Waste Management Analysis in Developing Countries through Unsupervised Classification of Mixed Data

Abstract

The increase in global population and the improvement of living standards in developing countries has resulted in higher solid waste generation. Solid waste management increasingly represents a challenge, but it might also be an opportunity for the municipal authorities of these countries. To this end, the awareness of a variety of factors related to waste management and an efficacious in-depth analysis of them might prove to be particularly significant. For this purpose, and since data are both qualitative and quantitative, a cluster analysis specific for mixed data has been implemented on the dataset. The analysis allows us to distinguish two well-defined groups. The first one is poorer, less developed, and urbanized, with a consequent lower life expectancy of inhabitants. Consequently, it registers lower waste generation and lower $C{O}_2$ emissions. Surprisingly, it is more engaged in recycling and in awareness campaigns related to it. Since the cluster discrimination between the two groups is well defined, the second cluster registers the opposite tendency for all the analyzed variables. In conclusion, this kind of analysis offers a potential pathway for academics to work with policy-makers in moving toward the realization of waste management policies tailored to the local context.

1. Introduction

In recent years, the term circular economy has gained much attention. It refers to a system of production and consumption providing minimal losses of materials and energy through extensive reuse, recycling, and recovery Haupt et al. (2017). In other words, it is an economic system for which is essential to recycle materials from waste in order “to close the cycle”.

Furthermore, the increase in global population and the improvement of living standards in developing countries has resulted in higher solid waste generation in the areas under investigation Abarca-Guerrero et al. (2013); Minghua et al. (2009). Consequently, nowadays waste management is at the center of a very lively debate Fernández-González et al. (2017); Schneider et al. (2017).

Waste disposal should be gradually eliminated, and where this is not possible, it should be monitored in order to be safe for human health and the environment. Consequently, solid waste management increasingly represents a challenge since municipalities must provide an efficient system to their population. However, they often have to struggle with complexity and with lack of both organization and financial resources Burntley (2007).

Among other things, sustainable waste management is able to reduce the incidence of health problems, the emission of greenhouse gases, and the deterioration of landscape, water, and air caused by landfilling Cucchiella et al. (2017). However, it not only represents a contribution to environmental protection, but it also pays off economically Nelles et al. (2016): waste management practices, indeed, can be cost-saving and generate revenue opportunities Romero-Hernández and Romero (2018).

In fact, waste can be a useful source of raw materials and energy, too. Metals, glass, and textiles have long been collected and put to new use; for example, the extraction of nickel and cobalt from raw materials, as well as from waste, is strategically important for industry and society Komnitsas et al. (2019). Waste can be turned into energy too, enabling the value of products, materials, and resources to be maintained on the market for as long as possible, minimizing waste and resource use in the wake of the objectives of circular economy Malinauskaite et al. (2017).

For all these reasons, and since the circular economy is an important issue for the future and the competitiveness of businesses Garcia-Muiña et al. (2018), solid waste management might represent an opportunity for the municipal authorities of developing countries, especially those characterized by a low-income Minghua et al. (2009).

In conclusion, in a circular economy wastes are considered a resource, especially from an economic point of view, consequently attracting an increasing number of industrial actors, policy-makers, and researchers Cucchiella et al. (2017). Thus, in the last years, a large number of studies have tried to detect factors influencing waste management systems in developing countries.

Therefore, since the awareness of a variety of factors related to waste management might prove to be particularly significant Ghinea and Gavrilescu (2019); Zeller et al. (2019); Zohoori and Ghani (2017), this paper aims to analyze some of them in the developing countries involved in the study.

From our analysis, it may be concluded that a study of this kind offers a potential pathway for academics to work with policy-makers in moving toward the realization of waste management policies tailored to the local context.

2. Materials and Methods

Real data often consists of mixed variables, that is, both continuous and categorical ones; an example is provided by the dataset analyzed in this paper, produced by Abarca-Guerrero (2014), the variables of which are described in

Table 1. Description of all of the analyzed variables.

