, and summarizes the characteristics and parameters of each technology, based on which scenario analysis is conducted to derive the importance of the relationship between indicators in different treatment technologies.

In this study, four typical wastewater treatment technologies used in the watershed were investigated, as shown in Figure 3, Figure 3a shows the percentage of treatment size of wastewater plants selected for the four processes, Figure 3b shows the percentage of the number of wastewater plants selected for each of the four processes. Figure 3 shows that: the number of wastewater plants with A²/O is 28.21%, but they are responsible for 87.03% of the wastewater treatment in the basin, whereas the number of wastewater plants with A/O+MBR and MBR is 41.03% and 25.64%, but they are responsible for only 2.17% and 0.71% of the wastewater treatment in the basin. This is because the A²/O process is the more mature, simplest process and the most widely used process for nitrogen removal and phosphorus removal, and is chosen by wastewater plants that have been in operation for a relatively long time in the basin, mainly in towns and cities, with a large average treatment scale. MBR and A/O+MBR process of which the MBR process is an emerging process, characterized by a small footprint, not limited by the setting occasion, easy to operate and manage, easy to achieve automatic control, in the basin by the new township small sewage plants used. Therefore, the WFZ contribution I2 should be of high importance in the A²/O (DS₁) and CASS processes (DS₂), whereas the MBR (DS₃) and A/O+MBR processes (DS₄) have less importance.

Figure 3. Number of wastewater plant processes and treatment scale.

The operation of wastewater plants cannot be separated from the financial support of the government, there are differences in the economic development of each region, so the financial support capacity for environmental management also varies greatly, the choice of a high level of effluent standards will incur higher operating costs, will put pressure on the local financial expenditure, Therefore, it is necessary to integrate the technical characteristics of each process and the economic situation of the town to optimize the effluent standard from the economic point of view.

Figure 4 shows the distribution of fiscal expenditures by county and municipal governments in the watershed, The data is obtained from the statistical yearbooks published in the latest year (2021) of Changchun, Siping and Liaoyuan cities. The statistical yearbook is an informative annual publication by each prefecture-level city in China on the statistical data of all areas within the city for the previous year, in order to fully reflect the economic and social development of China. Siping City District and Yitong County have high total expenditures, indicating that within the watershed, Siping City District and Yitong County are relatively well-funded; however, Yitong County, similar to other county-level cities in the watershed, does not have a high share of expenditures on general public health, public safety and energy consumption. In contrast, although the total fiscal expenditure in Liaoyuan city district is not high, it gives higher fiscal expenditure to the above three items, indicating that towns pay more attention to general public health, public safety and energy consumption. Therefore, DS₁ and DS₂ should have higher weights for power consumption cost (I₅), surrounding environmental impact (I₇) and risk assessment (I₁₀). Because the town’s funding budget is more adequate in the basin, economic efficiency (C₂) is therefore less important; the opposite is true for DS₃ and DS₄.

Figure 4. Distribution of financial expenditures by county and municipal governments in the basin.

3.4. AHP Weighting Analysis

The comprehensive decision background overview and scenario analysis allow the establishment of a two-by-two comparison matrix between each process index, a process supported by experts related to the water pollution prevention and control project in the Liaohe River Basin of Jilin Province. The comparison matrix between indicators consists of 1 target-to-criterion matrix and 3 criterion-to-indicator matrices. The comparison matrices between the indicators of CASS (DS₁), A²/O (DS₂), MBR (DS₃) and A/O+MBR (DS₄) are shown in Table 4, Table 5, Table 6 and Table 7, respectively.

Table 4. Comparison matrix between CASS indicators.

Objective Criteria	C1	C2	C3	C1 Indicators	I1	I2	I3
Environmental benefits (C1)	1	3	4	GHG generation (I1)	1	1/3	2
Economic benefits (C2)	1/3	1	1	WFZ contribution (I2)	3	1	3
Social benefits (C3)	1/4	1	1	Resource recovery (I3)	1/2	1/3	1
CR = 0.0089				CR = 0.0518
C2 Indicators	I4	I5	I6	C3 Indicators	I7	I8	I9	I10
Sludge treatment cost (I4)	1	1/4	1/4	Public evaluation (I7)	1	1/3	1/3	1/3
Power consumption cost (I5)	4	1	2	Affected by cold climate (I8)	3	1	1/2	1/2
OAM costs (I6)	4	1/2	1	Stability (I9)	3	2	1	1/2
				Risk assessment (I10)	3	2	2	1
CR = 0.0518				CR = 0.0456

Table 5. Comparison matrix between A²/O indicators.

