Towards facing uncertainties in biofuel supply chain networks: a systematic literature review

Biofuel supply chains (BSCs) face diverse uncertainties that pose serious challenges. This has led to an expanding body of research focused on studying these challenges. Hence, there is a growing need for a comprehensive review that summarizes the current studies, identifies their limitations, and provides essential advancements to support scholars in the field. To overcome these limitations, this research aims to provide insights into managing uncertainties in BSCs. The review utilizes the Systematic Reviews and Meta-Analyses (PRISMA) method, identifying 205 papers for analysis. This study encompasses three key tasks: first, it analyses the general information of the shortlisted papers. Second, it discusses existing methodologies and their limitations in addressing uncertainties. Lastly, it identifies critical research gaps and potential future directions. One notable gap involves the underutilization of machine learning techniques, which show potential for risk identification, resilient planning, demand prediction, and parameter estimations in BSCs but have received limited attention. Another area for investigation is the potential of agent-based simulation, which can contribute to analysing resilient policies, evaluating resilience, predicting parameters, and assessing the impact of emerging technologies on BSC resilience in the twenty-first century. Additionally, the study identifies the omission of various realistic assumptions, such as backward flow, lateral transshipments, and ripple effects in BSC. This study highlights the complexity of managing uncertainties in BSCs and emphasizes the need for further research and attention. It contributes to policymakers' understanding of uncertain sources and suitable approaches while inspiring researchers to address limitations and generate breakthrough ideas in managing BSC uncertainties.

An extended DNMA-based multi-criteria decision-making method and its application in the assessment of sustainable location for a lithium-ion batteries' manufacturing plant

Lithium-ion battery (LiB), a leading residual energy resource for electric vehicles (EVs), involves a market presenting exponential growth with increasing global impetus towards electric mobility. To promote the sustainability perspective of the EVs industry, this paper introduces a hybridized decision support system to select the suitable location for a LiB manufacturing plant. In this study, single-valued neutrosophic sets (SVNSs) are considered to diminish the vagueness in decision-making opinions and evade flawed plant location assessments. This study divided into four phases. First, to combine the single-valued neutrosophic information, some Archimedean-Dombi operators are developed with their outstanding characteristics. Second, an innovative utilization of the Method based on the Removal Effects of Criteria (MEREC) and Stepwise Weight Assessment Ratio Analysis (SWARA) is discussed to obtain objective, subjective and integrated weights of criteria assessment with the least subjectivity and biasedness. Third, the Double Normalization-based Multi-Aggregation (DNMA) method is developed to prioritize the location options. Fourth, an illustrative study offers decision-making strategies for choosing a suitable location for a LiB manufacturing plant in a real-world setting. Our outcomes specify that Bangalore (L ₂), with an overall utility degree (0.7579), is the best plant location for LiB manufacturing. The consistency and robustness of the presented methodology are discussed with the comparative study and sensitivity investigation. This is the first study in the current literature that has proposed an integrated methodology on SVNSs to select the best LiB manufacturing plant location by estimating both the objective and subjective weights of criteria and by considering ambiguous, inconsistent, and inexact manufacturing-based information.

Advanced deep learning approaches to predict supply chain risks under COVID-19 restrictions

The ongoing COVID-19 pandemic has created an unprecedented predicament for global supply chains (SCs). Shipments of essential and life-saving products, ranging from pharmaceuticals, agriculture, and healthcare, to manufacturing, have been significantly impacted or delayed, making the global SCs vulnerable. A better understanding of the shipment risks can substantially reduce that nervousness. Thenceforth, this paper proposes a few Deep Learning (DL) approaches to mitigate shipment risks by predicting "if a shipment can be exported from one source to another", despite the restrictions imposed by the COVID-19 pandemic. The proposed DL methodologies have four main stages: data capturing, de-noising or pre-processing, feature extraction, and classification. The feature extraction stage depends on two main variants of DL models. The first variant involves three recurrent neural networks (RNN) structures (i.e., long short-term memory (LSTM), Bidirectional long short-term memory (BiLSTM), and gated recurrent unit (GRU)), and the second variant is the temporal convolutional network (TCN). In terms of the classification stage, six different classifiers are applied to test the entire methodology. These classifiers are SoftMax, random trees (RT), random forest (RF), k-nearest neighbor (KNN), artificial neural network (ANN), and support vector machine (SVM). The performance of the proposed DL models is evaluated based on an online dataset (taken as a case study). The numerical results show that one of the proposed models (i.e., TCN) is about 100% accurate in predicting the risk of shipment to a particular destination under COVID-19 restrictions. Unarguably, the aftermath of this work will help the decision-makers to predict supply chain risks proactively to increase the resiliency of the SCs.

