No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Development of machine learning - based models to forecast solid waste generation in residential areas: A case study from Vietnam
Abstract
The main aim of this work was to compare six machine learning (ML) - based models to predict the municipal solid waste (MSW) generation from selected residential areas of Vietnam. The input data include eight variables that cover the economy, demography, consumption and waste generation characteristics of the study area. The model simulation results showed that the urban population, average monthly consumption expenditure, and total retail sales were the most influential variables for MSW generation. Among the ML models, the random forest (RF), and k-nearest neighbor (KNN) algorithms show good predictive ability of the training data (80% of the data), with an R² value > 0.96 and a mean absolute error (MAE) of 121.5–125.0 for the testing data (20% of the data). The developed ML models provided reliable forecasting of the data on MSW generation that will help in the planning, design and implementation of an integrated solid waste management action plan for Vietnam. The limitations of this work may be the heterogeneity of the dataset, such as the lack of data from lower administrative units in the country. In such cases, the predictive ML algorithm can be updated and re-trained in the future when the reliable data is added.
... In waste generation studies within the WM field, the most commonly used ML algorithms are ANN, DT, KNN, RF, and SVM [32,33]. These algorithms typically demonstrate exceptional performance in supervised learning tasks, handling non-linear data, identifying faults in datasets, and managing heterogeneous output parameters and numerical target variables [34]. Accordingly, many researchers often utilize these algorithms (i.e., ANN, DT, KNN, RF, and SVM), which continue to be widely employed, and these algorithms should be prioritized during the development of prediction models. ...
... Accordingly, many researchers often utilize these algorithms (i.e., ANN, DT, KNN, RF, and SVM), which continue to be widely employed, and these algorithms should be prioritized during the development of prediction models. In addition, the LR algorithm is straightforward and facilitates easy interpretation of results and is thus a recurrent choice for model development in the WM domain [32,34]. Ensemble algorithms including RF, offer benefits such as improved prediction performance and enhanced generalization results compared to individual learning algorithms [35,36]. ...
... As a supervised learning model for tackling classification and regression problems, the DT algorithm is used for efficiently extracting a set of rules from unfamiliar data [34] and offers numerous benefits. For example, the algorithm is particularly advantageous in terms of intuitively interpreting results, reducing computational expenses, and managing specific property data independent of omitted values. ...
A suitable waste management strategy is crucial for a sustainable and efficient circular economy in the construction sector, and requires precise data on the volume of demolition waste (DW) gen-erated. Therefore, we developed an optimal machine learning (ML) model to forecast the quantity of recycling and landfill waste based on the characteristics of DW. A dataset comprising infor-mation on the characteristics of 150 buildings, demolition equipment utilized, and volume of five waste types generated (i.e., recyclable mineral, recyclable combustible, landfill specified, and landfill mix waste, and recyclable minerals) was constructed. ML models were developed to predict the quantities of such waste. Artificial neural network, decision tree, gradient boosting machine, k-nearest neighbors, linear regression, random forest (RF), and support vector regression were applied, and optimal models were derived via hyperparameter tuning. The RF model demonstrated superior performance. In both validation and test phases, the “recyclable mineral waste” and “recyclable combustible waste” models achieved accuracies of of 0.987 and 0.972, re-spectively. The “recyclable metals” and “landfill specified waste” models achieved accuracies of 0.953 and 0.858 or higher, respectively. Moreover, the “landfill mix waste” model exhibited an accuracy of 0.984 or higher. SHapley Additive exPlanations analysis highlighted floor area as the primary input variable influencing model performance. The type of equipment employed in demolition emerged as another crucial input variable impacting the volume of recycling and landfill wastes generated. The developed model can provide precise data on waste management, thereby facilitating the decision-making process for industry professionals.
... In waste-generation studies within the WM field, the most commonly used ML algorithms are ANN, DT, KNN, RF, and SVM [33,34]. These algorithms typically demonstrate an exceptional performance in supervised learning tasks, handling non-linear data, identifying faults in datasets, and managing heterogeneous output parameters and numerical target variables [35]. Accordingly, many researchers often utilize these algorithms (i.e., ANN, DT, KNN, RF, and SVM), which continue to be widely employed and should be prioritized in developing prediction models. ...
... Accordingly, many researchers often utilize these algorithms (i.e., ANN, DT, KNN, RF, and SVM), which continue to be widely employed and should be prioritized in developing prediction models. Additionally, the LR algorithm is straightforward and facilitates the interpretation of results; it is thus a recurrent choice for model development in the WM domain [33,35]. Ensemble algorithms, including RF, offer benefits such as an improved prediction performance and enhanced generalization results compared to individual learning algorithms [36,37]. ...
... As a supervised learning model for tackling classification and regression problems, the DT algorithm is used for efficiently extracting a set of rules from unfamiliar data [35], and it offers numerous benefits; however, it can be vulnerable to the problem of overfitting with data [33]. To construct an effective DT model with an optimal performance, devising a model that avoids overfitting is crucial. ...
A suitable waste-management strategy is crucial for a sustainable and efficient circular economy in the construction sector, and it requires precise data on the volume of demolition waste (DW) generated. Therefore, we developed an optimal machine learning model to forecast the quantity of recycling and landfill waste based on the characteristics of DW. We constructed a dataset comprising information on the characteristics of 150 buildings, demolition equipment utilized, and volume of five waste types generated (i.e., recyclable mineral, recyclable combustible, landfill specified, landfill mix waste, and recyclable minerals). We applied an artificial neural network, decision tree, gradient boosting machine, k-nearest neighbors, linear regression, random forest, and support vector regression. Further, we derived the optimal model through data preprocessing, input variable selection, and hyperparameter tuning. In both the validation and test phases, the “recyclable mineral waste” and “recyclable combustible waste” models achieved accuracies (R2) of 0.987 and 0.972, respectively. The “recyclable metals” and “landfill specified waste” models achieved accuracies (R2) of 0.953 and 0.858 or higher, respectively. Moreover, the “landfill mix waste” model exhibited an accuracy of 0.984 or higher. This study confirmed through Shapley Additive exPlanations analysis that the floor area is the most important input variable in the four models (i.e., recyclable mineral waste, recyclable combustible waste, recyclable metals, and landfill mix waste). Additionally, the type of equipment employed in demolition emerged as another crucial input variable impacting the volume of recycling and landfill waste generated. The results of this study can provide more detailed information on the generation of recycling and landfill waste. The developed model can provide precise data on waste management, thereby facilitating the decision-making process for industry professionals.
... The most commonly used groups of predictors in the literature for estimating the waste accumulation rate can be divided into socio-cultural, environmental, and economic variables, as summarized in Table 1. [1,[17][18][19][20][21][22][23] 3. Economic economic trends employment rate household income work and occupation [10,12,17,[24][25][26][27][28][29] Predicting the amount of municipal waste is a task with a high degree of uncertainty, which is an important part of the decision-making process in waste management [29,30]. The proper identification and selection of modeling variables are very important to avoid over-fitting the model and to facilitate the interpretation of the results, which, in turn, increases their reliability [11]. ...
... The choice of modeling technique and the consideration of exogenous factors vary considerably by country, regional specificity, and level of development. Models based on statistical methods such as linear and multiple regression [15,28,29], fuzzy and rough set theories [7,[33][34][35], multivariate gray models [36], time series [37][38][39], and artificial neural networks [7,10,15,17,21,27,30,31,[39][40][41][42][43][44][45][46] are used to forecast municipal waste generation. Numerous studies emphasize that the lack of sufficient data on municipal waste forecasting and management is a significant obstacle to the development of effective waste management systems [7,34]. ...
... An analysis of the abovementioned literature items showed that it is difficult to compare the presented results with each other and choose the best method for forecasting the municipal waste accumulation rate due to different sets of independent variables, different forecast horizons, and different quality assessment indicators, as well as the lack of comprehensive assessment due to the simultaneous consideration of several assessment criteria. Therefore, it is necessary to continue research in this area to develop more precise and effective methods for forecasting municipal waste volumes and effectively managing this important aspect of the sustainable development of societies [1,11,27,32]. ...
As a public service, municipal waste management at the local and regional levels should be carried out in an environmentally friendly and economically justified manner. Information on the quantity and composition of generated municipal waste is essential for planning activities related to the implementation and optimization of the process. There is a need for reliable forecasts regarding the amount of waste generated in each area. Due to the variability in the waste accumulation rate, this task is difficult to accomplish, especially at the local level. The literature contains many reports on this issue, but there is a lack of studies indicating the preferred method depending on the independent variables, the complexity of the algorithm, the time of implementation, and the quality of the forecast. The results concerning the quality of forecasting methods are difficult to compare due to the use of different sets of independent variables, forecast horizons, and quality assessment indicators. This paper compares the effectiveness of selected forecasting models in predicting the amount of municipal waste collection generated in Polish municipalities. The authors compared nine methods, including artificial neural networks (ANNs), support regression trees (SRTs), rough set theory (RST), multivariate adaptive regression splines (MARS), and random regression forests (RRFs). The analysis was based on 31 socioeconomic indicators for 2451 municipalities in Poland. The Boruta algorithm was used to select significant variables and eliminate those with little impact on forecasting. The quality of the forecasts was evaluated using eight indicators, such as the absolute percentage error (MAPE), mean absolute error (MAE), and coefficient of determination (R2). A comprehensive evaluation of the forecasting models was carried out using the APEKS method. An analysis of the results showed that the best forecasting methods depended on the set of independent variables and the evaluation criteria adopted. Waste management expenditures, the levels of sanitation and housing infrastructure, and the cost-effectiveness of waste management services were key factors influencing the amount of municipal waste. Additionally, this research indicated that adding more variables does not always improve the quality of forecasts, highlighting the importance of proper selection. The use of a variable selection algorithm, combined with the consideration of the impact of various socioeconomic factors on municipal waste generation, can significantly improve the quality of forecasts. The SRT, CHAID, and MARS methods can become valuable tools for predicting municipal waste volumes, which, in turn, will help to improve waste management system.
... • use developed predictive models to plan and implement effective waste management strategies [20] • handle waste collection, disposal, and recycling [15]. ...
... Additionally, sensor devices are needed with the following functionalities: (1) detecting the movement of people throwing rubbish into containers using motion sensors [51] and passive infrared (PIR) sensors [12]; (2) capturing images of trash using an image sensor [41]; (3) detecting the presence of waste using ultrasonic sensors [9,17,43] and infrared sensor [44]; (4) weighing the waste stored in the container [20,42] and determining the type, weight, and ratio of waste impurities [53] using load-cell sensors [21,59]; (5) measuring the level of waste in the container using an ultrasonic sensor [21,22,37,49,59], overfill sensor [41], tracker sensor [43] infrared sensor [45]; (6) detecting the number of solid or liquid particles in the waste container using an ultrasonic sensor [12], (7) detecting wet or dry waste using a capacitive sensor based on water content [49], temperature and humidity sensors based on temperature and humidity levels [50], and moisture sensors and touch sensors [51]; (8) detecting the presence of metal waste using metallic sensors [49], induction sensors [46], and inductive proximity sensors [48,50]; (9) detecting the presence of plastic and wood waste using capacitance proximity sensors [48]; (10) storing and managing time in the waste recycling process using a real-time clock (RTC) sensor [12]. ...
... • The existence of hardware to detect the presence of waste, the Sensor The availability of waste container location-allocation algorithms helps determine the type, amount, and location of waste at collection locations using the mixed integer linear programming (MILP) model [10] and the maximal covering location problem (MCLP) [57]. Analytical data can predict patterns [20] and trends in waste collection during specific periods [42], supporting decision-making regarding price adjustments [19]. Integration with GIS is helpful in managing waste services by displaying layers containing road segments, traffic signs, and embedded buildings [11], displaying GPS coordinate locations [52], providing visualization of spatial parameters to support waste management plans [13], visualizing and analyzing the distribution of waste generation points and potential collec-tion locations [57], storing and managing spatial data related to bin attributes, routes, and schedules, finding optimal routes [18], analyzing spatial data, managing geographic data, and connecting with decision-making support processes like AHP [66]. ...