Objective Criteria	C1	C2	C3	C1 Indicators	I1	I2	I3
Environmental benefits (C1)	1	2	3	GHG generation (I1)	1	1/2	2
Economic benefits (C2)	1/2	1	2	WFZ contribution (I2)	2	1	3
Social benefits (C3)	1/3	1/2	1	Resource recovery (I3)	1/2	1/3	1
CR = 0.0089				CR = 0.0089
C2 Indicators	I4	I5	I6	C3 Indicators	I7	I8	I9	I10
Sludge treatment cost(I4)	1	1/3	1/3	Public evaluation (I7)	1	1/3	1/3	1/3
Power consumption cost (I5)	3	1	2	Affected by cold climate (I8)	3	1	1	1/2
OAM costs (I6)	3	1/2	1	Stability (I9)	3	1	1	1/2
				Risk assessment (I10)	3	2	2	1
CR = 0.0517				CR = 0.0694

Table 6. Comparison matrix between MBR indicators.

Objective Criteria	C1	C2	C3	C1 Indicators	I1	I2	I3
Environmental benefits (C1)	1	1/2	2	GHG generation (I1)	1	2	2
Economic benefits (C2)	2	1	3	WFZ contribution (I2)	1/2	1	2
Social benefits (C3)	1/2	1/3	1	Resource recovery (I3)	1/2	1/2	1
CR = 0.0089				CR = 0.0517
C2 Indicators	I4	I5	I6	C3 Indicators	I7	I8	I9	I10
Sludge treatment cost (I4)	1	1/3	1/4	Public evaluation (I7)	1	1/4	1/4	1/4
Power consumption cost (I5)	3	1	1/3	Affected by cold climate (I8)	4	1	1/2	1/2
OAM costs (I6)	4	3	1	Stability (I9)	4	2	1	2
				Risk assessment (I10)	4	2	1/2	1
CR = 0.013				CR = 0.0457

Table 7. Comparison matrix between A/O+MBR indicators.

Objectives Criteria	C1	C2	C3	C1 Indicators	I1	I2	I3
Environmental benefits (C1)	1	1/5	1/2	GHG generation (I1)	1	1	2
Economic benefits (C2)	5	1	5	WFZ contribution (I2)	1	1	3
Social benefits (C3)	2	5	1	Resource recovery (I3)	1/2	1/3	1
CR = 0.0521				CR = 0.0176
C2 Indicators	I4	I5	I6	C3 Indicators	I7	I8	I9	I10
Sludge treatment cost(I4)	1	1/4	1/2	Public evaluation (I7)	1	1/5	1/5	1/5
Power consumption cost (I5)	4	1	1/3	Affected by cold climate (I8)	5	1	1/3	1/3
OAM costs (I6)	2	3	1	Stability (I9)	5	3	1	1
				Risk assessment (I10)	5	3	1	1
CR = 0.013				CR = 0.0457

After calculating the consistency CR values for each indicator matrix, the calculation results showed that the indicator comparison matrices all had good consistency. In this study, the calculation of the criterion weights (W_i), the indicator-to-criterion weights (W_j) and the indicator-to-decision target weights (W_i) of the indicator matrices was completed using Matlab software, and the results are shown in Table 8.

Table 8. Weight distribution of indicator matrix.

