Keywords

1 Introduction

The Law 12305, which deals with the National Solid Waste Policy (PNRS), was implemented in 2010. It aims to eradicate the so-called ‘dumps’, prepare and implement municipal solid waste plans. Thinking in terms of the circular economy, Da Silva (2018) argues that investments in public policies of environmental education, sectoral and innovation policies are necessary to reorganize chains, turning a problem into an opportunity for municipalities.

An alternative to selective waste collection is to transform the traditional service (static) approach into a Product-Service-System (PSS). Thus, a service component (waste collection) is improved by a product component, i.e., technological alternatives to monitor waste, allowing the management of variable and dynamic waste streams (Elia et al. 2018). These technical options are part of a new way of thinking productive systems through smart cities. Díaz-Díaz et al. (2017) analyzed and compared business models in Intelligent Cities. Their results indicated that municipal services using smart technologies generally present a value proposition focused on service efficiency, which in consequence reduces environmental impact and lower costs.

Sustainability in operations is a necessity associated with waste management activities due to the complexity and amount of produced waste. Alternative systems and technologies for waste management have been researched as a way to solve or improve conventional systems, using dumpsters with sensors and Internet of Things (IoT), for example (Misra et al. 2018; Wen et al. 2017; Yerraboina et al. 2018).

However, these alternatives are not only technological. Others are related to paying schemes for the produced waste, by either weight or volume (Dahlen et al. 2010). These options can be very efficient when associated with waste bins technologies with RFID sensors tags. These bins can assist in charging through the measurement of weight and volume as well as in the inspection of the collection, transport and final destination (Wen et al. 2017).

Considering this context, the main objective behind the study is to structure a model to evaluate the implementation of sustainable operations and innovative technologies in the MSWM. A systematic review of the literature was carried out. The opinion of experts was used to select criteria and more adequate alternatives adjusted to the setting of a municipality in the Western region of the Brazilian state of Santa Catarina.

2 Theoretical Background

The management of solid waste by municipalities is crucial for public health, environmental protection and avoid visual pollution. It is necessary to properly manage all activities involving solid waste, from collection to final disposal (Al-Khatib et al. 2007). Hlatka et al. (2018) point out that residue separation is significantly influenced by the conditions of households for solid waste sustainable management. Topaloglu et al. (2018) emphasized that waste management must be environmentally sound, economically viable and socially acceptable.

According to Coban et al. (2018), solid waste management depends on the composition of waste produced by the population; it is strongly influenced by socioeconomic factors, seasons and family size. Considering an urban development with inefficient infrastructure and management, waste issues become increasingly complex. Authorities require effective tools to select appropriate technologies that meet the needs of the local infrastructure. Coban et al. (2018) state that multicriteria decision tools (MCDM) stand out as a group of techniques to evaluate MSWM scenarios. The authors also indicates that MCDM methods have gained popularity over the last decade in the area of MSWM, since complex and integrated processes involving distinct dimensions such as environmental, social and economic are very solvable with the use of MCDM. Among the MCDM tools proposed by several authors, TOPSIS has been the most prevalent, because of its ease of use and consistency of results. Additionally, Coban et al. (2018) show that using a single MCDM method to rank alternatives may lead to proposed solutions susceptible to uncertainties.

The uncertainties arise from qualitative parameters, better known as linguistic variables, collected in the study, which are essential for the decision-making. Thus, the fuzzy method proposes the solution by converting linguistic terms into diffuse numbers (Topaloglu et al. 2018). In the present study, two MCDM were used: AHP and fuzzy TOPSIS. The integration of these methods is explained and discussed in the research methodology section.

3 Research Design

This paper presents an evaluation model for the selection of practices and technologies for solid waste management (Fig. 1). The model initially consists of a systemic literature review for the identification and hierarchization of the evaluation criteria. The opinion of experts allows the selection of the most relevant strategy for solid waste management. Practices of sustainable and innovative operations, which associated with a specific context, may support technological and organizational alternatives for the construction of scenarios were identified through the literature review. Multicriteria decision support methods guide the selection of the best practices and technologies for waste management based on the selected criteria and organization of scenarios.

Fig. 1
figure 1

Source The authors

Selection model of practices and technologies for waste management.

Full size image

Using a systemic review of literature based on the method proposed by Ensslin et al. (2010), the Knowledge Development Process-Constructivist (Procknow-C), a ‘Paper Set’ was organized to analyze the content of the papers. It aimed to identify evaluation criteria and characteristics of innovative and sustainable operations for waste management. A questionnaire based on these results was developed (Table 1) and filled out by experts, selecting criteria relevant to the implementation of waste management sustainable and innovative operating systems.