Variable	Type	Description
Urban population	Continuous	% of urban population
Waste generation	Continuous	Waste generation rate (kg/capita/day)
CO${}_2$	Continuous	$C{O}_2$-emission/capita
		in percentage of disposable income
GDP	Continuous	Gross domestic product/capita/year
Life Expectancy	Continuous	Life expectancy at birth (years)
Municipality campaigns	Continuous	Recycling awareness campaigns
		supported by the municipality:
		1 = yes
		2 = no
Waste collection combination	Categorical	Waste collection combination:
		1 = no collection
		2 = animal power
		3 = man power
		4 = animal+man
		5 = mechanized
Climate	Categorical	Climate:
		1 = equatorial
		2 = arid
		3 = warm temperature
		4 = snow
Precipitation	Categorical	Precipitation:
		1 = desert
		2 = steppe
		3 = fully humid
		4 = summer dry
		5 = winter dry
		6 = monsoonal
Recyclable-material-buying companies	Categorical	Companies buying recyclable materials
		in the surroundings of the city:
		1 = none
		2 = few
		3 = some
		4 = many
		5 = very many
Recycling companies	Categorical	Recycling companies
		in the surroundings of the city:
		1 = none
		2 = few
		3 = some
		4 = many
		5 = very many

. Traditionally, cluster analysis has only focused on datasets composed of a single type of variable (all quantitative or all qualitative). For this reason, researchers dealing with mixed data usually convert them into a single data type, transform the categorical variables into binary ones and consequently apply methods for numeric variables, or transform continuous variables into categorical ones Dougherty et al. (1995); Ichino and Yaguchi (1994).

Indeed, clustering methods specific to mixed data are less encountered in the literature.

Some traditional methods are:

• data pre-processing, that is, all variables are converted to the same scale, either numerical to categorical or vice-versa;
• distance measures specifically developed for mixed datasets.

With regards to data pre-processing, these algorithms are essentially created for purely categorical attributes, although they have also been applied to mixed data after a transformation of numerical attributes to categorical ones (discretization). In general, these kinds of algorithms can be applied to mixed data through a discretization process that may, nevertheless, produce a loss of important information Caruso et al. (2018).

One example is represented by the dummy coding of all categorical variables. But this increases the dataset’s dimensionality, representing a problem when the number of categorical variables simultaneously increases with the size of the data. Another disadvantage is that any semantic similarity in the original dataset is lost in the transformed one. Finally, coding strategies imply a difficult choice of weights representing categorical attributes Foss et al. (2016).

An alternative to recoding categorical or continuous variables is to use a dissimilarity measure, taking into account the different types of data Caruso (2019). A common approach is to use the Gower distance Gower (1971).

2.1. Clustering Mixed Data

Let $X=\left\{x_1,x_2,\dots ,x_n\right\}$ denote a set of n objects and $x_i=\left[x_{i1},x_{i2},\dots ,x_{iL}\right]$ indicate an object constituted by L variables. Since the L variables of the considered dataset are both continuous and categorical, it is possible to write $L=Q+C$, where Q corresponds to the number of numeric variables and C to the number of categorical ones. $\mathcal{C}=\{l_1^{\mathcal{C}},\dots l_C^{\mathcal{C}}\}$ is a subset identifying the qualitative variables and $\mathcal{Q}=\{l_1^{\mathcal{Q}},\dots l_Q^{\mathcal{Q}}\}$ is a subset denoting the quantitative ones. The aim of clustering is to assign the n objects contained in X to K separate clusters. When clustering mixed datasets, the main problem is to determine how close or how far apart objects are from each other.

There are different approaches presenting different ways to combine distance measures for numerical variables and distance measures for categorical ones into a single cost function Caruso (2019); Caruso et al. (2019, in printb); Everitt (1974); Huang (1997).

2.2. The Huang Method

Huang (1998) presented a so-called K-prototypes algorithm, which is based on the K-means method but overcomes its quantitative data limitation, preserving, at the same time, its efficiency. The algorithm groups the objects in clusters against k prototypes. The updates occur in a dynamical manner, so as to minimize the following objective function:

$$E=\sum _{k=1}^K\sum _{i=1}^n{u}_{ik}{\Phi }_H(x_i,V_k),$$

(1)

where $u_{ik}$ is an element of a partition matrix $U_{n\times k}$, and ${\Phi }_H(x_i,V_k)$ is a dissimilarity measure for mixed data between the objects $x_i$ and $V_k$.