Household waste is the primary source of environmental pollution due to global population growth compared to other waste sources. This article aims to develop a framework for a smart and integrated household waste management system through a Systematic Literature Review (SLR) using the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA). The resulting framework not only focuses on information technology dimensions but also links them with other integrated dimensions. The framework’s design identifies the types of household waste management processes based on the Integrated Sustainable Waste Management (ISWM) framework, dimensions that support smart household waste management system, and the stakeholders involved. The SLR results, which include dimensions and subdimensions supporting the smart and integrated household waste management system framework, were validated by experts from the Indonesian Ministry of Environment and Forestry. The developed framework includes five main dimensions: Information Technology, Operational Infrastructure, Governance, Economy, and Social–Culture. It also addresses stakeholder engagement to support smart household waste management systems and identifies waste management processes based on the ISWM framework. This research uses the PRISMA technique to provide an initial framework for smart and integrated household waste management system. The proposed framework has been validated and can be further developed as a smart and integrated household waste management system. Additionally, it highlights the involvement of various dimensions identified to address waste problems.
... A deep learning approach using a multi-site Long Short Term Memory (LSTM) neural network was used to estimate the level of household waste generation in Denmark [14]. Machine learning modelling can predict waste generation in residential areas in Vietnam by identifying variables such as waste generation characteristics, demographics, economic level, and community consumption levels (Nguyen et al., 2021). The ensemble learning approach can also be used to predict waste generation and has been proven to be more accurate than other individual machine learning approaches [15]. ...
... Aside from identifying primary dimensions, this research also identifies waste management processes and data. Identification of waste management processes aims to improve the performance of smart systems in waste management [16,17], the effectiveness of waste management resource allocation [18], and opportunities for process integration with smart systems such as sensors [19], machine learning algorithms [20], and communication network technologies [21]. Meanwhile, identifying data managed by smart waste management can support real-time monitoring to determine appropriate actions for waste management [18], determine more efficient management through optimizing waste transport vehicle routes [11], and the development of IoT-based solutions [22]. ...
... A deep learning approach using a multi-site Long Short Term Memory (LSTM) neural network was used to estimate the level of household waste generation in Denmark [14]. Machine learning modelling can predict waste generation in residential areas in Vietnam by identifying variables such as waste generation characteristics, demographics, economic level, and community consumption levels [20]. The ensemble learning approach can also be used to predict waste generation and has been proven to be more accurate than other individual machine learning approaches [15]. ...
Household waste is the primary source of environmental pollution due to global population growth compared to other sources of waste. This research aims to develop a smart and integrated household waste management system framework. The resulting framework not only focuses on dimensions of information technology but also links it with other integrated dimensions. This research attempts to conduct a Systematic Literature Review (SLR) using the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) protocol to produce a smart and integrated household waste management framework. The framework's design is carried out by identifying the type of household waste management process based on the Integrated Sustainable Waste Management (ISWM) framework, dimensions that support smart household waste management, and stakeholders involved. The SLR results, namely dimensions and subdimensions that support the framework of a smart and integrated household waste management system, were validated by a resource person from the Indonesian Ministry of Environment and Forestry. The smart and integrated household waste management framework that has been developed in this research contains five main dimensions: Information Technology, Operational Infrastructure, Governance, Economy, and Social-culture. This framework also addresses stakeholder engagement to support smart household waste management systems and identifies waste management processes based on the ISWM framework. This research uses the PRISMA technique to provide a framework for smart and integrated household waste management systems in the initial stage. The proposed framework has been validated and can be further developed as a smart and integrated household waste management system. This research also contributes to the involvement of various dimensions identified to deal with waste problems.
... Data preprocessing plays a vital role in addressing these issues and ensuring the reliability of the findings (Ramírez-Gallego et al. 2017). Data preprocessing involves important steps prior to modelling, such as cleansing, instance selection, normalization, transformation, feature extraction, and feature selection (Nguyen et al. 2021). ...
... where y i represents the actual value for the ith observation, x i represents the predicted value, n represents the number of observations, and x i represents the average of predicted values. Models with higher R 2 values and lower RMSE, MAE, and MSE were more successful than others (Nguyen et al. 2021). ...
Effective planning and managing medical waste necessitate a crucial focus on both the public and private healthcare sectors. This study uses machine learning techniques to estimate medical waste generation and identify associated factors in a representative private and a governmental hospital in Bahrain. Monthly data spanning from 2018 to 2022 for the private hospital and from 2019 to February 2023 for the governmental hospital was utilized. The ensemble voting regressor was determined as the best model for both datasets. The model of the governmental hospital is robust and successful in explaining 90.4% of the total variance.
Similarly, for the private hospital, the model variables are able to explain 91.7% of the total variance. For the governmental hospital, the significant features in predicting medical waste generation were found to be the number of inpatients, population, surgeries, and outpatients, in descending order of importance. In the case of the private hospital, the order of feature importance was the number of inpatients, deliveries, personal income, surgeries, and outpatients. These findings provide insights into the factors influencing medical waste generation in the studied hospitals and highlight the effectiveness of the ensemble voting regressor model in predicting medical waste quantities.
... Because of the changing patterns of consumption and the growth of the urban population, solid waste management (SWM) has become a major concern [1]. Municipal solid waste (MSW) includes building and demolition debris, street sweeping, and marketable, institutional, and leftover cleaning materials [2,3]. ...
... where Const1 equals 2, and the dimensional vector dim generates a random count among [0,1]. Similarly, if 0.5, TO > the exploitation phase occurs with no collision between the object and random material best best best Density ( 1) ...
The rapid growth of smart cities and industry causes an increase in waste production. The amount of municipal solid waste (MSW) increases by several factors, including population growth, economic status, and consumption trends. The inadequacy of basic trash data is a major issue for managing MSW. Numer-ous existing models based on solid waste prediction have been presented so far, but none of them predict solid waste accurately and it consumes more time. To address these concerns, a deep convolutional spiking neural network for solid waste prediction (DCSNN-SWP) is proposed in this paper. Here, the real-time solid waste prediction data are gathered from the quantity of municipal corporation of Chennai (MCC), landfill, garden garbage, and coconut shell reports in Tamil Nadu (Chennai), such as Zone 9 (Nungambak-kam), Zone 10 (Kodambakkam) and Zone 13 (Adyar). Then the collected solid waste data are pre-processed using the kernel correlation model. Then the pre-processing data is given to DCSNN-hybrid BCMO and Archimedes optimization algorithm which accurately predicts the solid waste as wet waste, dry waste, horti-culture waste, and dumping yard for 2022–2032 years. The proposed DCSNN-SWP method has been implemented in Python.
... This algorithm has a simple structure with a fasttraining speed, but it is easy to fall into local optimality. ANN (Puntarić et al., 2022) is commonly employed in EOL product quantity prediction, which includes several algorithms such as feed forward neural network (FFNN) (Abbasi and El Hanandeh, 2016), back propagation neural network (BPNN) (Oguz-Ekim, 2021), deep neural network (DNN) (Nguyen et al., 2021), general regression neural network (GRNN) (Sodanil and Chatthong, 2014), adaptive network-based fuzzy inference system (ANFIS) (Golbaz et al., 2019), Elman neural network (ENN) (Meza et al., 2019), recurrent neural network (RNN) (Li and Ma, 2019), long short-term memory (LSTM) , and nonlinear autoregressive (NAR) . Multi-layer perceptrons (MLP) (Coskuner et al., 2021), which is a special type of FFNN with a more complex structure and stronger expressive power, have been put forward to predict EOL products. ...
... Random Forest (RF) RF is one of the classification and regression tree (CART) models based on Bagging integration. This algorithm has high accuracy in training results and good parallelism, but it performs poorly on small data sets (Nguyen et al., 2021). ...
The rapid development of machine learning algorithms provides new solutions for predicting the quantity of recycled end-of-life products. However, the Stacking ensemble model is less widely used in the field of predicting the quantity of recycled end-of-life products. To fill this gap, we propose a Stacking ensemble model that utilizes support vector regression, multi-layer perceptrons, and extreme gradient boosting algorithms as base models, and linear regression as the meta model. The k-nearest neighbor mega-trend diffusion method is applied to avoid overfitting problems caused by a small sample data set. The grid search and time series cross validation methods are utilized to optimize the proposed model. To verify and validate the proposed model, data related to China's end-of-life vehicles industry from 2006 to 2020 is used. The experimental results demonstrate that the proposed model achieves higher prediction accuracy and generalization ability in predicting the quantity of recycled end-of-life products.
... Furthermore, a sophisticated Convolutional Neural Network (CNN) architecture, utilizing water quality parameters, environmental factors, and additional atmospheric data as input variables, demonstrated excellent performance in predicting algal blooms in two large multi-purpose weirs (Pyo et al., 2020). While most previous studies focused on the reliability and accuracy of prediction, it is noteworthy that ML models are also J o u r n a l P r e -p r o o f Journal Pre-proof effective in disclosing the most important features of water-related data Nguyen et al., 2021b;Zhu et al., 2021), which suggests their potential application in minimizing data preparation and reducing costs associated with field sampling and analyses. ...
... Out of the tested models, the best prediction accuracy was found for XGBoost-Stacking, followed by LR, RR, and ANN, as shown by the decrease in the R 2 values and the increase in MAE, MSE, and RMSE results, following the order (Table S5). Importantly, it should be noted that the prediction performance of ML models on algal prediction far exceeds the simple linear bivariate correlation shown in Fig. S5 The superior performance of ensemble learning, such as Stacking and XGBoost, over conventional regression models is well-supported by many previous studies (Nguyen et al., 2021b;Shamshirband et al., 2019;Truong et al., 2021). Ensemble learning models were found to J o u r n a l P r e -p r o o f be more effective in addressing uncertainty, interactions between various predictors, and unbalanced datasets. ...
The prediction of algal blooms using traditional water quality indicators is expensive,
labor-intensive, and time-consuming, making it challenging to meet the critical
requirement of timely monitoring for prompt management. Using optical measures for forecasting algal blooms is a feasible and useful method to overcome these problems. This study explores the potential application of optical measures to enhance algal bloom prediction in terms of prediction accuracy and workload reduction, aided by machine learning (ML) models. Compared to absorption-derived parameters, commonly used fluorescence indices such as the fluorescence index (FI), humification index (HIX), biological index (BIX), and protein-like component improved the prediction accuracy. However, the prediction accuracy was decreased when all optical indices were considered for computation due to increased noise and uncertainty in the models. With the exception of chemical oxygen demand (COD), this study successfully replaced biochemical oxygen demand (BOD), dissolved organic carbon (DOC), and nutrients with selected fluorescence indices, demonstrating relatively analogous performance in either training or testing data, with consistent and good coefficient of determination (R2) values of approximately 0.85 and 0.74, respectively. Among all models considered, ensemble learning models consistently outperformed conventional regression models and artificial neural networks (ANNs). However, there was a trade-off between accuracy and computation efficiency among the ensemble learning models (i.e., Stacking and XGBoost) for algal bloom prediction. Our study offers a glimpse of the potential application of spectroscopic measures to improve accuracy and efficiency
in algal bloom prediction, but further work should be carried out in other water bodies to further validate our proposed hypothesis.
... Globally, 72% of GHGs are linked to household consumption activities [7]. Consumption upgrading (CU), as a key driver of economic growth, not only alters consumer purchasing behaviors and preferences, but also profoundly impacts the quantity, composition, and spatial distribution of MSW [8]. CU, particularly the increasing preference of residents for environmentally friendly products and services, plays a critical role in promoting the transition of MSW management toward eco-efficiency. ...