Table 1 Questionnaire for selection of criteria
Full size table

The Analytic Hierarchy Process (AHP) analyzes the criteria judgments, performed by the experts, through the correlation between the criteria, using the classification shown in Table 2.

Table 2 Numerical classification
Full size table

The TOPSIS methodology is based on the principle that there are ‘n’ criteria and ‘m’ alternatives. The selected alternative has a minimum distance from the ideal positive solution and a maximum distance from the ideal negative solution (Gupta and Barua 2018). The ideal positive solution is the solution that maximizes benefit criteria and minimizes cost criteria. The ideal negative solution is the solution that maximizes cost criteria and minimizes benefit criteria (Mesquita 2014). Chen (2000) extended TOPSIS as triangular Fuzzy Numbers (FN). The researcher introduced a vertex method to calculate the distance between two triangular FN. If \(\tilde{x} = ({\text{a}}1,{\text{b}}1,{\text{c}}1)\), \(\tilde{y} = ({\text{a}}2,{\text{b}}2,{\text{c}}2)\) are two triangular FN (1).

$$d\left( {\tilde{x},\tilde{y}} \right)\text{ := } \sqrt {\frac{1}{3}} \left[ {\left( {a1 - a2} \right)^{2} + \left( {b1 - b2} \right)^{2} + \left( {c1 - c2} \right)^{2} } \right]$$
(1)

In the study, TOPSIS Fuzzy procedure was applied as per the instructions by Nǎdǎban et al. (2016). Step 1 is the assignment of rating to the criteria and alternatives, assuming there is a decision group with K members. The Fuzzy classification of the decision makers kth about alternatives Ai w.r.t. criterion Cj is denoted \(\tilde{x}_{\text{ij}}^{\text{k}} = ({\text{a}}^{\text{k}}_{\text{ij}} ,{\text{b}}^{\text{k}}_{\text{ij}} ,{\text{c}}^{\text{k}}_{\text{ij}} )\) and the weight of the criterion Cj is denoted \(\tilde{w}_{kj} = ({\text{w}}_{{{\text{kj}}1}} ,{\text{w}}_{{{\text{kj}}2}} ,{\text{w}}_{{{\text{kj}}3}} )\). In Step 2, the aggregate diffuse classifications for the alternatives (Table 3) and diffuse weights aggregated for the criteria are calculated (Table 2).

Table 3 IVIFS linguistic values for linguistic terms
Full size table

The aggregated fuzzy classification \(\tilde{x}_{\text{ij}} = ({\text{a}}_{ij} ,{\text{b}}_{ij} ,{\text{c}}_{ij} )\) of ith alternative w.r.t. jth. The criterion is obtained as per Eq. (2).

$${\text{a}}_{\text{ij}} = \frac{\hbox{min} }{k}\{ a^{k}_{\text{ij}} \} ,{\text{b}}_{\text{ij}} = \frac{1}{K} \mathop \sum \limits_{k = 1}^{k} b^{k}_{ij} ,{\text{c}}_{\text{ij}} = \frac{\hbox{max} }{k}\{ c^{k}_{\text{ij}} \} .$$
(2)

The aggregate weight fuzzy \(\tilde{w}_{\text{j}} = ({\text{w}}_{j1} ,{\text{w}}_{j2} ,{\text{w}}_{j3} )\) for the criterion Cj is calculated by the formulas:

$${\text{w}}_{{{\text{j}}1}} = \frac{\hbox{min} }{k}\{ w^{k}_{{{\text{j}}1}} \} ,{\text{w}}_{{{\text{j}}2}} = \frac{1}{k}\mathop \sum \limits_{k = 1}^{k} w_{kj2} ,w_{j3} = \frac{\hbox{max} }{k}\{ w^{k}_{{{\text{j}}3}} \} .$$
(3)

The normalized fuzzy decision matrix is calculated in Step 3. The normalized fuzzy decision matrix is \(\tilde{R} = [\tilde{r}_{ij} ]\), (4) and (5).

$$\tilde{r}_{ij} = \left( {\frac{{a_{ij} }}{{c_{j}^{*} }},\frac{{bc_{i} }}{{c_{j}^{*} }},\frac{{cc_{i} }}{{c_{j}^{*} }}} \right){\text{e}}\,{\text{c}}_{\text{j}}^{*} = \frac{\hbox{max} }{i}\{ {\text{c}}_{\text{ij}} \} ({\text{benefit}}\;{\text{criterion}})$$
(4)
$$\tilde{r}_{ij} = \left( {\frac{{a_{j}^{ - } }}{{c_{ij} }} ,\frac{{a_{j}^{ - } }}{{b_{ij} }} ,\frac{{a_{j}^{ - } }}{{a_{ij} }}} \right) {\text{e}}\,{\text{c}}_{\text{j}}^{ - } = \frac{\hbox{min} }{i} \{ a_{ij} \} ({\text{cost}}\;{\text{criterion}})$$
(5)