	CASS			A2/O			MBR			A/O = MBR
	Wi	Wj	Wi	Wi	Wj	Wi	Wi	Wj	Wi	Wi	Wj	Wi
Environmental Benefit (C1)	0.6327			0.5390			0.2973			0.1149
GHG generation (I1)		0.2519	0.1594		0.2973	0.1602		0.4905	0.1458		0.3873	0.0445
WFZ contribution (I2)		0.5889	0.3726		0.5390	0.2905		0.3119	0.0927		0.4429	0.0051
Resource recovery (I3)		0.1593	0.1008		0.1638	0.0883		0.1976	0.0587		0.1698	0.0195
Economic benefits (C2)	0.0518			0.2973			0.5390			0.7028
Sludge treatment cost (I4)		0.1103	0.0212		0.1416	0.0421		0.1199	0.0646		0.1307	0.0919
Power consumption cost (I5)		0.5437	0.1046		0.5247	0.1560		0.2721	0.1467		0.6768	0.4757
OAM costs (I6)		0.3460	0.0666		0.3338	0.0992		0.6080	0.3277		0.1925	0.1353
Social benefits (C3)	0.1749			0.1638			0.1638			0.1149
Public evaluation (I7)		0.0981	0.0171		0.0954	0.0156		0.0756	0.0124		0.062	0.0113
Affected by cold climate (I8)		0.2081	0.0364		0.2535	0.0415		0.2134	0.0350		0.1786	0.0326
Stability (I9)		0.2875	0.0503		0.3514	0.0575		0.4166	0.0682		0.3797	0.0692
Risk assessment (I10)		0.4063	0.0710		0.2997	0.0491		0.2944	0.0482		0.3797	0.0692

As can be seen from Table 6, the environmental benefit weights of DS₁ and DS₂ at the criterion level reached 0.6327 and 0.5390, respectively, because the two processes are mainly distributed in towns and cities, with large average treatment scale and sufficient government financial expenditures, which attach the most importance to the improvement of the environment. In the remaining two criteria, DS₁ places more importance on social benefits than economic benefits, with a weight of 0.1749, whereas DS₂, in contrast to DS₁, places more importance on economic benefits, due to the high sludge treatment cost I₄, electricity consumption cost I₅ and OAM cost I₆ in the process parameters of DS₂, resulting in the need to pay more attention to its economic benefits. While the economic benefits weights of DS₃ and DS₄ reached 0.5390 and 0.7028, respectively, these two processes are mainly distributed in the countryside, mostly decentralized treatment units with small average treatment scale, and the local economy is underdeveloped with insufficient budget and more attention to economic savings. Of the two remaining criteria, DS₃ has a higher weight of environmental benefits than social benefits because the parameters related to indicator resource recovery (I₃) in the MBR process are low, so more attention should be paid to them, whereas DS₄ has the same weight of environmental and social benefits. The distribution of weights at the criterion level corresponds to the individual characteristics of each process and the geographical characteristics of the adopted location.

At the indicator level, the weight of the WFZ contribution of the three indicators DS₁ and DS₂ in environmental benefits reached 0.3726 and 0.2905, respectively, because of the large scale of treatment, so that the pressure to assess the water quality standards in the WFZ increased, DS₃ and DS₄ are more concerned about GHG emissions because of the small scale of treatment. The economic benefits of DS₁, DS₂ and DS₄ are more concerned about the cost of power consumption (I₅), whereas DS₃ is more concerned about the OAM cost (I₆), which is due to the MBR process itself process characteristics, the high cost of the core module membrane module, short life time, and the cost increase because other debris in the wastewater can cause pollution damage to the membrane module, whereas the A/O+MBR process avoids the above problems because the A/O process is coupled with the MBR process. The focus of attention for each of the four processes in the social benefits varies, which is determined by the respective technical characteristics and parameters of each process.

After the weighting analysis, it was found that the distribution of indicator weights for each process was reasonable and consistent with the scenario analysis in the previous section, the weights allowed for decision making on the most available options for the effluent criteria.

3.5. Options and Attribute Values

In this study, four different effluent standards (based on COD concentration) were selected, and the attribute values were calculated for different processes at different discharge targets. The four effluent standards are the widely used effluent standard of 50 mg/L in the study area and the most stringent 20 mg/L at present, and two gradients of 40 mg/L and 30 mg/L are set in them. For the quantified part of the available data, this paper calculates the annual treatment scale of different technologies in the study area as the benchmark, and the calculation formula is as in Section 2.2, and the electricity cost is referred to the local electricity cost in Jilin Province are calculated and finally converted to USD uniformly. For the attribute values that are difficult to determine, expert opinions are consulted to ensure the objectivity and accuracy of the data, and finally for different attribute values are finally classified into five levels as low, lower, medium, higher and high.

