Search

In this paper, a new responsive-green-cold vaccine supply chain network during the COVID-19 pandemic is developed for the first time. According to the proposed network, a new multi-objective, multi-period, multi-echelon mathematical model for the distribution-allocation-location problem is designed. Another important novelty in this paper is that it considers an Internet-of-Things application in the COVID-19 condition in the suggested model to enhance the accuracy, speed, and justice of vaccine injection with existing priorities. Waste management, environmental effects, coverage demand, and delivery time of COVID-19 vaccine simultaneously are therefore considered for the first time. The LP-metric method and meta-heuristic algorithms called Gray Wolf Optimization (GWO), and Variable Neighborhood Search (VNS) algorithms are then used to solve the developed model. The other significant contribution, based on two presented meta-heuristic algorithms, is a new heuristic method called modified GWO (MGWO), and is developed for the first time to solve the model. Therefore, a set of test problems in different sizes is provided. Hence, to evaluate the proposed algorithms, assessment metrics including (1) percentage of domination, (2) the number of Pareto solutions, (3) data envelopment analysis, and (4) diversification metrics and the performance of the convergence are considered. Moreover, the Taguchi method is used to tune the algorithm's parameters. Accordingly, to illustrate the efficiency of the model developed, a real case study in Iran is suggested. Finally, the results of this research show MGO offers higher quality and better performance than other proposed algorithms based on assessment metrics, computational time, and convergence. [ABSTRACT FROM AUTHOR]

By accurate predicting of pipe bursts, it is possible to schedule pipe maintenance, rehabilitation and improve level of services in Water Distribution Networks (WDNs). In recent years, artificial intelligence techniques have been used vastly in pipe burst prediction. These techniques need big input dataset for deep learning but big dataset might not be available in old WDNs. This paper aimed to compare the performance of multivariate adaptive regression splines (MARS) and regressive convolution neural network with least square support vector regression (RCNN-LSSVR) for predicting of Pipe Burst Rate (PBR) with limited dataset. Pipes age, diameter, depth of installation, length, average and maximum hydraulic pressure are considered as effective parameters of PBR. Data collected from real-case study in Iran include 158 cases for polyethylene and 124 cases for asbestos cement pipes during 2012–2019. The results indicate that RCNN-LSSVR model has a great performance of PBR prediction on limited dataset. [ABSTRACT FROM AUTHOR]

This paper employs machine learning algorithms to detect tax evasion and analyzes tax data. With the development of commercial businesses, traditional algorithms are not appropriate for solving the tax evasion detection problem. Hence, other algorithms with acceptable speed, precision, analysis, and data decisions must be used. In the case of assets and tax assessment, the integration of machine learning models with meta-heuristic algorithms increases accuracy due to optimal parameters. In this paper, intelligent machine learning algorithms are used to solve tax evasion detection. This research uses an improved particle swarm optimization (IPSO) algorithm to improve the multilayer perceptron neural network by finding the optimal weight and improving support vector machine (SVM) classifiers with optimal parameters. The IPSO-MLP and IPSO-SVM models using the IPSO algorithm are used as new models for tax evasion detection. Our proposed system applies the dataset collected from the general administration of tax affairs of West Azerbaijan province of Iran with 1500 samples for the tax evasion detection problem. The evaluations show that the IPSO-MLP model has a higher accuracy rate than the IPSO-SVM model and logistic regression. Moreover, the IPSO-MLP model has higher accuracy than SVM, Naive Bayes, k-nearest neighbor, C5.0 decision tree, and AdaBoost. The accuracy of IPSO-MLP and IPSO-SVM models is 93.68% and 92.24%, respectively. [ABSTRACT FROM AUTHOR]

The concept of leveraging the predictive capacity of predisposing factors for landslide susceptibility (LS) modeling has been continuously improved in recent work focusing on computational and machine learning algorithms. This paper explores the predictive capacity of different approaches to LS modelling using artificial intelligence. The key objective of this study is to estimate a LS map for the Taleghan-Alamut basin of Iran using Credal Decision Tree (CDT)-based (i.e. CDT-Bagging, CDT-Multiboost and CDT-SubSpace) hybrid machine learning approaches, which are state-of-the-art soft computing approaches that are hardly ever utilized in the assessment of LS. In this study, we used eighteen landslide predisposing factors (LPFs) that we considered to be the most important local morphological and geo-environmental factors influencing the occurrence of landslides. We calculated the significance of each of the LPFs in the landslide susceptibility assessment using the Random Forest Method. We also employed the Receiver Operating Characteristic curve, precision, performance, map robustness measurement and selection of the best-fitting models. The results shows that, compared to the other models, the CDT-Multiboost is the excellent model in this perspective with an average area under curve (AUC) of 0.993 based on a 4-fold cross-validation. We, therefore, consider the CDT-Multiboost models to be an effective method for improving spatial prediction of LS where landslide scarps or bodies are not clearly identified during the preparation of landslide inventory maps. Therefore, it will be helpful for preparing future landslide inventory maps and mitigate landslide damages. [ABSTRACT FROM AUTHOR]