The weighted normalized fuzzy decision matrix is calculated in Step 4. The weighted normalized fuzzy decision matrix is \(\tilde{V} = (\tilde{v}_{ij} )\), where \(\tilde{v}_{ij} = \tilde{r}_{ij} \times {\text{w}}_{j}\). In Step 5, the Ideal Positive Diffuse Solution (FPIS) (6) and the Ideal Fuzzy Negative Solution (FNIS) (7) are determined. FPIS and FNIS are calculated as per Eqs. (6) and (7):

$$A^{*} = (\tilde{v}_{1}^{*} , \tilde{v}_{2}^{*} , \ldots ,\tilde{v}_{n}^{*} ),\;{\text{where}}\;\tilde{v}_{j}^{*} = \frac{\hbox{max} }{i} \{ v_{ij3} \} ;$$
(6)
$$A^{ - } = (\tilde{v}_{1}^{ - } , \tilde{v}_{2}^{ - } , \ldots ,\tilde{v}_{n}^{ - } ),\;{\text{where}}\;\tilde{v}_{j}^{ - } = \frac{\hbox{min} }{i} \{ v_{ij1} \} ;$$
(7)

The distance from each alternative to FPIS and FNIS is determined (Step 6). Compute the distance from each alternative Ai to FPIS and FNIS, respectively (Eq. 8).

$$d_{i}^{*} = \mathop \sum \limits_{j = 1}^{n} d(\tilde{v}_{ij} ,\tilde{v}_{j}^{*} ),d_{j}^{ - } = \mathop \sum \limits_{j = 1}^{n} d(\tilde{v}_{ij} ,\tilde{v}_{j}^{ - } )$$
(8)

In Step 7, the closeness coefficient CCi for each alternative is determined. For each alternative Ai, the closeness coefficient CCi is calculated as per Eq. (9).

$${\text{CC}}_{\text{i}} = \frac{{d_{i}^{ - } }}{{d_{i}^{ - } + d_{I}^{*} }}$$
(9)

Finally, in Step 8, the alternatives are classified. The alternative with the highest closeness coefficient represents the best alternative. The TOPSIS Fuzzy method was applied using a spreadsheet program.

4 Results

The results are described in four subsections. The first subsection was a systematic review of the literature with content analysis on the methods, evaluation criteria and characteristics related to the topic ‘waste management’. The second involved interviews with experts for the criteria selection and hierarchization. The third subsection presents the contextualization of a real problem and organization of alternatives for a possible solution. The fourth subsection describes the application of the multicriteria method to select the best alternative.

5 Systemic Literature Review

The application of the Procknow-C methodology starts with the definition of keywords. A list of 23 research terms was divided into three research axes (Table 4).

Table 4 Axes and terms used in the research
Full size table

The collection of papers was performed in the Web of Science™ and Scopus® databases through combinations of the search terms and axes. The search was limited to the last 10 years (only papers). Resulting references were inserted in the Mendeley® software; duplicated papers were excluded. A total of 21,040 papers was obtained. The process continued with the analysis of the titles, which resulted in 503 papers aligned with the research theme.

In the scientific recognition analysis, 228 papers with more than five citations passed through the analysis of abstracts and 88 were regarded as aligned with the research theme. The authors of these papers composed a data set of 254 authors. A list of 275 papers with less than five citations went to the reanalysis process; other 22 were selected. The sum of the selected references resulted in 110 items, after being thoroughly read, the set of papers was organized and composed. Then, the paper set was divided into two other sets. The first aim was to identify waste management methodologies and evaluation criteria in different contexts. The second aim was to identify sustainable and innovative waste management systems. The references and evaluation methodologies of the first set can be verified in Table 5.

Table 5 Methods identified in the first set of papers
Full size table

Table 5 shows that the TOPSIS evaluation method was the most used as seen in Arıkan et al. (2017), Coban et al. (2018), Hlatka et al. (2018), Jovanovic et al. (2016), Mir et al. (2016), Pires et al. (2011) and Topaloglu et al. (2018). The criteria were identified (Frame 1). Nevertheless, due to different contexts for different indicators, it was decided to group similar criteria. For instance, Jovanovic et al. (2016) uses particulate matter, emission of gases (CH4, CO2 and N2O); Stefanović et al. (2016) used and classified emissions of greenhouse gases (CO2) and emissions of acid gases (NOx and SO2) as environmental indicators. In the current study, all of those are considered in the atmospheric emissions criterion. The second set of papers encompassed 13 references (Table 6). Content analysis aimed to identify solid waste management sustainable and innovative systems.