Intelligent model for contemporary supply chain barriers in manufacturing sectors under the impact of the COVID-19 pandemic

The COVID-19 pandemic has cast a shadow on the global economy. Since the beginning of 2020, the pandemic has contributed significantly to the global recession. In addition to the health damages of the pandemic, the economic impacts are also severe. The consequences of such effects have pushed global supply chains toward their breaking point. Industries have faced multiple obstacles, threatening the fragile flow of raw materials, spare parts, and consumer goods. Previous studies showed that supply chain barriers have multi-faceted impacts on industries and supply chains, which demand appropriate measures. In this regard, seven major barriers that directly impact industries have been identified to determine which industry is most affected by the COVID-19 pandemic. This paper utilized a hybrid multi-criteria decision-making (MCDM) approach under a neutrosophic environment using trapezoidal neutrosophic numbers to rank those barriers. The Analytical Network Process (ANP) quantifies the effects and considers the interrelationships between the determined barriers (criteria) involved in decision-making. Subsequently, the Measurement Alternatives and Ranking according to the COmpromise Solution (MARCOS) method was adopted to rank six industries according to the impact of those barriers. Results show that the lack of inventory is the largest barrier to influencing industries, followed by the lack of manpower. Sensitivity analysis is performed to detect the change in the rank of industries according to the change in the relative importance of the barriers.

An optimum balance among the reduction in ordering cost, product deterioration and carbon emissions: a sustainable green warehouse

This study involves an optimum balance between ordering cost reduction and product deterioration in controllable carbon emissions for a sustainable green warehouse. The sensitivity analysis is to simulate the impact of those attributes. Industries are foraging to find a proper balance between the use of fossil fuels and reducing carbon emissions, as burning fossil fuels is also indispensable for industrialization. Carbon can emit through inevitable logistic activities in the chains (e.g., lighting, heating, air-conditioning, product deterioration). An industry always attempts to curb those emissions through energy-efficient green technology. The green warehouse is a popular store system in present supply chains to limit the carbons. Product deterioration, particularly for perishable items, is also important for a practitioner to decide how to preserve a perishable product for maximum shelf-life. There is a common tendency among industries to increase order frequencies and volumes in search of a better preservation strategy, increasing the ordering cost and the probability of carbon emissions due to increased transportation. A realistic mathematical model is proposed based on those decision parameters by a sensitivity analysis to demonstrate the impacts. The results showed an increase of 46.30% profit is achieved when all three proposed reduction attributes, but shortages are considered. This improvement is significant without shortage, whereas the increased profit is 94.75%.

RESCOVIDTCNnet: A residual neural network-based framework for COVID-19 detection using TCN and EWT with chest X-ray images

Since the advent of COVID-19, the number of deaths has increased exponentially, boosting the requirement for various research studies that may correctly diagnose the illness at an early stage. Using chest X-rays, this study presents deep learning-based algorithms for classifying patients with COVID illness, healthy controls, and pneumonia classes. Data gathering, pre-processing, feature extraction, and classification are the four primary aspects of the approach. The pictures of chest X-rays utilized in this investigation came from various publicly available databases. The pictures were filtered to increase image quality in the pre-processing stage, and the chest X-ray images were de-noised using the empirical wavelet transform (EWT). Following that, four deep learning models were used to extract features. The first two models, Inception-V3 and Resnet-50, are based on transfer learning models. The Resnet-50 is combined with a temporal convolutional neural network (TCN) to create the third model. The fourth model is our suggested RESCOVIDTCNNet model, which integrates EWT, Resnet-50, and TCN. Finally, an artificial neural network (ANN) and a support vector machine were used to classify the data (SVM). Using five-fold cross-validation for 3-class classification, our suggested RESCOVIDTCNNet achieved a 99.5 percent accuracy. Our prototype can be utilized in developing nations where radiologists are in low supply to acquire a diagnosis quickly.