Translated with www.DeepL.com/Translator (free version). The attribute values for each process at different emission standards are shown in the Table 9, Table 10, Table 11 and Table 12.

Table 9. Property values of CASS.

Properties		50 mg/L (CA1)	40 mg/L (CA2)	30 mg/L (CA3)	20 mg/L (CA4)
GHG generation I1	I11 The amount of methane generated from the removal of COD	12,072	12,474	12,877	13,279
	I12 Amount of greenhouse gases from sludge treatment	3066	3168	3270	3373
	I13 CO2 emissions from electricity consumption	1014	1073	1134	1196
	Total	16,153	16,716	17,281	17,848
WFZ contribution I2 (t/a)		456	365	274	183
Reclaimed water reuse I3		Low	Lower	Mid	High
Sludge treatment cost I4 ($/a)		86,012	88,879	91,746	94,613
Electricity costs I5 ($/a)		92,080	95,149	98,219	101,288
OAM costsI6		Low	Lower	Mid	High
Odor generation I7		Low	Lower	Mid	High
Cold climate treatment effect I8		High	Mid	Mid	Low
Stability I9		High	Higher	Mid	Low
Risk Assessment I10		High	Mid	Mid	Low

Table 10. Property values of A²/O.

Properties		50 mg/L (CA1)	40 mg/L (CA2)	30 mg/L (CA3)	20 mg/L (CA4)
GHG generation I1	I11 The amount of methane generated from the removal of COD	18,921	19,551	20,182	20,813
	I12 Amount of greenhouse gases from sludge treatment	4805	4965	5126	5286
	I13 CO2 emissions from electricity consumption	1590	1682	1777	1874
	Total	25,316	26,199	27,085	27,973
WFZ contribution I2 (t/a)		715	572	429	286
Reclaimed water reuse I3		Low	Lower	Mid	High
Sludge treatment cost I4 ($/a)		134,890	139,386	143,883	148,379
Electricity costs I5 ($/a)		144,314	149,125	153,935	158,746
OAM costs I6		Low	Lower	Mid	High
Odor generation I7		Low	Lower	Mid	High
Cold climate treatment effect I8		High	Mid	Mid	Low
Stability I9		High	Higher	Mid	Low
Risk Assessment I10		High	Mid	Mid	Low

Table 11. Property values of MBR.

Properties		50 mg/L (CA1)	40 mg/L (CA2)	30 mg/L (CA3)	20 mg/L (CA4)
GHG generation I1	I11 The amount of methane generated from the removal of COD	169	174	180	185
	I12 Amount of greenhouse gases from sludge treatment	43	44	46	47
	I13 CO2 emissions from electricity consumption	14	15	16	17
	Total	226	234	242	250
WFZ contribution I2 (t/a)		6	5	4	3
Reclaimed water reuse I3		Low	Lower	Mid	High
Sludge treatment cost I4 ($/a)		1204	1244	1284	1325
Electricity costs I5 ($/a)		1289	1332	1375	1418
OAM costs I6		Low	Lower	Mid	High
Odor generation I7		Low	Lower	Mid	High
Cold climate treatment effect I8		High	Mid	Mid	Low
Stability I9		High	Higher	Mid	Low
Risk Assessment I10		High	Mid	Mid	Low

Table 12. Property values of A/O-MBR.

Properties		50 mg/L (CA1)	40 mg/L (CA2)	30 mg/L (CA3)	20 mg/L (CA4)
GHG generation I1	I11 The amount of methane generated from the removal of COD	324	335	346	356
	I12 Amount of greenhouse gases from sludge treatment	82	85	88	91
	I13 CO2 emissions from electricity consumption	27	29	30	32
	Total	434	449	464	480
WFZ contribution I2 (t/a)		12	10	7	5
Reclaimed water reuse I3		Low	Lower	Mid	High
Sludge treatment cost I4 ($/a)		2312	2389	2466	2543
Electricity costs I5 ($/a)		2475	2557	2640	2722
OAM costs I6		Low	Lower	Mid	High
Odor generation I7		Low	Lower	Mid	High
Cold climate treatment effect I8		High	Mid	Mid	Low
Stability I9		High	Higher	Mid	Low
Risk Assessment I10		High	Mid	Mid	Low

3.6. Indicator Values

In this study, the attribute values identified in the previous subsection were transformed into natural language and classified into five different grades: good, better, medium, poorer, and poor. After the assignment transformation and weighting calculated in Section 3, the indicator values for each process alternative are shown in the Table 13, Table 14, Table 15 and Table 16.