To achieve the goal of building zero-waste cities, managing greenhouse gas (GHG) emissions generated from municipal solid waste (MSW) treatment is a critical step toward carbon neutrality. Waste produced by consumption activities constitutes an essential component of MSW management. Using the Super Slacks-Based Measure Data Envelopment Analysis (SSBM-DEA) model and the Spatial Durbin Model (SDM), this study investigates the spatial impacts of consumption upgrading (CU) on municipal waste management across 30 provinces in China, with a particular focus on GHGs as undesirable outputs. In this study, we construct a framework from the dimensions of consumption level, consumption structure, and green consumption. Additionally, other socioeconomic factors influencing waste management are explored. The results indicate a convergence trend in the uneven distribution of consumption upgrading, with the gaps between regions gradually narrowing. Consumption upgrading significantly enhances the eco-efficiency of local waste management and exhibits notable spatial spillover effects, positively influencing the eco-efficiency of neighboring regions. Furthermore, the promotion effect of consumption upgrading on the central and western regions, compared with the eastern region, is more pronounced. This indicates that the technological catch-up resulting from consumption upgrading, supported by policies, can further enhance the eco-efficiency of MSW. This study also provides insights for other regions transitioning from scale expansion to high-quality development in waste management.
... Bayesian-optimized ANN models with ensemble uncertainty analysis were used to predict heterogeneous MSW generation rates [22]. The study demonstrated a correlation between MSW physical composition and other indices, resulting in more reliable predictions with lower relative standard deviations. ...
... CatBoost, designed for categorical data, was also explored in the WTE context; however, on the contrary, another recent work [4] indicated that XGBoost outperformed it for reliable energy recovery estimates. A study in Vietnam on ML for accurate MSW prediction [5] noted challenges with data gaps that necessitate model updates. ...
Accurate estimating energy recovery from trash is vital for optimizing the Waste-to-Energy (WTE) mechanisms essential in tackling global waste management processes and energy sustainability issues. This paper analyzes a synthetically expanded dataset generated with the help of SMOTE techniques to compare the performance of four machine learning (ML) models, Decision Tree Regression, Random Forest Regression, CatBoost Regression, and XGBoost Regression. Here, the dataset contains important waste parameters like composition, moisture content, and treatment procedures that help the models forecast energy output with high precision—datasets and evaluate additional machine learning techniques to boost prediction accuracy in industrial WTE systems further. Performance indicators like MAE, RMSE, MAPE, and R² scores have been assessed here to identify each model’s accuracy and computational efficiency. The final result of the analysis states that the ensemble based models, more precisely XG-Boost and CatBoost, outperformed the simpler ones like Decision Tree, where CatBoost achieved the best R² value of 0.9893 and the minimum MAPE of 12.90 percent. Though using little extra storage, CatBoost showed great performance. The obtained results bring useful insights into efficient model selection for WTE applications. Further studies shall therefore be exclusively focused on validating this study’s results in real life conditions.
... For example, algorithms like Random Forest (RF), Support Vector Machines (SVM), and Neural Networks can be utilized to identify patterns and make predictions based on historical data. Cuong et al. [24] conducted a study comparing six ML models to forecast MSW generation in selected residential areas of Vietnam. Among these models, the k-nearest neighbor and RF algorithms demonstrated strong predictive performance on the training dataset (which comprised 80% of the total data), achieving an R 2 value exceeding 0.96 and a mean absolute error ranging from 121.5 to 125.0 on the testing dataset (the remaining 20%). ...
The escalating challenges of municipal solid waste (MSW) management, exacerbated by the classification of MSW as hazardous waste due to the presence of heavy metals (HMs) and toxic compounds, necessitate innovative treatment strategies. Plasma pyrolysis has emerged as a promising technology for converting MSW into valuable energy byproducts, such as syngas, bio-oil, and slag, while significantly reducing waste volume. However, maintaining optimal operational parameters during the plasma pyrolysis process remains a complex challenge that can adversely affect both the efficiency and the quality and quantity of outputs. To address this issue, the integration of the Internet of Things (IoT) presents a transformative approach. By leveraging IoT technologies, real-time monitoring and advanced data analytics can be employed to optimize the operational conditions of plasma pyrolysis systems, ensuring consistent performance and maximizing resource recovery. This review explores the synergistic integration of plasma pyrolysis and IoT as a novel strategy for MSW management. The slag from plasma treatment can be efficiently channeled into anaerobic digestion (AD) systems, promoting resource recovery through biogas production and the generation of nutrient-rich digestate. This synergy not only mitigates the environmental impacts associated with traditional MSW disposal methods but also paves the way for sustainable energy recovery and resource management. Ultimately, this review presents a comprehensive framework for exploiting plasma pyrolysis and IoT in addressing the pressing issues of hazardous MSW, thereby fostering a circular economy through innovative waste-to-energy solutions.
... MARS (R 2 = 0.88), ANN (R 2 = 0.74), and RF (R 2 = 0.77). Nguyen et al. [25] developed a predictive model for MSW generation in residential areas of Vietnam, where the dataset (189 MSW samples) collected from 2015 to 2017 across nine features-urban population, total retail sales of consumer goods, average per capita monthly income, average per capita size of the home, population density, average per capita monthly consumption expenditure, total hospital beds, total residential land per province, and total solid waste collected per day) were used. The K-nearest neighbor (KNN), RF, and DNN algorithms were applied through hyper-parameters adjustment, and the resulting R 2 were 0.96, 0.97, and 0.91, respectively. ...
Smart management of construction and demolition (C&D) waste is imperative, and researchers have implemented machine learning for estimating waste generation. In Korea, the management of demolition waste (DW) is important due to old buildings, and it is necessary to predict the amount of DW to manage it. Thus, this study employed decision tree (DT)-based ensemble models (i.e., random forest—RF, extremely randomized trees—ET, gradient boosting machine—GBM), and extreme gradient boost—XGboost) based on data characteristics (i.e., small datasets with categorical inputs) to predict the demolition waste generation rate (DWGR) of buildings in urban redevelopment areas. As a result of the study, the RF and GBM algorithms showed better prediction performance than the ET and XGboost algorithms. Especially, RF (6 features, 450 estimators; mean, 1169.94 kg·m−2) and GBM (4 features, 300 estimators; mean, 1166.25 kg·m−2) yielded the top predictive performances. In addition, feature importance affecting DWGR was found to have a significant impact on the order of gross floor area (GFA) > location > roof material > wall material. The straightforward collection of features used here can facilitate benchmarking as a decision-making tool in demolition waste management plans for industry stakeholders and policy makers. Therefore, in the future, it is required to improve the predictive performance of the model by updating additional data and building a reliable dataset.
... Nguyen et al. [114] applied a data-driven approach and computational intelligence methods for forecasting the generation rate of solid waste in residential areas. The approach was validated for a case study in Vietnam, comprising selected residential areas. ...
This article reviews the literature surrounding innovative computational tools for waste management within smart cities. With the rise of urbanization and the increasing challenges of waste management, innovative technologies play a pivotal role in optimizing waste collection, sorting, recycling, and disposal processes. Leveraging computational tools such as artificial intelligence, Internet of Things, and big data analytics, smart waste management systems enable real-time monitoring, predictive modeling, and optimization of waste-related operations. These tools empower authorities to enhance resource efficiency , minimize environmental impact, and improve the overall quality of urban living. Through a comprehensive review of recent research and practical implementations, this article highlights the key features, benefits, and challenges associated with the development of cutting-edge computational tools for waste management. Emerging trends and opportunities for research and development in this rapidly evolving field are identified, emphasizing the importance of integrating technological innovations for building sustainable and resilient waste management in smart cities.
... The authors used a gradient boosting model to predict short-term waste generation in New York City, demonstrating the effectiveness of the GBM model for short-term waste prediction. Nguyen et al. [21] used six machine learning models to predict municipal household waste generation by selecting residential areas in Vietnam and found that the random forest and KNN algorithms had better predictive power for the training data, and the test data had R2 value exceeded 0.96. Kannangara et al. [22] conducted an empirical study of solid waste data from five Canadian provinces using a multi-layer perceptron (MLP) and support vector regression (SVR) machine learning approach to develop a predictive model for municipal solid waste management in Canada. ...
As urbanization accelerates, the management of urban solid waste poses increasingly intricate challenges. Traditional urban metrics, such as GDP and per capita consumption rates, have become inadequate for accurately reflecting the realities of waste generation; moreover, the linear correlation between these metrics and waste production is progressively diminishing. Consequently, this study introduces a novel methodology leveraging nighttime light remote sensing data to enhance the precision of urban solid waste production forecasts. By processing remote sensing data to mitigate noise and integrating it with conventional urban datasets, an innovative index system and predictive model were developed. Using Beijing as a case study, the gradient boosting regression algorithm yielded a prediction accuracy of 92%. Furthermore, in light of the substantial costs associated with waste recovery route planning and site selection for treatment facilities, this research further devised a location and distribution framework for waste treatment centers based on high-precision predictions of waste production while employing multi-objective evolutionary algorithms (MOEAs) alongside the non-dominated sorting genetic algorithm II (NSGA-II) for optimization. Distinct from prior studies, this study suggests that service point waste quantities are not fixed values but rather adhere to a normal distribution within specified ranges and thus provides a more realistic simulation of fluctuations in waste production while enhancing both the robustness and predictive accuracy of the model. In conclusion, by incorporating nighttime light remote sensing data along with advanced machine learning techniques, this study markedly improves forecasting accuracy for waste production while offering effective optimization strategies for site selection and recovery route planning—thereby establishing a robust data foundation aimed at refining urban solid waste management systems.
... In the specific sectors of renewable energy and waste management, there is significant value in being able to forecast the price or value of various variables. For instance, various models have been explored to forecast municipal solid waste (MSW), including correlation analysis (Daskalopoulos, Badr, and Probert 1998) A comparison of 6 ML models including random forest (RF) and k-nearest neighbor (KNN) algorithms was explored by (Nguyen et al. 2021) to forecast solid waste generation in residential areas in Vietnam. This study looked at eight variables including those related to the economy, demography, consumption and waste generation and found that urban population, monthly consumption expenditure, and retail sales were important variables to consider. ...
Time series forecasting methods have been adopted to forecast prices in various sectors. However, in certain non-transparent markets like the price of wood waste, due to the limitations of availability of information and several external factors, the task is more challenging. In this thesis, various time series models and forecasting methods are compared by using them to forecast future prices of wood waste based on historical prices recorded weekly for different categories of wood waste in 10 clusters (groups of postal codes) in Germany. The standard steps of data cleaning will be followed by exploration of characteristics of the time series including ACF, PACF, decomposition, differencing and stationarity test to select the necessary parameters for the models. Five different models are compared including parametric models such as ARIMA and non-parametric ones like Decision Trees, Random Forest, XGBoost and Prophet, as well as 3 distinct methods of forecasting and validation – static (multi-step) forecasting, rolling window and walk-forward validation. When the performance metrics such as RMSE and Direction Accuracy for each method and model were compared, it was found that the static method yielded poorer results on RMSE in most cases, and that the choice of the recommended model and forecasting method depended heavily on the specific time series of the categories in the clusters.
... У роботі [7] створено моделі машинного навчання з використанням соціально-економічних та екологічних даних для прогнозування утворення твердих побутових відходів у містах В'єтнаму. Авторами протестовано різні підходи до моделювання, такі як штучні нейронні мережі та методи на основі дерев рішень і опорних векторів. ...
У роботі запропоновано комплексний підхід до виявлення звалищ, який має дві основні складові. Перша базується на класифікації часових рядів мультиспектральних супутникових даних з використанням ансамблю нейронних мереж. Друга використовує модифіковану архітектуру U-Net для семантичної сегментації космічних знімків з метою безпосереднього виділення звалищ. Новизна представленого рішення полягає у використанні комбінованої функції втрат при навчанні моделі U-Net, що поєднує бінарну крос-ентропію та коефіцієнт Dice. Бінарна крос-ентропія забезпечує надійну попіксельну класифікацію, тоді як коефіцієнт Dice оптимізує сегментацію шляхом максимізації перетину між прогнозованими та істинними масками звалищ. Ця комбінація дозволяє досягти балансу між точністю класифікації та чутливістю до малих об’єктів, що важливо для виявлення звалищ за супутниковими даними. Запропоновано інтегрувати результати двох підходів на рівні прийняття рішень за допомогою ймовірнісного злиття з використанням ймовірностей кожної моделі. Такий гібридний метод дозволяє компенсувати недоліки кожного методу та підвищити загальну точність ідентифікації звалищ. Експериментальна перевірка на тестових даних продемонструвала ефективність розглянутого підходу. Загальна точність класифікації земного покриву склала 97,4 %, а точність визначення звалищ — 86,4 %. Розроблений метод застосовано для картографування звалищ на території Донецької області. Результати верифіковано експертами на місцевості. Отримані дані можуть бути використані місцевою владою для оперативного реагування та прийняття управлінських рішень.
