1. Introduction
During landfill disposal, the decomposition of organic matter in waste can easily lead to the production of landfill gases [
1,
2]. The gases, primarily composed of methane, are a significant source of greenhouse gas emissions, contributing up to 30% of global anthropogenic greenhouse gas emissions [
3]. In addition, odors and other health risks can become problematic in sanitary waste landfills [
4]. The selection and use of landfill cover materials are important in reducing landfill-related gas and odor emissions [
5]. Specifically, landfill covers can accelerate the conversion of methane into carbon dioxide by promoting the growth of methane-oxidizing bacteria [
6,
7]. Furthermore, the covering material can effectively filter and prevent the emission of odors and gas, prevent the breeding of mosquitos and flies, and reduce the spread of bacteria [
8,
9].
Currently, landfill covers are made of either natural materials, which primarily include compacted clay and modified clays with stabilizing additives or synthetic materials, such as geocomposites, polymer composites, and painting materials [
10,
11,
12,
13,
14]. Natural materials are easily obtainable, affordable, and accompanied by stable engineering performance. However, the high cost of the related construction machinery, which has poor seepage and airproof characteristics, has resulted in natural materials not meeting the requirements of modern landfills [
7,
15]. In contrast, artificial materials are ductile, have suitable tensile properties, and are water- and airproof. In addition, as artificial covers are light and thin, the related structures are relatively stable. Under normal circumstances, artificial covers are long-lasting. However, being lightweight, synthetic covers are easily damaged and cannot be reused, leading to a higher economic cost [
16]. Therefore, the various cover materials have specific advantages and disadvantages, and the selection of the most appropriate material with the best overall performance is becoming increasingly important. Typically, the selection of a landfill cover material is based on the subjective experience of engineers, from an economic feasibility perspective, or based on the performance of materials, with little consideration of the environmental impacts of the material [
17,
18].
Within this context and under the guidance of sustainable design, we developed a comprehensive material index system that considers the basic function of various materials, their economic feasibility, and their environmental impact. Thus, using multicriteria decision making (MCDM), we were able to compare and select the most appropriate cover material. In addition, we applied the model using a case study. The findings of our study provide insight into landfill management practices, leading to a reduction in associated landfill gases and, ultimately, increasing the sustainability of landfills.
2. Literature Review
The selection of appropriate materials can be complex as it involves multiple evaluation criteria and attribute information, with possible contradictory relationships between attributes, such as economic versus environmental factors [
18]. Within this context, MCDM is widely used for the selection of materials. Depending on the specific application, MCDM can be classified into multiple objective decision making (MODM), focusing on objective optimization; and multiple attribute decision making (MADM), focusing on objective comparisons [
19]. The decision variables of MODM are continuous and embedded in the region determined by the constraints, resulting in a multiobjective decision-making method with infinite schemes, which are primarily used for optimal design [
20,
21]. In comparison, MADM has a finite decision scheme, resulting in a discrete multi-objective decision-making method, which primarily focuses on the ranking or preference of finite decision schemes with multiple attributes or indicators [
22].
The selection of materials is a typical MADM problem. The classical MADM approach includes a number of material selection methods, such as the analytic hierarchical process (AHP), analytic network process (ANP), vise višekriterijumska optimizacija i kompromisno rešenje (VIKOR), technique for order preference by similarity to ideal solution (TOPSIS), elimination and choice translating reality (ELECTRE), simple additive weighted (SAW), weighted product method (WPM), and data envelopment analysis (DEA) [
23,
24,
25]. These methods allow materials to be ranked according to specified criteria, and thus offer assistance to decision makers. For example, in the classification and selection of soft and hard magnetic materials, Chauhan and Vaish [
26] used the VIKOR method with triangular intuitionistic fuzzy numbers to identify the optimal solution by order of preference. Similarly, Kumar et al. [
27] used TOPSIS to evaluate the best material for the design of exhaust pipes, using the cost of materials as an important indicator. In addition, within the context of selecting “sustainable” materials, Zhao [
28] used a gray correlation method within MADM for the selection of commercial materials. Using a plastic pipe as an example, the performance of the material was divided into related economic, environmental, and social aspects, with AHP used to assign weights to the performance indicators of alternative materials, while the weighted coefficient of the environmental indicators was increased. Zhou et al. [
29] used refrigerator shell material as a case study to investigate the process and mechanical performance, together with the economic and environmental aspects, based on an MODM model integrating neural networks and genetic algorithms. Notably, the selection of eco-friendly materials has not yet been standardized, and there is a need for quantitative research and analyses to guide the related material selection decisions.