$V_k=\left[v_{k1},v_{k2},\dots ,v_{kL}\right]$ is the prototype or representative vector for cluster k. U represents a hard partition matrix, where $u_{ik}\in \left\{0,1\right\}$, and $u_{ik}=1$ if $x_i$ is allocated to cluster k.

The Huang dissimilarity measure for mixed data is defined as

$${\Phi }_H(x_i,V_k)=\sum _{l\in \mathcal{Q}}{(x_{il}-v_{kl})}^2+{\gamma }_k\sum _{l\in \mathcal{C}}\delta (x_{il},v_{kl}),$$

(2)

where the first term is the squared Euclidean distance, whereas the second one is defined as $\delta (r,t)=0$ for $r=t$ and $\delta (r,t)=1$ for $r\ne t$. ${\gamma }_k$ is a weight for categorical variables in cluster k.

The internal term in Equation (1) can be defined as $E_k={\displaystyle \sum _{i=1}^n{u}_{ik}S(x_i,V_k)}$. It measures the total dissimilarity of objects in cluster k from their prototype $V_k$. The quantity $E_k$ could be considered as the total cost of allocating the objects $x_i(i\in C_k)$ to cluster k.

This term may be rewritten as

$$\begin{array}{cc}\hfill E_k& =\sum _{i=1}^n{u}_{ik}\sum _{l\in \mathcal{Q}}{(x_{il}-v_{kl})}^2+{\gamma }_k\sum _{i=1}^n{u}_{ik}\sum _{l\in \mathcal{C}}\delta (x_{il},v_{kl})\hfill \\ & =E_k^{\mathcal{Q}}+{\gamma }_k{E}_k^{\mathcal{C}},\hfill \end{array}$$

(3)

where $E_k^{\mathcal{Q}}$ and $E_k^{\mathcal{C}}$ represent the dissimilarity of the objects in cluster k for the quantitative and the qualitative variables, respectively. In order to minimize these two components, let $V_k^{\mathcal{Q}}$ and $V_k^{\mathcal{C}}$ be the prototypes for cluster k for the numerical and categorical variables, respectively.

$E_k^{\mathcal{Q}}$ is minimized with the usual update of the K-means algorithm for continuous variables. That is, the generic component of $V_k^{\mathcal{Q}}$ is the arithmetic mean:

$$v_{kl}=\frac{1}{n_k}\sum _{i=1}^n{u}_{ik}{x}_{il}l\in \mathcal{C},$$

(4)

where $n_k$ is the number of objects in cluster k. Let ${\mathcal{W}}_l=\{w_{l,1},w_{l,2},\dots ,w_{l,m_l}\}$ be the set enclosing the distinct values of the l-th categorical variable, and let $p_l\left(w_{l,j}\right|k)$ be the probability that value $w_{l,j}$ is observed in cluster k.

It is possible to rewrite $E_k^{\mathcal{C}}$ in (3) as

$$E_k^{\mathcal{C}}=\sum _{l\in \mathcal{C}}{n}_k\left[1-p(v_{kl}\in {\mathcal{W}}_l|k)\right].$$

(5)

In Equation (5), $E_k^{\mathcal{C}}$ is minimized by selecting the categorical values of the prototype $W_k^{\mathcal{C}}$, such that $p(v_{kl}\in {\mathcal{W}}_l|k)\ge p(w_{l_j}\in {\mathcal{W}}_l|k)$ for $v_{kl}\ne w_{l,j}$ for all categorical variables.

On the basis of the Huang algorithm, by minimizing (1), we implemented a cluster analysis with a number of clusters equal to $K=2$. This choice was made based on the Silhouette index Rousseeuw (1987); since higher values corresponds to better results, the resultant (optimal) maximum value precisely corresponds to 2.