... Currently, scholars mainly consider the influence of three dimensions: demographic, economic, and social in their studies on predicting MSW [24]. Among them, the demographic dimension mostly uses influencing factors such as gender ratio, resident population [25], inbound tourism [26], and education level [27] The economic dimension includes influencing factors such as GDP, per capita consumption expenditure, per capita disposable income [28], and total retail sales of consumer goods [29]. The social dimension includes urban road area, citywide centralized heating area, urban green coverage [30], and other influencing factors. ...
To cope with the increasing energy demand of people and solve the problem of a “Garbage Siege”, most cities have begun to adopt waste power generation (WTE). Compared to other WTE technologies, incineration has proven to be the most efficient technology for municipal solid waste (MSW) treatment. Therefore, to further explore the economic feasibility of MSW incineration plant construction, this study established a multi-factor prediction of MSW generation based on the GRA-BiLSTM model. By fully considering the relationship between the change in feed-in tariff (FIT) and the building of an incineration plant in Beijing, the economic feasibility of building an incineration plant is discussed based on the three scenarios set. The experimental results showed that (1) the combined model based on the GRA-BiLSTM showed good applicability for predicting MSW generation in Beijing, with MAE, MAPE, RMSE, and R2 values of 12.47, 5.97%, 18.5580, and 0.8950, respectively. (2) Based on the three scenarios set, the incineration power generation of Beijing MSW will show varying degrees of growth in 2022–2035. In order to meet future development, Beijing needs to build seven new incinerators, and the incineration rate should reach 100%. (3) According to setting different feed-in tariffs, based on the economic feasibility analysis, it is found that the feed-in tariff of MSW incineration for power generation in Beijing should be no less than $0.522/kWh. The government should encourage the construction of incineration plants and give policy support to enterprises that build incineration plants.
... It includes more technological procedures that result in complete management, followed by the removal of pollution without detrimental effects on the surroundings [6]. The NWT implementation has its economic justification since the reduction of the amount of waste and complete use of materials paves the way for better productivity and reduces the imports of raw materials [7]. Rarely, it may probably minimalize the usage of heat and electricity or by decreasing energy-consuming waste treatment procedures. ...
Intelligent manufacturing system incorporates a number of sensors including IoT devices, cameras, and scanners, for capturing real-time data about the manufacturing process. Based on their physical properties, colours, dimensions, or other relevant characteristics, these sensors can be used to track and identify waste objects. Waste object classification in intelligent manufacturing includes the usage of recent systems and technologies to detect and classify waste materials or objects produced during the manufacturing process. The objective is to enable effective waste management and recycling practices, optimizing resource utilization and reducing environmental impact. Manual waste classification is a laborious and expensive task, which results in the development of automatic waste classification models using artificial intelligence (AI) techniques. It remains a challenging process due to the significant variations in the solid waste because of varying shapes, colours, and sizes. Therefore, recent advances in deep learning (DL) methods can be employed to accomplish the solid waste classification process. The study introduces a chaotic African vulture optimization algorithm with a deep learning-based solid waste classification (CAVOA-DLSWC) system. The CAVOA-DLSWC technique aims to automatically detect waste objects and classify them into different categories using DL models. In the presented CAVOA-DLSWC approach, two major processes are involved such as object classification and detection. For the object detection method, the CAVOA-DLSWC technique uses a lightweight RetinaNet model with CAVOA based hyperparameter tuning process. The CAVOA is derived by integrating the chaotic concepts into the initial iteration values of the AVOA. Once the waste objects are identified, the classification process can be performed by the use of convolutional long short-term memory (CLSTM) network. The experimental values of the CAVOA-DLSWC approach can be tested employing the solid waste database including diverse kinds of waste objects. The comparative results show the remarkable performance of the CAVOA-DLSWC method over other techniques.
... Our literature review highlights a variety of methods and variables employed in forecasting MSW generation, as showcased in Table 1 [5,9,[14][15][16][17][18][19][20]. Notably, Wu et al. (2020) demonstrated the significance of geographic differentiation in achieving accurate predictions through their regional approach in China [14]. ...
This study integrates circular economy (CE) metrics with machine learning techniques, specifically XGBoost and Shapley additive explanations (SHAP), to forecast municipal solid waste (MSW) in the EU, analyzing data from 2010 to 2020. It examines key economic and consumption indicators, including GDP per capita and energy consumption, along with CE metrics such as resource productivity, the municipal waste recycling rate, and the circular material use rate. The model demonstrates high predictive accuracy, with an R2 of 99% for in-sample data and 75% for out-of-sample data. The results indicate a significant correlation between a higher GDP per capita and an increased gross municipal waste per capita (GMWp). Conversely, lower energy consumption is associated with reduced GMWp. Notably, the circular material use rate emerges as a crucial factor for sustainability, with increased use significantly decreasing the GMWp. In contrast, a higher resource productivity correlates with an increased GMWp, suggesting complex implications for waste generation. The recycling rate, while impactful, shows a more modest effect compared to the other factors. The culminating insights from this study emphasize the need for sustainable, integrated waste management and support the adoption of circular economy-aligned policies. They underscore the efficacy of merging CE metrics with advanced predictive models to bolster regional sustainability efforts.
... Zhang found that the number of urban population and GDP are the two most important predictors by selecting among demographic indicators and economic indicators [42]. Nguyen while four categories of economic, social, environmental, and population indicators were selected to assess the future growth rate of MSW in Vietnam [14]. ...
China MSW (Municipal solid waste) is being transformed from a strictly environmental problem to a renewable resource. Effective treatment of future MSW requires reliable forecasts of MSW generation, separation rates, incineration costs, and power generation, which are becoming more and more important. This study takes Beijing as the research object, establishes a combined model based on GRA(Grey relation analysis) and Bidirectional Long Short Term Memory, selects 10 influencing factors affecting municipal solid waste in Beijing as input indicators, realizes effective prediction of municipal solid waste generation, calculates electricity generation efficiency, cost and electricity generation according to different MSW source separation rate, and screens out the optimal MSW source separation rate. Firstly, GRA was used to select the influencing factors of MSW. Secondly, the screened indicators are used as input indicators for the model BILSTM (Bidirectional Long Short Term Memory) to forecast the amount of Beijing Municipal solid waste generation in 2035. The results show that the MAPE value of the established combined GRA-BILSTM model is 4.956, and the prediction performance of this model is better than the other eight models. Relevant suggestions should be made for the source separation rate of municipal solid waste in 2035.
... represents the actual value for the i th observation, represents the predicted value, n represents the 333 number of observations, and ̅ represents the average of predicted values. Models with higher R 2 values and 334 lower RMSE, MAE and MSE were more successful than others(Nguyen et al., 2021). ...
Effective planning and managing medical waste necessitate a crucial focus on both the public and private healthcare sectors. This study uses machine learning techniques to estimate medical waste generation and identify associated factors in a representative private and a governmental hospital in Bahrain. Monthly data spanning from 2018 to 2022 for the private hospital and from 2019 to February 2023 for the governmental hospital was utilized. The ensemble voting regressor was determined as the best model for both datasets. The model of the governmental hospital is robust and successful in explaining 90.4% of the total variance.
Similarly, for the private hospital, the model variables are able to explain 91.7% of the total variance. For the governmental hospital, the significant features in predicting medical waste generation were found to be the number of inpatients, population, surgeries, and outpatients, in descending order of importance. In the case of the private hospital, the order of feature importance was the number of inpatients, deliveries, personal income, surgeries, and outpatients. These findings provide insights into the factors influencing medical waste generation in the studied hospitals and highlight the effectiveness of the ensemble voting regressor model in predicting medical waste quantities.
... The 10-fold cross-validation is used for the data in training as it can improve the accuracy [50] and generalization ability of the model [51,52] so that the overfitting problem can be avoided. As shown in Figure 6, all data are divided into ten parts, one fold is used as validation set, and the others are used as training set in each part. ...
The durability degradation of reinforced concrete was mainly caused by chloride ingress. Former studies have used component parameters of concrete to predict chloride diffusion by machine learning (ML), but the relationship between microstructure and macroparameter of concrete need to be further clarified. In this study, multi-layer perceptron (MLP) and support vector machine (SVM) were used to establish the prediction model for chloride diffusion coefficient in concrete, especially for the solid waste concrete. A database of concrete pore parameters and chloride diffusion coefficients was generated by the algorithm based on the Gaussian mixture model (GMM-VSG). It is shown that both MLP and SVM could make good predictions, in which the data using the normalization preprocessing method was more suitable for the MLP model, and the data using the standardization preprocessing method was more adapted to the SVM model.
... Its performance on MSW prediction may vary, reaching R 2 as high as 0.9 in studies of some specific areas (Chhayet al., 2018) but less than 0.7 in other studies (Azadi & Karimi-Jashni, 2016;Kumar & Samadder, 2017). Support vector machine (SVM), random forest, and k-nearest neighbor (KNN) show good prediction accuracy in some studies, but they are rarely used in recent studies (Abbasi & El Hanandeh, 2016;Nguyen et al., 2021). The artificial neural network (ANN) model has been used in most research (Xia et al., 2022), but it often has problems with overfitting and insufficient generalization ability (Abbasi & El Hanandeh, 2016). ...
... YOLO, characterized as a proficient CNN model, excels in real-time object detection. The entire image is processed through a singular computational model, subsequently subdividing it into regions and determining boundary boxes and probabilities for each [22]. Among object detection algorithms, YOLO is recognized for its prowess, especially when handling extensive datasets, marking it as exceptionally fast and accurate [23]. ...
... Addition on-ground services, investigation and contextual information is required to enforce GPS tracking D.Lee et al., 2018 -In Machine learning based approaches sufficient data set is required for training the model. Lack of large data set and variety of data set is biggest challengeNguyen et al., 2021 Waste disposal techniques: Waste handling and disposal is major challenge with waste management. After selecting the best alternate for waste collection and transport, main issue is disposing the waste. ...
Urbanization and rising populations are contributing significantly to the problem of garbage generation in cities. For a quite long time, big cities all around the world have been particularly concerned about rising waste. For countries with inadequate waste management procedures and faster-rising garbage volumes, this is a severe problem. Many of the waste management systems in use in cities today do not come with features like operational transparency, traceability, audit, security, and trusted data provenance. The way solid waste is managed has evolved as a result of the Internet of Things, Global Positioning System, and cloud computing methods. The research community has also developed numerous waste management methods based on the Internet of Things and blockchain, and machine learning. Using the method of a systematic literature review, this research examines several waste management approaches. Based on the considered 178 research papers, this paper provides the opportunities brought about by the Internet of Things, machine learning, Global Positioning System, and Blockchain in various waste management techniques, their application scenarios, real-time tracing and tracking of waste, reliable channelization and compliance with waste treatment, efficient waste resource management, protection of waste management documentation, and fleet management. Based on the conducted review, this paper presents open challenges associated with waste management techniques that act as future research directions. Waste bin placement and its security, internet connectivity with waste management components, transportation of waste, and waste disposal techniques are the main categories of challenges associated with waste management. This paper also provides a comparison of the presented review with other published review papers.