Table 13. Weighted indicator values of CASS.

Indicator	CA1			CA2			CA3			CA4
GHG generation I1	0.0956	0.1275	0.1594	0.0478	0.0797	0.1116	0.0000	0.0319	0.0638	0.0000	0.0000	0.0319
WFZ contribution I2	0.0000	0.0000	0.0745	0.1118	0.1863	0.2608	0.2236	0.2981	0.3726	0.2981	0.3726	0.3726
Reclaimed water reuse I3	0.0837	0.1046	0.1046	0.0628	0.0837	0.1046	0.0314	0.0523	0.0732	0.0000	0.0209	0.0418
Sludge treatment cost I4	0.0170	0.0212	0.0212	0.0127	0.0170	0.0212	0.0064	0.0106	0.0148	0.0000	0.0042	0.0085
Electricity costs I5	0.0533	0.0666	0.0666	0.0400	0.0533	0.0666	0.0200	0.0333	0.0466	0.0000	0.0133	0.0266
OAM costs I6	0.0103	0.0137	0.0171	0.0051	0.0086	0.0120	0.0000	0.0034	0.0068	0.0000	0.0000	0.0034
Odor generation I7	0.0000	0.0202	0.0403	0.0302	0.0504	0.0706	0.0605	0.0806	0.1008	0.0806	0.1008	0.1008
Cold climate treatment effect I8	0.0218	0.0291	0.0364	0.0109	0.0182	0.0255	0.0000	0.0073	0.0146	0.0000	0.0000	0.0073
Stability I9	0.0302	0.0402	0.0503	0.0151	0.0252	0.0352	0.0000	0.0101	0.0201	0.0000	0.0000	0.0101
Risk Assessment I10	0.0426	0.0568	0.0710	0.0213	0.0355	0.0497	0.0000	0.0142	0.0284	0.0000	0.0000	0.0142

Table 14. Weighted indicator values of A²/O.

Indicator	CA1			CA2			CA3			CA4
GHG generation I1	0.1282	0.1602	0.1602	0.0961	0.1282	0.1602	0.0481	0.0801	0.1121	0.0000	0.0320	0.0641
WFZ contribution I2	0.0000	0.0000	0.0581	0.0872	0.1453	0.2034	0.1743	0.2324	0.2905	0.2324	0.2905	0.2905
Reclaimed water reuse I3	0.0000	0.0177	0.0353	0.0265	0.0442	0.0618	0.0530	0.0706	0.0883	0.0706	0.0883	0.0883
Sludge treatment cost I4	0.0253	0.0337	0.0421	0.0126	0.0211	0.0295	0.0000	0.0084	0.0168	0.0000	0.0000	0.0084
Electricity costs I5	0.0936	0.1248	0.1560	0.0468	0.0780	0.1092	0.0000	0.0312	0.0624	0.0000	0.0000	0.0312
OAM costs I6	0.0794	0.0992	0.0992	0.0595	0.0794	0.0992	0.0298	0.0496	0.0694	0.0000	0.0198	0.0397
Odor generation I7	0.0094	0.0125	0.0156	0.0047	0.0078	0.0109	0.0000	0.0031	0.0062	0.0000	0.0000	0.0031
Cold climate treatment effect I8	0.0249	0.0332	0.0415	0.0125	0.0208	0.0291	0.0000	0.0083	0.0166	0.0000	0.0000	0.0083
Stability I9	0.0345	0.0460	0.0575	0.0173	0.0288	0.0403	0.0000	0.0115	0.0230	0.0000	0.0000	0.0115
Risk Assessment I10	0.0295	0.0393	0.0491	0.0147	0.0246	0.0344	0.0000	0.0098	0.0196	0.0000	0.0000	0.0098

Table 15. Weighted indicator values of MBR.