In this study, the selection of study materials was based on a combination of function, cost, and environmental impact, with a specific focus on the economic and environmental feasibility of the materials, i.e., the ratio of cost to function or the ratio of environmental impact to cost, which is a typical efficiency-based evaluation. For this task, DEA has stronger adaptability and flexibility.
First proposed by Charnes [
30], DEA is now widely used, among others, in business decision making, technology evaluation, and especially in material selection, where it has been successfully applied across a number of different contexts [
31,
32]. Based on DEA, Dickson [
33] used an evaluation index system in the selection of suppliers of construction materials, allowing the suppliers to be ranked according to selected indicators. In addition, Peng et al. [
34] developed an improved cross-efficiency DEA model based on entropy weight TOPSIS to solve the optimization problem of wood for furniture. This development allowed for an analysis of the effectiveness and feasibility of the material selection and evaluation model. Furthermore, to improve the economic feasibility of construction material suppliers, Hemmati et al. [
35] combined DEA with TOPSIS and used a G-C
2R model in generalized DEA to address various problems, including the reverse order of TOPSIS and the incomplete ordering of DEA. In addition, TOPSIS was used to rank the effective suppliers, thereby providing a basis for decision making. Similarly, Safa et al. [
36] combined the AHP and DEA methods and used suppliers of exterior curtain wall engineering as a case study to establish an evaluation index and demonstrate the applicability of the AHP/DEA method in the selection of construction material suppliers.
Because the input and output indexes of the traditional DEA do not contain weighted coefficients, they cannot reflect decision makers’ preferences via evaluation indexes, which may lead to a large deviation between theory and practice [
37,
38]. In particular, this deviation can occur when the DEA model contains a large number of input and output indicators together with unconstrained weights. This results in a disproportionate validity of the decision units and reduces the effectiveness of the model. To solve the constraint of index weight distribution, Charnes and Cooper [
39] developed a DEA model with a cone ratio (C
2WH model), which incorporates the preferences of decision makers. On this basis, Wu [
40] introduced the concept of the AHP constraint cone by applying the AHP to the relative evaluation of DEA to better reflect the preferences of decision makers. Hahn et al. [
41] applied the cone ratio C
2WH model to a technology evaluation index system of road networks and calculated the relative efficiency of each road network system by using the C
2R model and the C
2WH model, separately. Their results indicated that the C
2WH model, which considers the preference of decision makers, was more appropriate than the C
2R model. Thus, the DEA model has unique advantages in analyzing the efficiency of systems with multiple inputs and outputs as well as different dimensions.
In this study, we addressed the issue of material selection using a multiobjective input and output system that considers the function, economic value, and environmental impact of various landfill cover materials. Specifically, we calculated the relative efficiency of each evaluation unit using a DEA-C2WH model with a preference cone to obtain ranked evaluation values, allowing for the identification of the optimum landfill cover material.