3. Results

3.1. The Dataset

The dataset Abarca-Guerrero (2014) used in this application has been extracted from the data archive of the “4TU.Centre for Research Data” in the Netherlands, and it regards the period of 1985–2011. It contains information on factors influencing the municipal waste management system in 22 developing countries; each of these is associated with more than one observation.

The dataset considers some key factors affecting waste management systems, in particular the country performance in terms of public health (life expectancy at birth), economy (gross domestic product/capita/year), and environment ($C{O}_2$-emissions/capita).

Other general parameters characterizing the countries are the urban population, the kind of climate, and precipitation. Furthermore, waste-specific parameters have been considered: the waste generation rate (kg/capita/day) and the sophistication of waste collection. The latter can be articulated as 1 = no organized collection of solid waste; 2 = collection based on manpower only; 3 = collection based on both manpower and draught animal; 4 = collection based on motorized transport but no compactor used; and 5 = collection based on motorized transport and compactor used. Other parameters include the existence of a recycling culture and the presence of municipality awareness campaigns, of recyclable-material-buying companies, and of recycling companies, the latter two specifically in the surroundings of the city. In conclusion, the analyzed data consist of a selection of 11 variables (6 continuous and 5 categorical), described in detail in Table 1, for a total of 50 observations Abarca-Guerrero (2014).

3.1.1. Internal Indexes

Since the ground truth (i.e., an empirical evidence) Han et al. (2011) is not given for this dataset, it is not possible to compute the external indexes. The internal ones are shown below.

We compared several methods for clustering mixed data types, namely those of: Huang (1997), Ahmad and Dey (2007), and Cheung and Jia (2013).

The relevant validity of cluster results was evaluated through the above-mentioned indexes and the Huang method proved to be the one yielding the best results, namely the highest values of the Calinski–Harabasz index (CH) Calinski and Harabasz (1974) and of the Silhouette index (SHI) Rousseeuw (1987), both computed on quantitative variables. The results provided by these methods are shown in

Table 2. Internal indexes. CH—Calinski–Harabasz index; SHI—Silhouette index.

Method	CH	SHI
Huang	13.23	0.21
Ahmad & Dey	10.15	0.209
Cheung & Jia	10.68	0.189

.

3.1.2. Analysis of Quantitative Variables

First of all, we provide the descriptive analysis of the quantitative variables used: the mean, the standard deviation, and the values corresponding to the 1st, the 2nd, and the 3rd quartiles, as shown in

Table 3. Descriptive statistics of quantitative variables.

Variables	Mean	Standard Deviation	${\mathit{Q}}_1$	${\mathit{Q}}_2$	${\mathit{Q}}_3$
%UrbPop	51.12	19.35	33.50	57	65.75
WasGen	0.61	0.28	0.41	0.50	0.82
CO2	2.28	2.51	0.80	1.40	3.50
GDP	3825	6747.90	1069	2349	4469
LifeExp	68.46	8.27	66	71	73

.

Table 4. Mean values of quantitative variables for each cluster.

Cluster	Size	%UrbPop	WasGen	CO2	GDP	LifeExp
1	29	42.24	0.57	1.72	2096.38	66.24
2	21	63.38	0.68	3.06	6212.29	71.52

displays, for each cluster, the mean value of the analyzed quantitative attributes. With regards to the variable “waste generation”, the two groups have a weak cluster structure, that is clusters values are very similar between them, whereas they have a strongest structure with regards to the variable “GDP”. With regards to the “percentage of urban population”, the relevant overall mean equals 51.12. Thus, the first cluster mean is lower than the average one, whereas the second one is higher. For what concerns the “waste generation rate”, instead, the overall mean corresponds to 0.61. In this case as well, the first cluster mean is lower than the overall one, whereas the second one is higher. However, in this case the separation between clusters is less marked. With regards to the “$C{O}_2$ emissions”, the overall mean value is 2.28, so the first cluster mean is lower than the overall one, whereas the second one is higher. For what concerns the “GDP”, the overall mean value is equal to 3825, so the first cluster mean is lower than the average, whereas the second one is significantly higher. With regards to “life expectancy”, the overall mean equals 68.46; in this case as well, the first mean is lower and the second one is higher. In summary, for all of the analyzed quantitative variables, the first cluster is below the overall mean whereas the second cluster overcomes it.