Indicator	CA1			CA2			CA3			CA4
GHG generation I1	0.0875	0.1166	0.1458	0.0437	0.0729	0.1021	0.0000	0.0292	0.0583	0.0000	0.0000	0.0292
WFZ contribution I2	0.0000	0.0000	0.0185	0.0278	0.0464	0.0649	0.0556	0.0742	0.0927	0.0742	0.0927	0.0927
Reclaimed water reuse I3	0.0000	0.0117	0.0235	0.0176	0.0294	0.0411	0.0352	0.0470	0.0587	0.0470	0.0587	0.0587
Sludge treatment cost I4	0.0517	0.0646	0.0646	0.0388	0.0517	0.0646	0.0194	0.0323	0.0452	0.0000	0.0129	0.0258
Electricity costs I5	0.1174	0.1467	0.1467	0.0880	0.1174	0.1467	0.0440	0.0734	0.1027	0.0000	0.0293	0.0587
OAM costs I6	0.2622	0.3277	0.3277	0.1966	0.2622	0.3277	0.0983	0.1639	0.2294	0.0000	0.0655	0.1311
Odor generation I7	0.0074	0.0099	0.0124	0.0037	0.0062	0.0087	0.0000	0.0025	0.0050	0.0000	0.0000	0.0025
Cold climate treatment effect I8	0.0210	0.0280	0.0350	0.0105	0.0175	0.0245	0.0000	0.0070	0.0140	0.0000	0.0000	0.0070
Stability I9	0.0409	0.0546	0.0682	0.0205	0.0341	0.0477	0.0000	0.0136	0.0273	0.0000	0.0000	0.0136
Risk Assessment I10	0.0289	0.0386	0.0482	0.0145	0.0241	0.0337	0.0000	0.0096	0.0193	0.0000	0.0000	0.0096

Table 16. Weighted indicator values of A/O-MBR.

Indicator	CA1			CA2			CA3			CA4
GHG generation I1	0.0267	0.0356	0.0445	0.0134	0.0223	0.0312	0.0000	0.0089	0.0178	0.0000	0.0000	0.0089
WFZ contribution I2	0.0000	0.0000	0.0010	0.0015	0.0026	0.0036	0.0031	0.0041	0.0051	0.0041	0.0051	0.0051
Reclaimed water reuse I3	0.0000	0.0039	0.0078	0.0059	0.0098	0.0137	0.0117	0.0156	0.0195	0.0156	0.0195	0.0195
Sludge treatment cost I4	0.0735	0.0919	0.0919	0.0551	0.0735	0.0919	0.0276	0.0460	0.0643	0.0000	0.0184	0.0368
Electricity costs I5	0.3806	0.4757	0.4757	0.2854	0.3806	0.4757	0.1427	0.2379	0.3330	0.0000	0.0951	0.1903
OAM costs I6	0.1082	0.1353	0.1353	0.0812	0.1082	0.1353	0.0406	0.0677	0.0947	0.0000	0.0271	0.0541
Odor generation I7	0.0068	0.0090	0.0113	0.0034	0.0057	0.0079	0.0000	0.0023	0.0045	0.0000	0.0000	0.0023
Cold climate treatment effect I8	0.0196	0.0261	0.0326	0.0098	0.0163	0.0228	0.0000	0.0065	0.0130	0.0000	0.0000	0.0065
Stability I9	0.0415	0.0554	0.0692	0.0208	0.0346	0.0484	0.0000	0.0138	0.0277	0.0000	0.0000	0.0138
Risk Assessment I10	0.0415	0.0554	0.0692	0.0208	0.0346	0.0484	0.0000	0.0138	0.0277	0.0000	0.0000	0.0138

3.7. Simulation Results

According to the equation, the posting schedule Ci of the model was calculated, and the results are shown in Table 17.

Table 17. C_i values for each process option.