Table 6 Sustainable and innovative features identified in the second set of papers
Full size table

As shown in Table 6, the IoT information technologies are the most prevalent features due to the number of papers that address them within the set. They have been studied by Díaz-Díaz et al. (2017), Elia et al. (2015, 2018), Misra et al. (2018), Wen et al. (2017) and Yerraboina et al. (2018).

6 Selection and Hierarchization of Criteria

Three experts were selected. All of them graduated in Environmental Engineering; one has a master’s degree in Building Engineering and is a lecturer in the subject of solid waste management. The other two have master’s degrees in Environmental Engineering, with experience in municipal waste management. The experts were requested to indicate relevant criteria for the implementation and operations of sustainable and innovative waste management systems, using a questionnaire with closed questions. Each of the experts received a questionnaire to evaluate the criteria, individually and without any consultation with the other interviewees. The 11 selected criteria (Table 7) are observed in the literature and considered relevant by the experts. For the application of multicriteria methods, the selected criteria were divided into three categories (environmental, economic and social).

Table 7 Indicators observed in the literature and selected by the experts
Full size table

The experts, according to the AHP methodology and Saaty’s classification (Table 2), performed peer comparison. Table 8 shows the results of the weights for each criterion after the judgement by the experts through the AHP method.

Table 8 Weights of the criteria
Full size table

7 Contextualization of the Solid Waste Management in a Municipality in the Western Region of the Brazilian State of Santa Catarina to Build Possible Scenarios

According to the Environmental Department of a municipality located in the Western region of the Santa Catarina state, the city does not have landfills or its own machinery for garbage collection. This service is the responsibility of a private company. The municipality pays a fixed amount for the collection of recyclable waste and a variable rate, according to the amount of residue. Decreasing the amount of recyclable organic matter mixed with residue decreases the value of the variable rate to be paid. Thus, the issue can be summarized in the following question: What are the systems alternatives for the optimization of solid waste management?

Considering the results observed in the second set of papers in the portfolio (Table 6), the selected indicators and context, seven possible scenarios were developed. They are used to compare and apply the multicriteria methodology and select a possible ideal scenario. These scenarios are described in Table 9.

Table 9 Description of scenarios
Full size table

8 Analysis Using the TOPSIS Fuzzy Methodology

Through the analysis of the three experts, seven scenarios or alternatives for the collection of recyclable solid waste were evaluated. The context of the city was studied, considering the 11 selected criteria and their respective weights. It should be indicated that the classifications, environmental, economic and social, have equal weights in the study. Hence, the sum of the environmental criteria has the same weight as the sum of the social criteria, which in turn is equal to the sum of the economic criteria.

For the application of the TOPSIS Fuzzy methodology, the experts filled out a spreadsheet with linguistic variables (Table 3), relating the scenarios to the criteria. Tables 10, 11 and 12 show the linguistic judgments regarding the performance of the alternatives.

Table 10 Expert assessment matrix 1 on the performance of alternatives
Full size table
Table 11 Expert assessment matrix 2 on the performance of alternatives
Full size table
Table 12 Expert assessment matrix 3 on the performance of alternatives
Full size table

The values presented in the tables were converted into fuzzy numbers, and the normalized results were then multiplied by the respective weight of each criterion. The ideal positive and negative solutions were calculated according to (4) and (5). Using the method according to (6), (7), and (8), the distances between the values ​​and the ideal positive solutions (FPIS) and the negative (FNIS) were determined. Using (9), the closeness coefficient (CCi) was calculated. Table 13 lists the rankings of the alternatives and their respective Overall Performance (OP) according to the judgment of each expert. Table 13 shows a final ranking, that is, the result of a weighted average of the results.

Table 13 Result with the ranking of alternatives
Full size table

It is possible to see that the Alternative A7, that is, recyclable garbage collection points using IoT trash cans in the center of the municipality, the center of the neighborhoods, in strategic locations, with a discount in property taxes or other similar financial incentives, has the highest ranking positions. This means that the collection of recyclable solid waste comes closest to the ideal positive solution; it is also the furthest from the ideal negative solution.

9 Conclusion

The study achieved its main goal, i.e., structuring a model to evaluate the implementation of MSWM sustainable operations and technologies. Based on these results, it is not yet possible to ensure that technological and innovative systems are a final solution. However, it is clear, considering the alternatives selected by experts, current waste management methods are not the more adequate. The interpretation of the indicators by experts is regarded as a limitation related to this investigation. The experts selected the criteria based on their experiences. As a future agenda is possible to move beyond the replication of the search, choosing other criteria. In organizing scenarios and selecting possible alternatives, the MCDM approach should also consider the managers’ opinions in the criteria selection and hierarchy.