The intensive generation of industrial solid waste (ISW) in industrial parks poses a severe threat to the water and soil environment. To address the lack of system‐level guidance strategies for ISW management (ISWM), the paper proposes a systematic ISWM framework within the context of the circular economy principle and applies it to a prominent eco‐industrial park in China. This framework is founded on the metabolic characteristics of the entire ISW process, supported by multi‐stakeholder cooperation, and guaranteed by a shareable information platform and infrastructure. A holistic ISW flow analytical process is introduced to facilitate the implementation of this framework. In the case study, a high‐resolution inventory of ISW is established by 157 enterprises in 2021, covering 116 types of general ISW and 61 types of hazardous waste (HW). By integrating material flow analysis, questionnaires, and interviews, this study quantitatively reveals the underperformance of mini‐scale HW recycling, with a recycling rate of 5.6% in 2021, well below the park‐wide rate of 50.8%. Collaborative inboundary and transboundary disposal emerges as a crucial direction for ISWM, with over 80% of ISW being disposed of outside the park. Following the implementation of circular economy strategies guided by the management framework, notable improvements are observed, including a 26.5% increase in the recycling rate of mini‐scale HW and a 13.5% increase in the proportion of disposal outside Beijing. To foster sustainable ISWM, this study recommends enhancing the decentralized infrastructure of collection–storage–transportation integration for mini‐scale ISW in industrial parks. The developed management framework provides valuable insights and guidance for park‐specific ISWM strategies.
This study presents the ‘Dual Path CNN-MLP’, a novel hybrid deep neural network (DNN) architecture that merges the strengths of convolutional neural networks (CNNs) and multilayer perceptrons (MLPs) for regional groundwater flow simulations. This model stands out from previous DNN approaches by managing mixed input types, including both imagery and numerical vectors. Such flexibility allows the diverse nature of groundwater data to be efficiently utilized without the need to convert it into a uniform format, which often leads to oversimplification or unnecessary expansion of the dataset. When applied to the northeast Qatar aquifer, the model demonstrates high accuracy in simulating transient groundwater flow fields, benchmarked against the well-established MODFLOW model. The model's efficacy is confirmed through k-fold cross-validation, showing an error margin of less than 12% across all examined locations. The study also examines the model's ability to perform uncertainty analysis using Monte Carlo simulations, finding that it achieves around 1% average absolute percentage error in estimating the mean hydraulic head. Errors are mostly found in areas with significant variations in the hydraulic head. Switching to this ML model from the conventional MODFLOW simulator boosts computational efficiency by about 99%, showcasing its advantage for tasks like uncertainty analysis in repetitive groundwater simulations.
Waste management poses a pressing global challenge, necessitating innovative solutions for resource optimization and sustainability. Traditional practices often prove insufficient in addressing the escalating volume of waste and its environmental impact. However, the advent of Artificial Intelligence (AI) technologies offers promising avenues for tackling the complexities of waste management systems. This review provides a comprehensive examination of AI’s role in waste management, encompassing collection, sorting, recycling, and monitoring. It delineates the potential benefits and challenges associated with each application while emphasizing the imperative for improved data quality, privacy measures, cost-effectiveness, and ethical considerations. Furthermore, future prospects for AI integration with the Internet of Things (IoT), advancements in machine learning, and the importance of collaborative frameworks and policy initiatives were discussed. In conclusion, while AI holds significant promise for enhancing waste management practices, addressing challenges such as data quality, privacy concerns, and cost implications is paramount. Through concerted efforts and ongoing research endeavors, the transformative potential of AI can be fully harnessed to drive sustainable and efficient waste management practices.
Improper municipal solid waste (MSW) management contributes to greenhouse gas emissions, necessitating emissions reduction strategies such as waste reduction, recycling, and composting to move towards a more sustainable, low-carbon future. Machine learning models are applied for MSW-related trend prediction to provide insights on future waste generation or carbon emissions trends and assist the formulation of effective low-carbon policies. Yet, the existing machine learning models are diverse and scattered. This inconsistency poses challenges for researchers in the MSW domain who seek to identify and optimize the machine learning techniques and configurations for their applications. This systematic review focuses on MSW-related trend prediction using the most frequently applied machine learning model, artificial neural network (ANN), while addressing potential methodological improvements for reducing prediction uncertainty. Thirty-two papers published from 2013 to 2023 are included in this review, all applying ANN for MSW-related trend prediction. Observing a decrease in the size of data samples used in studies from daily to annual timescales, the summarized statistics suggest that well-performing ANN models can still be developed with approximately 33 annual data samples. This indicates promising opportunities for modeling macroscale greenhouse gas emissions in future works. Existing literature commonly used the grid search (manual) technique for hyperparameter (e.g., learning rate, number of neurons) optimization and should explore more time-efficient automated optimization techniques. Since there are no one-size-fits-all performance indicators, it is crucial to report the model's predictive performance based on more than one performance indicator and examine its uncertainty. The predictive performance of newly-developed integrated models should also be benchmarked to show performance improvement clearly and promote similar applications in future works. The review analyzed the shortcomings, best practices, and prospects of ANNs for MSW-related trend predictions, supporting the realization of practical applications of ANNs to enhance waste management practices and reduce carbon emissions.
Municipal solid waste (MSW) is the derivative of urban development and it is harmful to the environment and residents' health. But with sustainable MSW management, MSW can be applied as an important renewable energy. In order to achieve sustainable MSW management, it is necessary to understand the mechanism of MSW generation. Consumption patterns differ in various regions of China, which make the influencing factors of MSW have unique characteristics. To explore the factors influencing MSW generation in China, this study builds a global model based on the panel data of 30 Chinese provinces. Considering regional heterogeneity, provinces are clustered into three groups according to economic and consumption indicators. Each group has its own local model of MSW generation. The results show that household expenditure on housing and the tertiary industry proportion show opposite impacting directions in high-level and low-level provinces. Finally, with the combination of the grey model (1,1) (GM(1,1)) and multiple linear regression (MLR), we find that developing provinces will generate more MSW than developed regions. According to this, different provinces should control MSW by optimizing consumption pattern and efficient fiscal expenditure, and developing provinces should pay attention to MSW management and learn from the experience of developed provinces.
The shear strength of rockfill materials (RFM) is an important engineering parameter in the design and audit of geotechnical structures. In this paper, the predictive reliability and feasibility of random forests and Cubist models were analyzed by estimating the shear strength from the relative density, particle size, distribution (gradation), material hardness, gradation and fineness modulus, and confining (normal) stress. For this purpose, case studies of 165 rockfill samples have been applied to generate training and testing datasets to construct and validate the models. Thirteen key material properties for rockfill characterization were selected to develop the proposed models. Validation and comparison of the models have been performed using the root mean square error (RMSE), coefficient of determination (R2), and mean estimation error (MAE) between the measured and estimated values. A sensitivity analysis was also conducted to ascertain the importance of various inputs in the prediction of the output. The results demonstrated that the Cubist model has the highest prediction performance with (RMSE = 0.0959, R2 = 0.9697 and MAE = 0.0671), followed by the random forests model with (RMSE = 0.1133, R2 = 0.9548 and MAE= 0.0665), the artificial neural network (ANN) model with (RMSE = 0.1320, R2 = 0.9386 and MAE = 0.0841), and the conventional multiple linear regression technique with (RMSE = 0.1361, R2 = 0.9345 and MAE = 0.0888). The results indicated that the Cubist and random forests models are able to generate better predictive results of the shear strength of RFM than ANN and conventional regression models. The Cubist model was considered to be more promising for interpreting the complex relationships between the influential properties of RFM and the shear strengths of RFM to some extent, which can be extremely helpful in estimating the shear strength of rockfill materials.
A prognosis model has been developed for solid waste generation from households in Hoi An City, a famous tourist city in Viet Nam. Waste sampling, followed by a questionnaire survey, was carried out to gather data. The Bayesian model average method was used to identify factors significantly associated with waste generation. Multivariate linear regression analysis was then applied to evaluate the impacts of significant factors on household waste production. The model obtained from this study indicated that household location, household size, house area per person, and family economic activity are important determinants of the waste generation rate. The models could explain about 34% of the variation of the per capita daily waste generation rate. Diagnostic tests and model validation results showed that the regression model could provide reliable results of estimated household waste. The study revealed that per capita urban household waste generation is 70–80% higher compared to a rural household. The models also showed that if a family ran a business from home, the household waste generation rate would increase by about 35%. This result provides reliable information for better waste collection and management planning. Two other significant variables (family size and house area per capita) do not contribute much (less than 20%) to waste generation. Variables accounting for household income, presence of a garden, number of rooms in a house, and percentage of members of different ages were proven to be not significant. The study provides a reliable method for estimating household waste generation, providing decision makers useful information for waste management policy development.
This paper presents the development of a general regression neural network (GRNN) model for the prediction of annual municipal solid waste (MSW) generation at the national level for 44 countries of different size, population and economic development level. Proper modelling of MSW generation is essential for the planning of MSW management system as well as for the simulation of various environmental impact scenarios. The main objective of this work was to examine the potential influence of economy crisis (global or local) on the forecast of MSW generation obtained by the GRNN model. The existence of the so-called structural breaks that occur because of the economic crisis in the studied period (2000–2012) for each country was determined and confirmed using the Chow test and Quandt–Andrews test. Two GRNN models, one which did not take into account the influence of the economic crisis (GRNN) and another one which did (SB-GRNN), were developed. The novelty of the applied method is that it uses broadly available social, economic and demographic indicators and indicators of sustainability, together with GRNN and structural break testing for the prediction of MSW generation at the national level. The obtained results demonstrate that the SB-GRNN model provide more accurate predictions than the model which neglected structural breaks, with a mean absolute percentage error (MAPE) of 4.0 % compared to 6.7 % generated by the GRNN model. The proposed model enhanced with structural breaks can be a viable alternative for a more accurate prediction of MSW generation at the national level, especially for developing countries for which a lack of MSW data is notable.
Predicting the mass of solid waste generation plays an important role in integrated solid waste management plans. In this study, the performance of two predictive models, Artificial Neural Network (ANN) and Multiple Linear Regression (MLR) was verified to predict mean Seasonal Municipal Solid Waste Generation (SMSWG) rate. The accuracy of the proposed models is illustrated through a case study of 20 cities located in Fars Province, Iran. Four performance measures, MAE, MAPE, RMSE and R were used to evaluate the performance of these models. The MLR, as a conventional model, showed poor prediction performance. On the other hand, the results indicated that the ANN model, as a non-linear model, has a higher predictive accuracy when it comes to prediction of the mean SMSWG rate. As a result, in order to develop a more cost-effective strategy for waste management in the future, the ANN model could be used to predict the mean SMSWG rate.
A strategy of household waste data analysis is proposed in this work, based on the WEKA workbench, developed on the Pattern Recognition community. The analysis was conducted on data collected from homes at residential areas in the city of Mexicali, México. The data included information about solid waste produced and also the householder's commitment to the environment, assessed in a Likert's scale based on a questionnaire. A cluster analysis and a tree classifier constructed using the clustered data are presented. An analysis of the decision tree allowed to translate the resulting tree in a set of production rules. After the interpretation of these rules, we were able to predict an environmental behavior, based on the information about waste generation and the questionnaire answers. The rules showed a tendency to bring together members of the same family, concurring in all cases the reference to the same period of waste generation. The elements identified on each rule indicate that the socioeconomic stratum is an important factor, related to behavioral attributes and consumption habits, the main relationship is based on attributes of waste generation as glass, organics, paper, inert, mixer containers and sanitary. The discovered relationships between the cluster, socio-demographic, behavioral and waste attributes are presented and discussed.