3.1.3. Analysis of Qualitative Variables

Table 5. Overall distribution of the categorical variable “municipality campaigns”.

Municipality Campaigns	Distribution
Yes	0.68
No	0.32

shows the overall distribution of the variable “municipality campaigns”. The prevailing modality is represented by the presence of recycling awareness campaigns supported by the municipality.

Table 6. Distribution of the categorical variable “municipality campaigns”.

Municipality Campaigns	Clusters
	1	2
Yes	0.90	0.38
No	0.10	0.62

, instead, shows more in detail the distribution of the categorical variable “municipality campaigns” for each of the two clusters. The first one, which has the strongest cluster structure, is characterized by the presence of recycling awareness campaigns supported by the municipality, whereas the second cluster is characterized by their absence.

Table 7. Overall distribution of the categorical variable “waste collection combination”.

Waste Collection Combination	Distribution
No collection	0.02
Animal power	0.42
Man power	0.12
Animal + man	0.12
Mechanized	0.32

shows the overall distribution of the variable “waste collection combination”. The prevailing modality is represented by “animal power”, followed by “mechanized methods”.

In

Table 8. Distribution of the categorical variable “waste collection combination”.

Waste Collection Combination	Clusters
	1	2
No collection	0.00	0.05
Animal power	0.66	0.10
Man power	0.17	0.05
Animal + man	0.00	0.29
Mechanized	0.17	0.52

, the two clusters are clearly outlined. In the first, the prevailing modality is represented by “animal power”, whereas in the second one it corresponds to “mechanized” methods. By comparing the obtained clusters with the relevant overall distribution, the first cluster registers a higher use of methods based on “animal power” than in the overall distribution. The same applies to the modality “mechanized” methods in the second cluster.

Table 9. Overall distribution of the categorical variable “climate”.

Climate	Distribution
Arid	0.18
Equatorial	0.72
Snow	0.02
Warm	0.08

shows the overall distribution of the variable “climate”. The modality with the overwhelming majority is “equatorial”.

Table 10. Distribution of the categorical variable “climate”.

Climate	Clusters
	1	2
Arid	0.17	0.19
Equatorial	0.76	0.67
Snow	0.03	0.00
Warm	0.03	0.14

shows that in both clusters, the prevailing modality is represented by the modality “equatorial”, in the wake of the overall distribution.

Table 11. Overall distribution of the categorical variable “precipitation”.

Precipitation	Distribution
Desert	0.06
Fully humid	0.36
Monsoonal	0.24
Steppe	0.08
Summer dry	0.06
Winter dry	0.20

, instead, shows the overall distribution of the variable “precipitation”. The prevailing modality is represented by “fully humid”.

In

Table 12. Distribution of the categorical variable “precipitation”.

Precipitation	Clusters
	1	2
Desert	0.10	0.00
Fully humid	0.21	0.57
Monsoonal	0.41	0.00
Steppe	0.07	0.10
Summer dry	0.07	0.05
Winter dry	0.14	0.29

, the distribution of the two clusters is shown. The first cluster is characterized by the prevalence of the modality “monsoonal”, whereas in the second cluster the most frequent modality is “fully humid”, exactly like in the overall distribution.

Table 13. Overall distribution of the categorical variable “recyclable-material-buying companies”.

Recyclable-Material-Buying Companies	Distribution
none	0.28
few	0.18
some	0.38
many	0.16
very many	0.00

shows the overall distribution of the variable “recyclable-material-buying companies”. The prevailing modality is represented by “some”.

In

Table 14. Distribution of the categorical variable “recyclable-material-buying companies”.

Recyclable-Material-Buying Companies	Clusters
	1	2
none	0.00	0.67
few	0.10	0.29
some	0.62	0.05
many	0.28	0.00
very many	0.00	0.00

, the distribution of the two clusters is shown. The first cluster is characterized by a prevalence of the modality “some”, exactly as in the overall distribution, whereas in the second cluster the most frequent modality is “none”.