4. Results and Discussion
Using MATLAB (R2019b, Mathworks, Natick, MA, USA), the data in matrices C and B were entered and run to obtain the input and output data:
The C
2WH model of the DMU
1, i.e., clay covering material, was as follows:
Similarly, the C
2WH model of DMU
2, i.e., HDPE geomembrane, was as follows:
The C
2WH model of DMU
3, i.e., PVC covering material, was as follows:
The C
2WH model of DMU
4, i.e., GCL waterproof blanket, was as follows:
MATLAB software was run to solve the above four planning equations. See
Table 6 for the output results. Using the developed C
2WH model, we evaluated the efficiency (E) of the four investigated materials. The outputs of the model were E1 = 0.2600, E2 = 0.5757, E3 = 0.7815, and E4 = 1.0000. Our results indicated that DMU
4 was relatively effective among the four materials, with GCL having the highest efficiency evaluation value. In terms of performance, the four investigated materials could be ranked as follows: GCL > PVC > HDPE > clay.
As GCL obtained the highest efficiency evaluation value, it could be interpreted as being the optimal landfill cover material. Specifically, GCL waterproof blanket material prevents seepage, occupies less space, is easy to install, and is cost-effective. In addition, GCL waterproof blanket material is self-repairing and can increase the self-waterproofing effect of the cover [
55,
56]. These characteristics explain the recent trend of replacing or partially replacing clay cover materials with GCL waterproof blankets. In terms of sustainable design, GCL waterproofing blankets have lesser environmental and social impacts than other materials. This result is consistent with the results of other studies on the mechanics, permeability, and stability of long-term GCL landfill covers [
57,
58]. This agreement indicates that the C
2WH model used in our study can be applied to further research on landfill cover materials.
PVC materials and HDPE geomembranes are both composed of artificial composite materials. Notably, PVC has slightly better environmental properties, a less energy-intensive production process, and less of an environmental impact related to the production life cycle than HDPE geomembranes, resulting in better overall performance. HDPE geomembranes and clay covers are currently the most commonly used cover materials for landfills. This highlights the discrepancy between the results of our model and commonly used engineering practices. Furthermore, this discrepancy between theory and practice underlines the shortcomings of material selection through traditional empirical methods, resulting in a greater impact on the environment. Notably, based on sustainability theory, HDPE geomembranes and clay materials, which have a high energy consumption, high ecological index, and low recycling rate, are not optimal cover materials and should gradually be replaced by materials with better performance [
59,
60]. Within this context, we suggest that GCL waterproof blankets, composed of a variety of materials, should be used for landfill covers. GCL covers prevent seepage, are easy to install, and can decrease the environmental problems caused by the antiseepage defects of monocomponent cover materials.
5. Conclusions and Further Study
Combined with the market research on the current situation of sanitary landfill cover material use, in this study, we evaluated the performance, economic value, and environmental properties of selected materials using a comprehensive analysis that combined qualitative and quantitative aspects. The C2WH model with preference cone was used to establish the evaluation model of cover material to achieve the optimal selection of sanitary landfill cover material.
Based on the concept of sustainable design, four typical landfill cover materials were investigated, namely clay, HDPE, PVC, and GCL, using a combined AHP and DEA material selection and evaluation method. Specifically, we developed an evaluation index system, with economic and environmental performance representing the input index and the basic performance of material as the output index. The C2WH model with a preference cone was successfully used to evaluate and rank the investigated landfill cover materials, with results indicating that GCL material was the optimal cover material. This result provides a reference and data support for multiobjective decision-making problems regarding material selection.
Although we included a number of factors in the developed evaluation index system, some influencing factors, such as landfill gas generated in the landfill process, fell beyond the scope of this study. In addition, we used the C2WH model with preference cones to reflect the subjective judgment of decision makers. In light of clear individual differences in subjective preferences, which can easily influence the assignment of weights and evaluation results, further research is needed regarding sensitivity analyses. In the next step, we will construct a more complete index system and select representative performance indicators from a broader range. In addition to a single cover layer material, according to the specific engineering application, a composite liner system can be considered, such as HDPE + GCL or clay + HDPE geomembrane and other composite cover structures, and then more comprehensively examine the material performance of sanitary landfill cover layer.