Quantity prediction of municipal solid waste (MSW) is crucial for design and programming municipal solid waste management system (MSWMS). Because effect of various parameters on MSW quantity and its high fluctuation, prediction of generated MSW is a difficult task that can lead to enormous error. The works presented here involve developing an improved support vector machine (SVM) model, which combines the principal component analysis (PCA) technique with the SVM to forecast the weekly generated waste of Mashhad city. In this study, the PCA technique was first used to reduce and orthogonalize the original input variables (data). Then these treated data were used as new input variables in SVM model. This improved model was evaluated by using weekly time series of waste generation (WG) and the number of trucks that carry waste in week of t. These data have been collected from 2005 to 2008. By comparing the predicted WG with the observed data, the effectiveness of the proposed model was verified. Therefore, in authors' opinion, the model presented in this article is a potential tool for predicting WG and has advantages over the traditional SVM model. © 2008 American Institute of Chemical Engineers Environ Prog, 2009
The waste management processes typically involve numerous technical, climatic, environmental, demographic, socio-economic, and legislative parameters. Such complex nonlinear processes are challenging to model, predict and optimize using conventional methods. Recently, artificial intelligence (AI) techniques have gained momentum in offering alternative computational approaches to solve solid waste management (SWM) problems. AI has been efficient at tackling ill-defined problems, learning from experience, and handling uncertainty and incomplete data. Although significant research was carried out in this domain, very few review studies have assessed the potential of AI in solving the diverse SWM problems. This systematic literature review compiled 85 research studies, published between 2004 and 2019, analyzing the application of AI in various SWM fields, including forecasting of waste characteristics, waste bin level detection, process parameters prediction, vehicle routing, and SWM planning. This review provides comprehensive analysis of the different AI models and techniques applied in SWM, application domains and reported performance parameters, as well as the software platforms used to implement such models. The challenges and insights of applying AI techniques in SWM are also discussed.
Environmentally sound management of plastic packaging waste is an issue of concern around the world because it causes potential threats to oceans and the environment upon disposal and mismanagement. This study examines the current efforts on recycling of the waste by extended producer responsibility (EPR) in South Korea as well as other countries. Material flow analysis (MFA) was performed on plastic packaging by life cycle. Based on the results in this study, material footprint of common single use plastics (i.e., PET water bottles, plastic cups, plastic bags, and plastic containers and cutlery by food delivery) by consumption was estimated to be on average 11.8 kg or 638 disposable plastics per capita a year, resulting in 32.6 billion disposable plastics and 603,000 ton of waste for disposal in South Korea. Approximately, 3 million ton of plastic packaging waste from household waste streams in 2017 in South Korea was generated and treated by energy recovery with solid refuse fuels and heat recovery, incineration without energy recovery, material recycling, and landfilling. Material recycling and recovery rates of plastic packaging waste from households were relatively low at 13.5% and 50.5%, respectively. It was estimated that as much as 3.6 million ton of CO2eq was generated from 2.7 million ton of plastic waste by incineration in 2017. Approximately 6.6 million ton CO2eq could be avoided by material recycling. Challenges and efforts have been discussed to improve current recycling system of plastic packaging waste towards a circular economy.
There are existing inadequate and ineffective practices that are not only common in Vietnam but also explicit in each municipal area. This study compares the municipal solid waste management attributes of cities in Vietnam under uncertainty. The uncertainties include the interrelationships among the attributes, linguistic preferences and qualitative information on the attributes. This study applies exploratory factor analysis to test the validity and reliability of the proposed attributes. Fuzzy set theory is used to translate the linguistic references into the qualitative attributes of municipal solid waste management. The decision-making trial and evaluation laboratory is used to address the interrelationships among the attributes. This study identifies the causal interrelationships among attributes using qualitative information, and a set of 14 attributes is defined and found to be valid and reliable for measurement. The results show that technical integration and social acceptability are the aspects that drive municipal solid waste management. Treatment innovations, safety and health, economic benefits, and technology functionality and appropriateness are determined to be the linkage criteria. The distinctions between cities are identified, Hanoi focuses on the institutional and organizational administration framework, whereas resource efficiency is an aspect of specific concern in Danang, and Ho Chi Minh City prioritized financial and operational requirements and facilities and infrastructure requirements. The implications for theory and practice are discussed.
Although industrial accident rates are gradually decreasing in Korea, the construction industry's accident rate is still higher compared with other industries. Human errors, mentally unstable workers, insufficient safety training, and safety policy affect the occurrence of construction accidents. Owing to the characteristics of this industry, occupational accident types, such as fall from height, collision with objects, rollover, and those due to falling objects, can be related to the weather data.
Therefore, to reduce and prevent occupational injury, it is necessary to classify and predict occupational accident types in detail. In this study, we built a model to classify and predict occupational accident types using a random forest (RF). We extracted important factors that affect the occupational accident types at construction sites using feature importance, and we analyzed the relationship between these factors and occupational accident types. The accuracy score of the RF model was obtained as 71.3%, and we presented key construction safety factors considering the feature importance. For future research, we will collect data and develop models to predict occupational accident types in real-time. Real-time construction accident prediction research will reduce accident at construction sites.
Growing world population along with fast economic growth and increased living standards have increased the municipal waste generation making its management be a foremost global issue. The problem is even more serious in urban areas as its improper management prompts tainting of soil, water, and environment which create public health risks. These problems of waste disposal and management were usually assessed by traditional methods which require loads of data. The recent development in the new software technologies and Internet along with the introduction of gradually more compact and dependable hardware products have presented the ability to accurately deal with these procedures more easily than costly and tedious field experiments. This paper presents an outline of the utilization of different scientific models in solving the environmental problems of municipal waste disposal. The examination of past literature uncovered that usually optimization models were used to find the answer of 'what is the best' under an explicit arrangement of conditions, while, simulation models were usually helpful to get an answer to 'what if? ' due to their predictive capability. An indication of the municipal waste disposal problems and its management alongside the ramifications of the investigation is provided. The rationale and backdrop of the waste disposal issues are described. The applications of optimization modeling, multi-objective approach, multi-criteria decision analysis, and artificial neural networks in waste management are presented and applications of these modeling techniques in diverse case studies
worldwide are described. And finally, the conclusions of the analysis are summarized.
Machine learning methods are now an important tool for scientists, researchers, engineers and students in a wide range of areas. This book is written for people who want to adopt and use the main tools of machine learning, but aren’t necessarily going to want to be machine learning researchers. Intended for students in final year undergraduate or first year graduate computer science programs in machine learning, this textbook is a machine learning toolkit. Applied Machine Learning covers many topics for people who want to use machine learning processes to get things done, with a strong emphasis on using existing tools and packages, rather than writing one’s own code.
A companion to the author's Probability and Statistics for Computer Science, this book picks up where the earlier book left off (but also supplies a summary of probability that the reader can use).
• Emphasizing the usefulness of standard machinery from applied statistics, this textbook gives an overview of the major applied areas in learning
• Covers the ideas in machine learning that everyone going to use learning tools should know, whatever their chosen specialty or career.
• Broad coverage of the area ensures enough to get the reader started, and to realize that it’s worth knowing more in-depth knowledge of the topic.
• Practical approach emphasizes using existing tools and packages quickly, with enough pragmatic material on deep networks to get the learner started without needing to study other material.
Purpose:
A successful hospital solid waste (HSW) management needs an accurate estimation of waste generation rates. The conventional regression methods upon increasing the number of input variables hardly can predict the HSW generation rate and require more complex modeling. In return, application of machine learning methods seems to be able to increase the power of predicting the produced wastes.
Methods:
To predict the HSW, Multiple Linear Regression(MLR) and several Neuron- and Kernel-based machine learning methods were employed to analyze data from hospitals of Karaj metropolis. The number of wards, active and occupied beds, staffs and inpatients, and ownership type and activity years of hospital were defined as the model inputs. In addition, proposed models performance was evaluated based on coefficient of determination (R2) and Mean-Square Error (MSE).
Results:
The performance of Neuron- and Kernel-based machine learning methods indicated that both models were satisfactory in predicting HSW. However, the better results of 0.82–0.86 for average R2 value and 0.003–0.008 for average MSE value, indicated relative superiority of Kernel-based models compared to Neuron based (average R2 = 0.68–0.74, average MSE = 0.009–0.023) and MLR models. Number of staffs and hospital ownership type were the most influential model variables in predicting the HSW generation rate.
Conclusions:
The machine learning methods could interpret the relationship between waste generation rate and model inputs, appropriately. Thus, they may play an effective role in developing cost-effective methods for suitable HSW management.
We used the experiences of Vietnam following the economic reform, known as Doimoi, to study urbanization, economic development, and environmental and social changes in transitional economies at multi-scales. The country underwent rapid urban land expansion, as indicated by the increase in the mean value of nighttime light data from −1.4 in 1992 to 4.4 in 2012. The urban population grew at a faster annual rate following Doimoi (1986–2015) compared to the pre-Doimoi period (1960–1985). At the inter-city level, cities with populations more than 1 million experienced more rapid growth of built-up land intensity and population size compared to the national average. At the intra-city level, conversion from farmland contributed significantly to built-up land in Hanoi and Ho Chi Minh City from 1990 to 2010. As indicated by PM2.5 and NO2 concentrations, urban environments in large cities deteriorated; yet poverty was alleviated, as measured by populations falling under the poverty line and the proportion of the urban population living in slums. Coupled dynamics of urbanization, economic development, and environmental and social changes were modeled and the main findings are: (1) economic development strongly influenced urbanization and (2) urbanization and economic development contributed to environmental deterioration while promoting the social conditions. How urban land expansion was facilitated by local institutional interventions such as frequent changes of administrative boundaries, master plans, and policies is also discussed. Our study highlighted a multi-scale and multi-dimensional perspective, the independent and coupled relationships between economic development, urbanization, and environmental/social changes, and a hybrid approach of examining the influences of the institutional intervention and the market mechanism on urbanization in transitional economies.
Plastic waste generation is an inevitable product of human activities, however its management faces challenges in many cities. Understanding the existing patterns of plastic waste generation and recycling is essential for effective management planning. The present study established a relationship between plastic waste generation rate and the identified socioeconomic groups, higher socioeconomic group (HSEG), middle socioeconomic group (MSEG), and lower socioeconomic group (LSEG) of the study area(Dhanbad, India). For identification of the socioeconomic groups, four different socioeconomic parameters were considered (total family income, education, occupation and type of houses). The information related to the identified parameters were obtained using questionnaire survey conducted in the selected households. One week plastic waste sampling was carried out in the house holds of all the socioeconomic groups. The plastic waste generated in the study area was 5.7% of the total municipal solid waste. In terms of total plastic waste generation rate, it was found that HSEG had maximum (51 g/c/d) and LSEG had minimum (8 g/c/d) generation rate. The present study area does not have any formal waste recycling system. Thus, the amount of plastic waste recovered and the revenue generated from recycling of plastic waste by the active informal recyclers (waste pickers, itinerant waste buyers and scrap dealers) in the study area have been evaluated. Additionally, three non-linear machine learning models i.e., artificial neural network (ANN), support vector machine (SVM) and random forest (RF) have been developed and compared for the prediction of plastic waste generation rate.
Electronic waste has become one of the fastest growing waste streams in Vietnam, of which adequate information on its size and its flows is very crucial to efficiently manage it and prevent associated environmental problems. To obtain such information, Material Flow Analysis (MFA) was used to quantify and systematically analyze the flow of obsolete TVs from households in urban areas in Vietnam. The results showed a gradual increase of obsolete TVs during the period 1966-2035. Further investigation showed that 66% of the total obsolete TVs in 2012 was directly reused or reused after repair/refurbishing, 3% was domestically recycled or open burned to recover valuable materials, 9% was illegally exported and the rest of 22% was open dumped. Substance flow analysis showed that 75% of total base metals (i.e., Cu, Al, Fe/steel) contained in obsolete TVs was reused or recycled; the rest (25%) was exported or emitted, representing the loss of materials for the Vietnamese economy. For precious metals (i.e., Au, Ag, Pd), plastic and glass, a larger material leaching was noticed. About 34% was illegally exported (in case of precious metals and plastics) or open dumped (in case of glass). The analysis also revealed the heavy involvement of the informal sector in the TV waste management system, making it more complicated, difficult to control and resulting in potential risks to the environment and human health. Based on the above analysis, an integrated management system was proposed to manage the secondary material source and prevent potential harmful effects.