In this paper, the monthly long-term precipitation of the city of Ardabil from 1976 to 2020 is simulated by a modern hybrid learning machine. To this end, the Wavelet and Outlier Robust Extreme Learning Machine (ORELM) models are integrated to produce a hybrid model called "Wavelet-Outlier Robust Extreme Learning Machine (WORELM)". First, the observed data are normalized, and the best normalization coefficients for the study are acquired. About 70% of the observed data are employed to train the artificial intelligence models and the remaining (i.e., 30%) to test them. After that, the optimal mother wavelet and the best decomposition level of the wavelet model are computed. Then, the optimal number of the hidden layer neurons and the best activation function of the ORELM model is chosen by performing a trial and error procedure. In this study, the regularization parameter of the ORELM model is also optimized. Moreover, using the autocorrelation function (ACF), the most influencing lags of the time-series data are detected, and 14 WORELM models are developed. By analyzing the WORELM models, the best model and the effective lags of the time-series data are introduced. The outcomes of the hybrid WORELM model are finally compared with other machine learning (ML) algorithms to prove the superiority of the WORELM model. [ABSTRACT FROM AUTHOR]

A scientometric approach is utilized to investigate the dynamic maps of relationships among researchers, institutes, and countries in the field of Artificial Intelligence for Fluid-flow and Heat-transfer (AIFH). The Web of Science database was searched for related publications during the last 40 years (1982 and 2022). A total of 6151 articles were discovered, which were analyzed in detail. Using a bibliometric analysis, the most relevant and most cited sources of publications were identified. The most active researchers, institutions, and countries leading AIFH were reported. Then, the worldwide dynamic collaboration maps and coupling maps of relationships were reported. The Islamic Azad University (1893 T.C.), the Chinese Academy of Sciences (1374 T.C.), and Beihang University (1266 T.C.) were the most influential institutes in AIFH. The most influential countries were China, the USA, and Iran. The dynamic map of collaborations shows a good worldwide collaboration distribution. The USA and China established the most connection with the rest of the world. ANNs are the most studied topic (19.5% of publications), followed by Machine Learning (17.9%) and Neural Networks (15.4%). Support Vector Machines lag behind at 1.4%. ANNs boast the highest total citations (17,064) and H-index (63). Most ANIF papers were published by Medical Physics (119 T.P.). Half of the articles in AIFH were published by five journals of Medical Physics, Neurocomputing, International Journal of Heat and Mass Transfer, International Journal of Radiation Oncology Biology Physics, and IEEE Access. The International Journal of Heat and Mass Transfer received the most citations in AIFH. [ABSTRACT FROM AUTHOR]

• A Decision Support System for flash-flood risk management in the arid areas is designed. • An integrated GIS and Watershed Modelling System is applied for flood flow simulation. • AI and Classic Delphi are employed for smart sustainable management of flash-flood risks. • Flood risk is analysed using Failure Mode and Effects Analysis & Hazard & Operability techniques. This paper presents a novel framework for smart integrated risk management in arid regions. The framework combines flash flood modelling, statistical methods, artificial intelligence (AI), geographic evaluations, risk analysis, and decision-making modules to enhance community resilience. Flash flood is simulated by using Watershed Modelling System (WMS). Statistical methods are also used to trim outlier data from physical systems and climatic data. Furthermore, three AI methods, including Support Vector Machine (SVM), Artificial Neural Network (ANN), and Nearest Neighbours Classification (NNC), are used to predict and classify flash flood occurrences. Geographic Information System (GIS) is also utilised to assess potential risks in vulnerable regions, together with Failure Mode and Effects Analysis (FMEA) and Hazard and Operability Study (HAZOP) methods. The decision-making module employs the Classic Delphi technique to classify the appropriate solutions for flood risk control. The methodology is demonstrated by its application to the real case study of the Khosf region in Iran, which suffers from both drought and severe floods simultaneously, exacerbated by recent climate changes. The results show high Coefficient of determination (R2) scores for the three AI methods, with SVM at 0.88, ANN at 0.79, and NNC at 0.89. FMEA results indicate that over 50% of scenarios are at high flood risk, while HAZOP indicates 30% of scenarios with the same risk rate. Additionally, peak flows of over 24 m3/s are considered flood occurrences that can cause financial damage in all scenarios and risk techniques of the case study. Finally, our research findings indicate a practical decision support system that is compatible with sustainable development concepts and can enhance community resilience in arid regions. [ABSTRACT FROM AUTHOR]