Table 15. Overall distribution of the categorical variable “recycling companies in the surroundings of the city”.

Recycling Companies	Distribution
none	0.34
few	0.30
some	0.24
many	0.10
very many	0.02

shows the overall distribution of the variable “recycling companies in the surroundings of the city”. The prevailing modality is represented by “none”.

In

Table 16. Distribution of the categorical variable “recycling companies in the surroundings of the city”.

Recycling Companies	Clusters
	1	2
none	0.14	0.62
few	0.31	0.29
some	0.38	0.05
many	0.17	0.00
very many	0.00	0.00

, the distribution of the two clusters is shown. The first cluster has a prevalence of the modality “some”, whereas in the second cluster the most frequent modality is “none”, exactly as in the overall distribution.

4. Discussion

On the basis of our analysis of quantitative variables, it appears that the first cluster is characterized by a lower percentage of urban population, lower levels of GDP, and a lower life expectancy. As a consequence of limited urbanization and greater poverty, this group registers lower rates of waste generation and of $C{O}_2$ emissions. Since the cluster discrimination between the two groups is well defined, the second cluster registers the opposite tendency for all of the above-mentioned variables, namely higher levels of GDP and a stronger percentage of urban population, with a consequently higher life expectancy. The higher urbanization corresponds to higher levels of waste generation and of $C{O}_2$ emissions. In more detail, in order to better describe the distribution of the quantitative variables analyzed, each of these has been represented through a box plot Cleveland (1993).

With regards to the percentage of urban population (Figure 1), the overall median of the distribution equals 57.00, whereas the overall mean is 51.12; the mean of the first cluster is lower than this, whereas the one associated with the second cluster is higher. In the first cluster, the interquartile distance is much higher than in the second one, denoting a greater dispersion of the $50\%$ most central observations around the median. On the other hand, since the interquartile distance of the second cluster is lower, the $50\%$ most central observations are highly concentrated around the median. Furthermore, since in the first cluster the distances between each quartile and the median are quite different from one another, the distribution is asymmetric. In the second cluster, instead, the distances are more similar between these, denoting a lower asymmetry of the distribution.

For what concerns the waste generation (Figure 2), the overall median of the distribution equals 0.50, whereas the overall mean is 0.61; the mean of the first cluster is lower than this, whereas the one associated with the second cluster is higher. In the first cluster, the interquartile distance is lower than in the second one. Thus, in the first group there is a low dispersion of the $50\%$ most central observations around the median. On the other hand, since the interquartile distance of the second cluster is higher, the $50\%$ most central observations are less concentrated around the median. Furthermore, in the first cluster the two distances are quite different from one another, denoting a very asymmetric distribution, whereas in the second group these distances are more similar, resulting in a slightly lower asymmetry of the distribution.

With regards to the $C{O}_2$ emissions (Figure 3), the overall median of the distribution equals 1.40, whereas the overall mean is 2.28; the mean of the first cluster is lower than this, whereas the one associated to the second cluster is higher. In the first cluster, the interquartile distance is lower than in the second one. Thus in the first group, the dispersion of the $50\%$ most central observations around the median is lower. On the other hand, since the interquartile distance of the second cluster is higher, the $50\%$ most central observations are less concentrated around the median. Furthermore, in the first cluster the two distances are very similar to one another, denoting a symmetric distribution, whereas in the second cluster they are less similar, indicating the asymmetry of the distribution.

With regards to the GDP (Figure 4), the overall median of the distribution equals 2349, whereas the mean is 3825; the median of the first cluster is lower than this, whereas the one associated with the second cluster is higher. In the first cluster, the interquartile distance is lower than in the second one. Thus, in the first group the dispersion of the the $50\%$ most central observations around the median is lower. Since the interquartile distance of the second cluster is higher, instead, the $50\%$ most central observations are less concentrated around the median. Furthermore, in the first cluster the two interquartile distances are quite different from one another, so the distribution is asymmetric. In the second cluster, instead, they are more similar, denoting the lower asymmetry of the distribution.