In the developing countries, the inadequacy of basic waste data is a significant obstacle for municipal solid waste management. To evaluate an effective waste management plan, identification of influencing socio-economic factors and projection of municipal solid waste generation (MSWG) plays a crucial role. Yet, several forecasting methods have been utilized to quantify future MSWG. In this study, we investigated the influencing socio-economic factors for MSWG in China using the fuzzy logic method, and a short-term forecasting of MSWG was conducted using multi-model approach. The Grey (1, 1), linear regression, and artificial neural network (ANN) models were evaluated for short-term forecasting. The factor analysis results show that urban population growth is the most influencing socio-economic factor for MSWG, and the influence of GDP on waste generation is not so obvious. Afterward, the multi-model forecasting results indicate an increasing trend of MSWG. Based on the absolute percentage error, root mean-squared error, mean absolute error, and coefficient of determination (R²), ANN found as the most acceptable model to forecast MSWG in China. Hence, the forecasting by ANN model illustrates that MSWG in China will be 24666.65 (10⁴ tons) by 2030.
Municipal solid waste management represents an increasingly significant environmental, fiscal, and social challenge for cities. Understanding patterns of municipal waste generation behavior at the household and building scales is a critical component of efficient collection routing and the design of incentives to encourage recycling and composting. However, high spatial resolution estimates of building refuse and recycling have been constrained by the lack of granular data for individual properties. This paper presents a new analytical approach, which combines machine learning and small area estimation techniques, to predict weekly and daily waste generation at the building scale. Using daily collection data from 609 New York City Department of Sanitation (DSNY) sub-sections over ten years, together with detailed data on individual building attributes, neighborhood socioeconomic characteristics, weather, and selected route-level collection data, we apply gradient boosting regression trees and neural network models to estimate daily and weekly refuse and recycling tonnages for each of the more than 750,000 residential properties in the City. Following cross-validation and a two-stage spatial validation, our results indicate that our method is capable of predicting building-level waste generation with a high degree of accuracy. Our methodology has the potential to support collection truck route optimization based on expected building-level waste generation rates, and to facilitate new equitable solid waste management policies to shift behavior and divert waste from landfills based on benchmarking and peer performance comparisons.
High-dimensional data analysis is a challenge for researchers and engineers in the fields of machine learning and data mining. Feature selection provides an effective way to solve this problem by removing irrelevant and redundant data, which can reduce computation time, improve learning accuracy, and facilitate a better understanding for the learning model or data. In this study, we discuss several frequently-used evaluation measures for feature selection, and then survey supervised, unsupervised, and semi-supervised feature selection methods, which are widely applied in machine learning problems, such as classification and clustering. Lastly, future challenges about feature selection are discussed.
The main objective of this study was to develop models for accurate prediction of municipal solid waste (MSW) generation and diversion based on demographic and socio-economic variables, with planned application of generating Canada-wide MSW inventories. Models were generated by mapping residential MSW quantities with socio-economic and demographic parameters of 220 municipalities in the province of Ontario, Canada. Two machine learning algorithms, namely decision trees and neural networks, were applied to build the models. Socio-economic variables were derived from Canadian Census data at regional and municipal levels. A data pre-processing and integration framework was developed in Matlab® computing software to generate datasets with sufficient data quantity and quality for modeling. Results showed that machine learning algorithms can be successfully used to generate waste models with good prediction performance. Neural network models had the best performance, describing 72% of variation in the data. The approach proposed in this study demonstrates the feasibility of creating tools that helps in regional waste planning by means of sourcing, pre-processing, integrating and modeling of publically available data from various sources.
The management of municipal solid waste (MSW) is one of the main costs incurred by local authorities in developing countries. According to some estimates, these costs can account for up to 50% of city government budgets. It is therefore of importance that policymakers, urban planners and practitioners have an adequate understanding of what these costs consist of, from collection to final waste disposal. This article focuses on a specific stage of the MSW value chain, the treatment of waste, and it aims to identify cost patterns associated with the implementation and operation of waste treatment approaches in developing Asian countries. An analysis of the capital (CAPEX) and operational expenditures (OPEX) of a number of facilities located in countries of the region was conducted based on a database gathering nearly 100 projects and which served as basis for assessing four technology categories: composting, anaerobic digestion (AD), thermal treatment, and the production of refuse-derived fuel (RDF). Among these, it was found that the least costly to invest, asa function of the capacity to process waste, are composting facilities, with an average CAPEX per ton of 21,493 USD2015/ton. Conversely, at the upper end featured incineration plants, with an average CAPEX of 81,880 USD2015/ton, with this treatment approach ranking by and large as the most capital intensive of the four categories assessed. OPEX figures of the plants, normalized and analyzed in the form of OPEX/ton, were also found to be higher for incineration than for biological treatment methods, although on this component differences amongst the technology groups were less pronounced than those observed for CAPEX. While the results indicated the existence of distinct cost implications for available treatment approaches in the developing Asian context, the analysis also underscored the importance of understanding the local context asa means to properly identify the cost structure of each specific plant. Moreover, even though CAPEX and OPEX figures are important elements to assess the costs of a waste treatment system, these should not be considered on a standalone basis for decision making purposes. In complement to this internal cost dimension, the broader impacts - to the economy, society and the environment - resulting from the adoption of a certain treatment approach should be properly understood and, ideally, measured and expressed in monetary terms.
Accurate prediction of the quantity of household solid waste generation is very much essential for effective management of municipal solid waste (MSW). In actual practice, modelling methods are often found useful for precise prediction of MSW generation rate. In this study, two models have been proposed that established the relationships between the household solid waste generation rate and the socioeconomic parameters, such as household size, total family income, education, occupation and fuel used in the kitchen. Multiple linear regression technique was applied to develop the two models, one for the prediction of biodegradable MSW generation rate and the other for non-biodegradable MSW generation rate for individual households of the city Dhanbad, India. The results of the two models showed that the coefficient of determinations (R²) were 0.782 for biodegradable waste generation rate and 0.676 for non-biodegradable waste generation rate using the selected independent variables. The accuracy tests of the developed models showed convincing results, as the predicted values were very close to the observed values. Validation of the developed models with a new set of data indicated a good fit for actual prediction purpose with predicted R² values of 0.76 and 0.64 for biodegradable and non-biodegradable MSW generation rate respectively.
Understand deep learning, the nuances of its different models, and where these models can be applied.
The abundance of data and demand for superior products/services have driven the development of advanced computer science techniques, among them image and speech recognition. Introduction to Deep Learning Using R provides a theoretical and practical understanding of the models that perform these tasks by building upon the fundamentals of data science through machine learning and deep learning. This step-by-step guide will help you understand the disciplines so that you can apply the methodology in a variety of contexts. All examples are taught in the R statistical language, allowing students and professionals to implement these techniques using open source tools.
What You Will Learn:
• Understand the intuition and mathematics that power deep learning models
• Utilize various algorithms using the R programming language and its packages
• Use best practices for experimental design and variable selection
• Practice the methodology to approach and effectively solve problems as a data scientist
• Evaluate the effectiveness of algorithmic solutions and enhance their predictive power
The Handbook of Statistical Analysis and Data Mining Applications is a comprehensive professional reference book that guides business analysts, scientists, engineers and researchers (both academic and industrial) through all stages of data analysis, model building and implementation. The Handbook helps one discern the technical and business problem, understand the strengths and weaknesses of modern data mining algorithms, and employ the right statistical methods for practical application. Use this book to address massive and complex datasets with novel statistical approaches and be able to objectively evaluate analyses and solutions. It has clear, intuitive explanations of the principles and tools for solving problems using modern analytic techniques, and discusses their application to real problems, in ways accessible and beneficial to practitioners across industries - from science and engineering, to medicine, academia and commerce. This handbook brings together, in a single resource, all the information a beginner will need to understand the tools and issues in data mining to build successful data mining solutions. Written "By Practitioners for Practitioners" Non-technical explanations build understanding without jargon and equations Tutorials in numerous fields of study provide step-by-step instruction on how to use supplied tools to build models using Statistica, SAS and SPSS software Practical advice from successful real-world implementations Includes extensive case studies, examples, MS PowerPoint slides and datasets CD-DVD with valuable fully-working 90-day software included: "Complete Data Miner - QC-Miner - Text Miner" bound with book.
Historical municipal solid waste (MSW) collection data supplied by the New York City Department of Sanitation (DSNY) was used in conjunction with other datasets related to New York City to forecast municipal solid waste generation across the city. Spatiotemporal tonnage data from the DSNY was combined with external data sets, including the Longitudinal Employer Household Dynamics data, the American Community Survey, the New York City Department of Finance’s Primary Land Use and Tax Lot Output data, and historical weather data to build a Gradient Boosting Regression Model. The model was trained on historical data from 2005 to 2011 and validation was performed both temporally and spatially. With this model, we are able to accurately ( ) forecast weekly MSW generation tonnages for each of the 232 geographic sections in NYC across three waste streams of refuse, paper and metal/glass/plastic. Importantly, the model identifies regularity of urban waste generation and is also able to capture very short timescale fluctuations associated to holidays, special events, seasonal variations, and weather related events. This research shows New York City’s waste generation trends and the importance of comprehensive data collection (especially weather patterns) in order to accurately predict waste generation.
For an adequate planning of waste management systems the accurate forecast of waste generation is an essential step, since various factors can affect waste trends. The application of predictive and prognosis models are useful tools, as reliable support for decision making processes. In this paper some indicators such as: number of residents, population age, urban life expectancy, total municipal solid waste were used as input variables in prognostic models in order to predict the amount of solid waste fractions. We applied Waste Prognostic Tool, regression analysis and time series analysis to forecast municipal solid waste generation and composition by considering the Iasi Romania case study. Regression equations were determined for six solid waste fractions (paper, plastic, metal, glass, biodegradable and other waste). Accuracy Measures were calculated and the results showed that S-curve trend model is the most suitable for municipal solid waste (MSW) prediction.
Municipal solid waste (MSW) management is a major concern to local governments to protect human health, the environment and to preserve natural resources. The design and operation of an effective MSW management system requires accurate estimation of future waste generation quantities. The main objective of this study was to develop a model for accurate forecasting of MSW generation that helps waste related organizations to better design and operate effective MSW management systems. Four intelligent system algorithms including support vector machine (SVM), adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN) and k-nearest neighbours (kNN) were tested for their ability to predict monthly waste generation in the Logan City Council region in Queensland, Australia. Results showed artificial intelligence models have good prediction performance and could be successfully applied to establish municipal solid waste forecasting models. Using machine learning algorithms can reliably predict monthly MSW generation by training with waste generation time series. In addition, results suggest that ANFIS system produced the most accurate forecasts of the peaks while kNN was successful in predicting the monthly averages of waste quantities. Based on the results, the total annual MSW generated in Logan City will reach 9.4 × 107 kg by 2020 while the peak monthly waste will reach 9.37 × 106 kg.
Deep neural networks with several layers have recently become a highly successful and popular research topic in machine learning due to their excellent performance in many benchmark problems and applications. A key idea in deep learning is to learn not only the nonlinear mapping between the inputs and outputs but also the underlying structure of the data (input) vectors. In this chapter, we first consider problems with training deep networks using backpropagation-type algorithms. After this, we consider various structures used in deep learning, including restricted Boltzmann machines, deep belief networks, deep Boltzmann machines, and nonlinear autoencoders. In the latter part of this chapter, we discuss in more detail the recently developed neural autoregressive distribution estimator and its variants.
Up to now, there has been no robust, homogenized and standardized method for the design of mechatronic systems. This chapter presents a number of methods to optimize the reliability of mechatronic systems. These methods are based on knowledge and expertise in deterministic and stochastic modeling. The objective is to combine a finite element numerical model describing the physical behavior of a mechatronic system with a stochastic model of its behavior. The results of numerical modeling make it possible to build a meta-model using the response surface. By using this meta-model, the level of control factors is adjusted, the sensitivity of the mechatronic system to sources of variability is reduced (noise factors) and the response of the system can be adjusted to meet the target (objective).