• Reviewed more than 200 papers that used machine learning in groundwater quality modeling. • Neural networks are the most used machine learning model in groundwater quality modeling. • Nitrate is the most modeled contaminants. • Suggestions for further works are proposed. Groundwater is a crucial resource across agricultural, civil, and industrial sectors. The prediction of groundwater pollution due to various chemical components is vital for planning, policymaking, and management of groundwater resources. In the last two decades, the application of machine learning (ML) techniques for groundwater quality (GWQ) modeling has grown exponentially. This review assesses all supervised, semi-supervised, unsupervised, and ensemble ML models implemented to predict any groundwater quality parameter, making this the most extensive modern review on this topic. Neural networks are the most used ML model in GWQ modeling. Their usage has declined in recent years, giving rise to more accurate or advanced techniques such as deep learning or unsupervised algorithms. Iran and the United States lead the world in areas modeled, with a wealth of historical data available. Nitrate has been modeled most exhaustively, targeted by nearly half of all studies. Advancements in future work will be made with further implementation of deep learning and explainable artificial intelligence or other cutting-edge techniques, application of these techniques for sparsely studied variables, the modeling of new or unique study areas, and the implementation of ML techniques for groundwater quality management. [ABSTRACT FROM AUTHOR]

The structure of the electricity market in Iran is based on a pay-as-bid auction mechanism. In such a market, hydropower generators need to have accurate estimates of energy price in peak hours of the day-ahead market to optimally operate the reservoir and maximize the revenue. This paper aims at providing a robust model with the best predictors for forecasting the maximum daily electricity price (MDEP) in Iran's electricity market. To reach the goal, hourly electricity prices in 2020 and 2021 were used and several artificial intelligence models were employed to predict the MDEP and ADEP (average daily electricity price). A sensitivity analysis of the inputs showed that in most of the models for forecasting the day-ahead MDEP (P t max), the best predictors were P t − 1 max , P t − 7 max , and P t − 30 m a x . Also, the convolutional neural-long short-term memory network (CNN-LSTM) had the best performance for forecasting both MDEP and ADEP in Iran's energy market. Compared to the multivariate linear regression model, the CNN-LSTM dealt well with sinusoidal characteristics and fluctuation of electricity prices. Therefore, it could improve the accuracy and correlation of the MDEP forecasts by 31% and 3.5%, respectively. • The maximum daily electricity price (MDEP) is very important for hydropower plant's operation. • Nonlinearity, instability, and hourly fluctuations of electricity price make it difficult to forecast. • Several artificial intelligent models, SVM, ANN, ANFIS, ANN-GA, and CNN-LSTM, were used for forecasting MDEP. • The best predictors (P t − 1 max , P t − 7 max , and P t − 30 max) were determined in a sensitivity analysis. • The CNN-LSTM model improved the accuracy of forecasts for MDEP by 31% in comparison with the conventional MLR model. [ABSTRACT FROM AUTHOR]

Globally, most industrial parks that have not gone through a proper site selection process have been neglected or underutilised and such active sites tend to burn a lot of fuels and generate greenhouse gases. Technology must be used to conduct on-site investigations of the spatial and non-spatial factors of industrial parks, especially the emerging Eco-Industrial Park (EIP). The paper seeks to appraise the global application, capability and acceptability of Geographic Information System (GIS) and Multi-Criteria Decision-Making (MCDM) tools in the EIP sites selection and planning. Also, to identify the benefits of mixing MCDM tools when integrating with GIS. There have been some reviews on the use of GIS and MCDM methods, but so far, there is none on the role of GIS and MCDM methods in EIP site selection. Geospatial and MCDM tools can be used to carefully select EIP sites, where materials and energy can be shared to achieve cleaner production, reduce environmental pollution and promote sustainable development. To understand the usage of these tools and their role in the EIP site selection process, a critical review of twenty-two papers published from 2007 to 2020 indexed by Google Scholar, Springer, IEEE Xplore and Web of Science have been developed for the first time. The articles are classified based on country/region, journal of publication, year and citations of publication, the authors and technologies applied. The results show that most studies on EIP site selection come from Iran and China. The Journal of Environmental Monitoring and Assessment has the highest number of articles published at 28%. The highest number of articles was published in 2014 with 312 citations. GIS and Analytic Hierarchy Process (AHP) topped the technologies used, accounting for 42% and 24% respectively, and hybrid tools 34%. As categorised by criteria content, there was an overlap of criteria in most journals. Though, water bodies, utilities, infrastructure, and natural characteristics were prominent. Renewable energy resources were not given attention by most researchers, and risk assessment was absent from the site selection literature. A blend of traditional MCDM tools constitutes the state-of-the-art tools, and when these are combined with GIS in EIP site surveys, they execute to obtain actual results. Future EIP site selection that is integrated with risk assessment is envisioned to be industry 4.0 (4IR)-driven through a combination of GIS and artificial intelligence, and the recently developed MCDM hybrid modules. [Display omitted] • Poor site selection has led to industrial park being neglected or underutilised. • Abandoned/underutilised active industrial parks can burn fuels to cause GHG. • First time of critical review of role of GIS and MCDM tools in EIP site selection. • GIS and AHP currently mainly used for large-scale EIP site criteria selection. • EIP site choice risk study using GIS, AI and MCDM hybrid can sustain industry 4.0 [ABSTRACT FROM AUTHOR]