For what concerns life expectancy (Figure 5), the overall median of the distribution equals 71.00, whereas the overall mean is 68.46, thus the mean of the first cluster is lower than this, whereas the one associated with the second cluster is higher. In the first group, the interquartile distance is higher than in the second one. Thus, in the first group the dispersion of the $50\%$ most central observations around the median is higher. On the other hand, since the interquartile distance of the second cluster is lower, the $50\%$ most central observations are highly concentrated around the median. Furthermore, in the first cluster the two distances are quite different from one another, so the distribution is asymmetric, whereas in the second group the distances are more similar, denoting the lower asymmetry of the distribution.

With regards to qualitative variables, instead, the first cluster is characterized by the overwhelming majority of recycling awareness campaigns supported by the municipality, the waste is mainly collected through animal power, and there are some recyclable-material-buying companies and some recycling companies in the surrounding areas of the cities. The countries falling under this category are mostly characterized by monsoonal precipitation and are the following: Ethiopia, Sri Lanka, Thailand, China, Peru, Tanzania, India, Bangladesh, Nepal, Malawi, Zambia, Nicaragua, Kenya, and the Philippines.

The second cluster, instead, is mainly characterized by the absence of recycling awareness campaigns supported by the municipality, the waste is mainly collected through mechanized tools, but it is mostly characterized by the absence of recyclable-material-buying companies and of recycling companies in the surrounding areas of the cities. Furthermore, it is characterized by the prevalence of a fully humid climate. The countries falling into this cluster are Turkey, Suriname, Costa Rica, Ecuador, Pakistan, and Bhutan, whereas Indonesia and South Africa are in the overlapping area of the two clusters.

5. Conclusions

Since nowadays more and more applications are based on datasets composed of mixed data, there is an ever-growing interest in cluster analysis. Due to their characteristics, traditional methods are unable to capture, store, manage, and analyze these datasets. A cluster analysis implemented on such a dataset has huge potential; however, most clustering algorithms are designed to exclusively handle one type of data at a time, being unable to analyze mixed data simultaneously. The use of cluster analysis for mixed data represents an element of innovation, especially in the waste management sector, since until now only the traditional cluster analysis has been applied in this framework.

Certainly, the research in this area is far from being complete. There are quite a few methods in the literature, but further advancements in this field are needed Caruso (2019). Furthermore, in the wake of this work, future research will be focused on the development of new cluster analysis techniques for mixed data and on the consequent creation of dedicated software packages, also with the aim of widening the number of potential users of this method Caruso et al. (2019).

The basis for future developments will take into consideration the results yielded from the applications described in Section 3 and from an interesting insight provided by the work of Diday and Govaert Diday and Govaert (1977). They propose an adaptive clustering that consists in a dynamic procedure and is useful for calibrating the weights of variables used in the clustering.

Usually, indeed, all of the variables participate in the cluster analysis with the same importance, but since some of them may be more discriminant than others, or better characterize a cluster, there are some ways to correctly consider their different values Irpino et al. (2016).

One strategy consists in assigning a weight to each variable in advance, on the basis of a prior knowledge, and then performing a cluster analysis; a future development could consist in computing the weights for each variable in an automatic way Caruso (2019). In this context, Diday and Govert Diday and Govaert (1977) proposed using an adaptive distance when clustering real data. It is necessary to introduce a weighting step in the optimization process, generating a set of weights; each of these corresponds to a variable and measures its importance in the cluster analysis. While Diday and Govert’s proposal is only focused on quantitative variables, a further advancement could be to extend it to both quantitative and qualitative data.

Furthermore, an additional input for our future research could be to extend this kind of analysis to our recent study Caruso et al. (in printa). Moreover, since clustering is also at the center of a very lively debate Di Battista et al. (2016), Di Battista and Fortuna (2016), Fortuna and Maturo (2018), Fortuna et al. (2018) in the functional framework, a further and interesting possible development could be to also consider this kind of approach in our future research.