Data preprocessing techniques generally refer to the addition, deletion, or transformation of the training set data. Preprocessing data is a crucial step prior to modeling since data preparation can make or break a model’s predictive ability. To illustrate general preprocessing techniques, we begin by introducing a cell segmentation data set (Section 3.1). This data set contains common predictor problems such as skewness, outliers, and missing values. Sections 3.2 and 3.3 review predictor transformations for single predictors and multiple predictors, respectively. In Section 3.4 we discuss several approaches for handling missing data. Other preprocessing steps may include removing (Section 3.5), adding (Section 3.6), or binning (Section 3.7) predictors, all of which must be done carefully so that predictive information is not lost or erroneous information is added to the data. The computing section (3.8) provides R syntax for the previously described preprocessing steps. Exercises are provided at the end of the chapter to solidify concepts.
This study was undertaken to evaluate the characteristics of household solid waste (HSW) generation and to identify opportunities and benefits for waste recycling in a typical developed city of Suzhou in East China. A four-stage systematic tracking survey of 240 households was conducted for one week in each season starting from the summer of 2011 to the spring of 2012. And the driving forces behind HSW generation were analyzed using a multiple linear regression model. Results show that Suzhou's HSW generation rate was 280.5 g/cap/day, and the annual generation of HSW reached 568 thousand tons, among which, 89.3% were compostable and recyclable waste. Education level of the household daily manager has a major impact on HSW generation rate. And other factors, such as local customs and culture, consumption patterns and residential lifestyles could also influence HSW generation. It could achieve annual economic benefit of 15.9 million RMB, reduce 32.6 million tons of CO2 equivalent emissions, and supply nearly 3500 job opportunities in Suzhou if source separation practice well. Implications of our results for HSW management in Suzhou and other Chinese cities were also discussed.
The quantities and composition of municipal solid waste (MSW) are important factors in the planning and management of MSW. Daily human activities were classified into three groups: maintenance activities (meeting the basic needs of food, housing and personal care, MA); subsistence activities (providing the financial support requirements, SA); and leisure activities (social and recreational pursuits, LA). A model, based on the interrelationships of expenditure on consumer goods, time distribution, daily activities, residents groups, and waste generation, was employed to estimate MSW generation by different activities and resident groups in five provinces (Zhejiang, Guangdong, Hebei, Henan and Sichuan) of China. These five provinces were chosen for this study and the distribution patterns of MSW generated by different activities and resident groups were revealed. The results show that waste generation in SA and LA fluctuated slightly from 2003 to 2008. For general waste generation in the five provinces, MA accounts for more than 70% of total MSW, SA approximately 10%, and LA between 10% and 16% by urban residents in 2008. Females produced more daily MSW than males in MA. Males produced more daily MSW than females in SA and LA. The wastes produced at weekends in MA and LA were far greater than on weekdays, but less than on weekdays for SA wastes. Furthermore, one of the model parameters (the waste generation per unit of consumer expenditure) is inversely proportional to per-capita disposable income of urban residents. A significant correlation between gross domestic product (GDP) and waste generation by SA was observed with a high coefficient of determination.
Copyright © 2015 Elsevier Ltd. All rights reserved.
The increasing construction and demolition (C&D) waste causes both cost inefficiency and environmental pollution. Many countries have developed regulations to minimize C&D waste. Implementation of these regulations requires an understanding of the magnitude and material composition of waste stream. Construction waste generation index is a useful tool for estimating the amount of construction waste and can be used as a benchmark to enhance the sustainable performance of construction industry. This paper presents a model for quantifying waste generation per gross floor area (WGA) based on mass balance principle for building construction in China. WGAs for major types of material are estimated using purchased amount of major materials and their material waste rate (MWR). The WGA for minor quantities of materials is estimated together as a percentage of total construction waste. The model is applied to a newly constructed residential building in Shenzhen city of South China. The WGA of this project is 40.7 kg/m2, and concrete waste is the largest contributor to the index. Comparisons with transportation records in site, empirical index in China and data in other economies reveal that the proposed model is valid and practical. The proposed model can be used to setup a benchmark WGA for Chinese construction industry by carrying out large-scale investigations in the future.
Random forests are a combination of tree predictors such that each tree depends on the values of a random vector sampled independently and with the same distribution for all trees in the forest. The generalization error for forests converges a.s. to a limit as the number of trees in the forest becomes large. The generalization error of a forest of tree classifiers depends on the strength of the individual trees in the forest and the correlation between them. Using a random selection of features to split each node yields error rates that compare favorably to Adaboost (Y. Freund & R. Schapire, Machine Learning: Proceedings of the Thirteenth International conference, ∗∗∗, 148–156), but are more robust with respect to noise. Internal estimates monitor error, strength, and correlation and these are used to show the response to increasing the number of features used in the splitting. Internal estimates are also used to measure variable importance. These ideas are also applicable to regression.
Forecasting-the art and science of predicting future outcomes-has become a crucial skill in business and economic analysis. This volume introduces the reader to the tools, methods, and techniques of forecasting, specifically as they apply to financial and investing decisions. With an emphasis on "earnings per share" (eps), the author presents a data-oriented text on financial forecasting, understanding financial data, assessing firm financial strategies (such as share buybacks and R&D spending), creating efficient portfolios, and hedging stock portfolios with financial futures. The opening chapters explain how to understand economic fluctuations and how the stock market leads the general economic trend; introduce the concept of portfolio construction and how movements in the economy influence stock price movements; and introduce the reader to the forecasting process, including exponential smoothing and time series model estimations. Subsequent chapters examine the composite index of leading economic indicators (LEI); review financial statement analysis and mean-variance efficient portfolios; and assess the effectiveness of analysts' earnings forecasts. Using data from such firms as Intel, General Electric, and Hitachi, Guerard demonstrates how forecasting tools can be applied to understand the business cycle, evaluate market risk, and demonstrate the impact of global stock selection modeling and portfolio construction. © Springer Science+Business Media New York 2013. All rights are reserved.
Most cities are actually very concerned about the economic viability of waste management and also about the impact they may have on the environment. Economical, social and cultural factors in the population will determine the characteristics in waste and the value of the design parameters used in the calculations of a collection system. A clear understanding of these factors is fundamental to plan and to implement efficient and sustainable collecting strategies. Our goal in this work is to model municipal waste separation rates in Spanish cities with over 50,000 inhabitants taking their different socio-economic, demographic and logistic covariates into account. Several statistical regression models to manage continuous proportion data are compared, these being: Generalized linear models (GLM) with Binomial, Poisson and Gamma errors after several transformations of the data and Beta regression on the raw data. The best fits are obtained by using GLM Gamma and beta regression. Significant covariates for the different separation rates are obtained from these models, and the strength of the influence of all these factors on the response variable is calculated. All these results could be taken into account to design and to evaluate selective collection systems, and will allow us to make predictions on cities not included in this study.
This study presents a new approach—preprocessing for reaching the stationary chain in time series—to unravel the interpolating problem of artificial neural networks (ANN) for long-term prediction of solid waste generation (SWG). To evaluate the accuracy of the prediction by ANN, comparison between the results of the multivariate regression model and ANN is performed. Monthly time series datasets, by the yrs 2000–2010, for the city of Mashhad, are used to simulate the generated solid waste. Different socioeconomic and environmental factors are assessed, and the most effective ones are used as input variables. The projections of these explanatory variables are used in the estimated model to predict the generated solid waste values through the yr 2032. Ultimately, various structures of ANN models are examined to select the best result based on the performance indices criteria. Research findings clearly indicate that such a new approach can be a practical method for long-term prediction by ANNs. Furthermore, it can reduce uncertainties and lead to noticeable increase in the accuracy of the long-term forecasting. Results indicate that multilayer perception approach has more advantages in comparison with traditional methods in predicting the municipal SWG. © 2011 American Institute of Chemical Engineers Environ Prog, 2011
The economically-viable and environmentally-acceptable disposal of municipal solid waste (MSW) is a major concern in many industrialised countries. The main problem facing policy makers in the waste management sector is how to predict the amount and the composition of MSW that is likely to be generated in the near future in order to devise the most appropriate treatment/disposal strategy. Published data on MSW arisings in European countries during the period 1980–1993 and those for the USA between 1960 and 1993 have been correlated with the corresponding figures for the gross domestic product (GDP) and population. The typical composition of MSW has been expressed in terms of the fraction of the total consumer expenditure on goods and products resulting in the generation of MSW, i.e. related total consumer expenditure (RTCE). A model linking RTCE to GDP has been developed and utilised to estimate the amounts of the individual fractions in the total MSW generated. The correlations permit highly-accurate predictions of the total amount of MSW arisings to be obtained both for European countries and the USA. Deviations between the predicted and measured values are, however, much lower in the case of the USA, because the corresponding model is based on data for a single country. Good matches between the predicted and measured figures for the individual fractions of the MSW also ensue. However, the model for predicting the individual fractions in the MSW for the European countries is based solely on information available for the UK.
Does public-service performance depend on privatisation policy? In the case of urban hygiene it would appear that firm dimension and inter-firm relationships, alternative technologies and landfill capacity are more important. This study focuses specifically on the Italian case and belongs to a series of comparative international studies. The general aim of the paper is to provide public-sector administrators with clear guidelines. Waste collection should evolve towards a highly capitalised, medium/large-sized waste separation system with a long-term view. At the same time, the landfill-based waste disposal model should develop into a recovery model. In order for this to happen the extension of collection areas should be increased. If these policies are introduced promptly, the system may gradually evolve towards higher environmental standards and lower management costs.
The uncertainty of the availability of water resources during the boreal winter has led to significant economic losses in recent years in Taiwan. A modified support vector machine (SVM) based prediction framework is thus proposed to improve the predictability of the inflow to Shihmen reservoir in December and January, using climate data from the prior period. Highly correlated climate precursors are first identified and adopted to predict water availability in North Taiwan. A genetic algorithm based parameter determination procedure is implemented to the SVM parameters to learn the non-linear pattern underlying climate systems more flexibly. Bagging is then applied to construct various SVM models to reduce the variance in the prediction by the median of forecasts from the constructed models. The enhanced prediction ability of the proposed modified SVM-based model with respect to a bagged multiple linear regression (MLR), simple SVM, and simple MLR model is also demonstrated. The results show that the proposed modified SVM-based model outperforms the prediction ability of the other models in all of the adopted evaluation scores. Copyright © 2009 Royal Meteorological Society
This paper concerns learning tasks that require the prediction of a continuous value rather than a discrete class. A general method is presented that allows predictions to use both instance-based and model-based learning. Results with three approaches to constructing models and with eight datasets demonstrate improvements due to the composite method.
This study was undertaken to evaluate the quantity and composition of household solid waste to identify opportunities for waste recycling in Can Tho city, the capital city of the Mekong Delta region in southern Vietnam. Two-stage survey of 100 households was conducted for dry season and rainy season in 2009. Household solid waste was collected from each household and classified into 10 physical categories and 83 subcategories. The average household solid waste generation rate was 285.28 g per capita per day. The compostable and recyclable shares respectively accounted for 80.02% and 11.73%. The authors also analyzed the relations between some socioeconomic factors and household solid waste generation rates by physical categories and subcategories. The household solid waste generation rate per capita per day was positively correlated with the population density and urbanization level, although it was negatively correlated with the household size. The authors also developed mathematical models of correlations between the waste generation rates of main physical categories and relevant factors, such as household size and household income. The models were proposed by linear models with three variables to predict household solid waste generation of total waste, food waste, and plastic waste. It was shown that these correlations were weak and a relationship among variables existed. Comparisons of waste generation by physical compositions associated with different factors, such as seasonal and daily variation were conducted. Results presented that the significant average differences were found by the different seasons and by the different days in a week; although these correlations were weak. The greenhouse gas baseline emission was also calculated as 292.25 g (CO(2) eq.) per capita per day from biodegradable components.