Water scarcity is increasing worldwide due to population growth and climate variability/change. As a supplementary water resource, Rainwater harvesting (RWH) is a possible solution for dealing with water scarcity, particularly in arid and semi-arid regions with considerable water demand and high variability in precipitation and unexpected extreme events (floods and droughts). The success of RWH systems significantly depends on the location of RWH structures and usually selecting suitable sites is challenging for decision-makers and managers. This paper presents an approach for mapping suitable sites for RWH structures using socio-environmental variables and artificial intelligence algorithms (AIAs). Based on FAO recommendations, the most important conditioning variables for RWH systems are elevation, slope, aspect, precipitation, temperature, distance from the river, curve number (CN), land use, geology, soil type, population density, distance from road, and distance from lakes. An ensemble model was developed based on AIAs, socio-environmental variables, and existing RWH projects, and used for RWH suitability mapping in the large Maharloo-Bakhtegan basin, Iran. Model performance was evaluated using receiver operating characteristic (ROC) and Kappa index. Using the best-performing model, threshold values for conditioning variables were determined from probability curves (PC). The results showed that land use, precipitation, soil type, CN and slope were the most important variables for RHW sites, with the lowest correlation and autocorrelation. The suitability map indicated that 9.7% (3070 km2) of Maharloo-Bakhtegan basin had very high suitability for RWH systems. Thus, in RWH suitability mapping for large area, climate, hydrological, geological, agricultural, topographical, human and socio-economic parameters should be considered to enable efficient RWH planning. Probability curves revealed that the optimum parameter range (α) in Maharloo-Bakhtegan basin was precipitation 357–428 mm, temperature 12.80–15.16 °C, slope 3–6%, elevation 1612–1975 m asl, distance from lake 32–45 km, distance from river 11.4–15.9 km, distance from road 2.59–4.80 km. The RWH suitability map presented can assist decision-makers, hydrologists, and natural resources planners in finding suitable locations for constructing RWH systems. [Display omitted] • Large-scale rainwater harvesting (RWH) was planned using socio-environmental data. • An ensemble model was developed based on AIAs in RWH planning. • Land use and precipitation were major factors in RWH planning. • New probability curves and optimum variable ranges are presented for RWH systems. • RWH suitability mapping confirmed data reliability without expensive field surveys. [ABSTRACT FROM AUTHOR]

Finding an optimal design for hydraulic structures and devices, which work together in irrigation networks, can be formulated as a multi-objective optimization problem. In this paper, a novel framework is proposed for simultaneous optimization of an open channel section and a labyrinth weir geometry. A recently proposed optimizer called Multi-Objective Multi-Verse Optimization (MOMVO) algorithm is employed and its results are compared with Pareto Envelope-based Selection Algorithm II (PESA-II) and Non-dominated Sorting Genetic Algorithm II (NSGA-II) using five metrics, including spacing (SP), Maximum Spread (MS), Non-uniformity of Pareto Front (NPF), Mean Ideal Distance (MID), and Coverage Measure (CM). The first objective function is defined to minimize the construction costs per unit length of the open channel, and the second one is to minimize total concrete volume of the labyrinth weir. The results showed significant differences between MOMVO Pareto optimal solutions and the other two algorithms. The least values of SP, NPF, and MID metrics were provided by MOMVO, which meant its solutions had better conditions regarding uniformity and the closeness to the ideal point. To optimize irrigation network as a system, penalty functions were applied to satisfy hydraulic conditions (flow velocity, Froude number, and nappe interference). Results showed that if the proposed model had been employed in Isfahan Irrigation Networks (IINs) design, it would have reduced the construction costs of open channel and labyrinth weir approximately 11% and 74%, respectively. It can be reported that the most cost-effective design has the least channel wetted perimeter, channel cross-section top width, cross-sectional area, and labyrinth apex length; and the highest channel depth, and weir sidewall angles. • Proposing a novel hybrid framework to optimal design of irrigation networks. • Reducing construction cost of two hydraulic structures using artificial intelligence. • Comparing the performance of three AI algorithms to find the best tradeoff. • Considering a real design problem (Isfahan Irrigation Networks in Iran). • Approximating trade-offs between open channel and labyrinth weir design solutions. [ABSTRACT FROM AUTHOR]