	CASS		A2/O		MBR		A/O-MBR
	Ci	Sort	Ci	Sort	Ci	Sort	Ci	Sort
CA1	0.333	4	0.427	4	0.765	1	0.962	1
CA2	0.527	3	0.563	3	0.718	2	0.774	2
CA3	0.686	1	0.630	1	0.480	3	0.474	3
CA4	0.674	2	0.584	2	0.323	4	0.275	4

From the calculation results, it can be seen that CA₃ has the highest score among CASS process and A²/O process, which indicates that CA₃ is a suitable effluent standard for operating CASS and A²/O process wastewater plants in the Liaohe River Basin, whereas CA₁ is optimal for MBR and A/O-MBR processes. CA₄, as the most stringent effluent criterion, did not achieve a high priority ranking among the four technologies, and even ranked last among MBR and A/O-MBR, limited by the economy of the basin and more greenhouse gases produced, even though CA₄ achieved the optimum in I₂ and I₃, it could not achieve the optimum in the final weighting, indicating that the stringent effluent criterion was not suitable for the basin. CA₃ is optimal in the CASS process and A²/O process. CA₃ is the most appropriate effluent standard as a more stringent effluent standard, balancing environmental, economic and social benefits in large water plants. CA₂ does not show outstanding advantages among the four technologies and is not suitable as the optimal choice for any one of them. CA₁ has the highest score among MBR and A/O-MBR processes and has a weight of 0.962 in the A/O-MBR process species, indicating that a lenient effluent standard is more appropriate for small wastewater treatment facilities located in this region of the township.

4. Conclusions

This study provides an opinion on what effluent standards to implement in the Liaohe River Basin through the coupled AHP and fuzzy TOPSIS model. The study firstly analyzed the scenarios of the four main wastewater treatment technologies used in the basin, analyzed the different technical characteristics of each technology and the background of their actual use in the basin, calculated the index weights of each technology using the AHP method, then determined the attribute values of each process under different effluent standards through extensive calculations and interviews with experts, and finally ranked the process alternatives by the fuzzy TOPSIS method. The ranking of the process alternatives was obtained by the fuzzy TOPSIS method. The following conclusions were obtained from this study.

(1)

The model results show that, with the effluent COD as the index, the effluent standard is 30 mg/L for A2/O and CASS processes as appropriate, whereas the effluent standard is 50 mg/L for MBR and A/O-MBR.

(2)

At the criterion level, different processes pay different attention to the environment (C₁) and economy (C₂) for their own technical characteristics and the actual context of use in the basin, with the technologies DS₁ and DS₂ chosen for large water plants paying more attention to C₁ and the technologies DS₃ and DS₄ chosen for small water plants paying more attention to C₂. The focus of attention of each technology at the indicator level is different, I₂ is the most important indicator for DS₁, I₂, I₁ and I₅ are important indicators for DS₂, I₆ is the most important indicator for DS₃ and I₅ is the most important indicator for DS₄, the fluctuation of the indicator weight has a greater potential to affect the choice of the optimal solution, when making similar environmental decision problems, the weight of the indicator needs to be focused on the situation.

(3)

Although some of the results of each process are similar, the actual weight distribution is not the same, the same process used in large water plants, CASS process CA₃ and CA₄ weights are similar, whereas CA₃ in the A²/O advantage is obvious, because CASS process operating costs are relatively low, the economic benefits of low weight, local pollution control pressure can also choose CA₄. MBR and A/O-MBR The technology chosen for small water plants, the options are ranked the same, with CA₁ weighting of 0.962 for the A/O-MBR process, which is more suitable for the process in the basin’s lenient effluent standards.

Based on the above findings, it can be found that the coupled model can scientifically and effectively optimize the effluent standards of different technologies for wastewater treatment plants in the Liaohe River Basin. Considering the similarity of economic development and climate in Northeast China and the pressure of pollution control, the model proposed in this study can be extended to other provinces and cities in Northeast China to provide a decision support framework for the selection of effluent standards and can be used as a reference method for decision making in other industries.

In addition, there are still shortcomings and need for subsequent improvement in this study.

The process parameters and the weights of the indicators of the alternative options play an important role in the results of the preferences, and every effort should be made to ensure the integrity and accuracy of the data and the objectivity of the evaluation of the indicators. The four processes in this study were simply categorized as urban and rural by the location of the wastewater plant, and although the towns and villages in the same region are similar on an economic, environmental and social level, there is still some variability, and more refinement of the study area is needed in subsequent studies of this kind. In addition, the pollutants in wastewater plants are not only COD; however, NH3-N is also one of the indicators to evaluate the wastewater quality, and more indicators need to be considered to make the results more accurate in subsequent studies.