Inventory decisions on the transportation system and carbon emissions under COVID-19 effects: A sensitivity analysis

The COVID-19 pandemic has created multiple problems in the existing transportation system. The contribution of this study is to guide logistics managers as they make ordering decisions within a disrupted transportation system. In the overall supply chain system, inventory decisions have been either compromised or challenged. Traditional inventory decisions that consider preplanned transportation facilities (and speeds) are currently becoming obsolete, predominantly in post-COVID times due to delays in the delivery of products and higher delivery costs. Therefore, businesses such as retailers must align ordering and pricing decisions to maintain a sustainable profit. To address this issue, this study investigates optimum inventory decisions under the pandemic's effects while considering the transportation cost as proportional to COVID-19 intensity. This study also considers product deterioration, time-dependent holding costs, price-dependent demands, and carbon emissions from vehicle operation and intends to establish a harmonious relationship among these attributes. The optimization of green technology investment is studied to reduce emissions due to transportation. Some theoretical derivations and numerical examples are given, and they are followed by a sensitivity analysis to extract important managerial insights into the effect of COVID-19. The manager can set an optimal selling price and the cycle length by carefully planning the number of trips in considering the rate of the outbreak and its effect on the increasing transportation cost.

Performance Improvement of a Parsimonious Learning Machine Using Metaheuristic Approaches

Autonomous learning algorithms operate in an online fashion in dealing with data stream mining, where minimum computational complexity is a desirable feature. For such applications, parsimonious learning machines (PALMs) are suitable candidates due to their structural simplicity. However, these parsimonious algorithms depend upon predefined thresholds to adjust their structures in terms of adding or deleting rules. Besides, another adjustable parameter of PALM is the fuzziness in membership grades. The best set of such hyper parameters is determined by experts' knowledge or by optimization techniques such as greedy algorithms. To mitigate such experts' dependency or usage of computationally expensive greedy algorithms, in this work, a meta heuristic-based optimization technique, called the multimethod-based optimization technique (MOT), is utilized to develop an advanced PALM. The performance has been compared with some popular optimization techniques, namely, the greedy search, local search, genetic algorithm (GA), and particle swarm optimization (PSO). The proposed parsimonious learning algorithm with MOT outperforms the others in most cases. It validates the multioperator-based optimization technique's advantages over the single operator-based variants in selecting the best feasible hyperparameters for the autonomous learning algorithm by maintaining a compact architecture.

An MPPT method using hybrid radial movement optimization with teaching-learning based optimization under fluctuating atmospheric conditions

Due to its clean and abundant availability, solar energy is popular as a source to generate electricity. Solar photovoltaic (PV) technology converts sunlight incident on the solar PV panel or array directly into non-linear DC electricity. However, the non-linear nature of the solar panels’ power needs to be tracked for its efficient utilization. The problem of non-linearity becomes more prominent when the solar PV array is shaded, even leading to high power losses and concentrated heating in some areas (hotspot condition) of the PV array. Bypass diodes used to eliminate the shading effect cause multiple peaks of power on the power versus voltage (P-V) curve and make the tracking problem quite complex. Conventional algorithms to track the optimal power point cannot search the complete P-V curve and often become trapped in local optima. More recently, metaheuristic algorithms have been employed for maximum power point tracking. Being stochastic, these algorithms explore the complete search area, thereby eliminating any chance of becoming trapped stuck in local optima. This paper proposes a hybridized version of two metaheuristic algorithms, Radial Movement Optimization and teaching-learning based optimization (RMOTLBO). The algorithm has been discussed in detail and applied to multiple shading patterns in a solar PV generation system. It successfully tracks the maximum power point (MPP) in a lesser amount of time and lesser fluctuations.

Reactive power compensation for grid by Packed-U-Cell inverter using model predictive control strategy with intelligent multi-objective scheme

This paper proposes a model predictive control strategy for 15 level Packed-U-Cell inverter that satisfies multiple-objectives of low current total harmonic distortion (THD), capacitor voltage balances, supply of desired active and reactive power, as well as lower switching and lower voltage stresses on the switching devices. The proposed device performs well under dynamic conditions and can successfully track the current command during step changes in the power demand. A detailed modeling is presented and discussed. MATLAB/Simulink is used for obtaining the simulation results, and the results are validated in the real time by using a hardware-in-the-loop (HIL) Typhoon 402 real-time emulator.

An accurate artificial intelligence system for the detection of pulmonary and extra pulmonary Tuberculosis

Tuberculosis (TB) is the greatest irresistible illness in humans, caused by microbes Mycobacterium TB (MTB) bacteria and is an infectious disease that spreads from one individual to another through the air. It principally influences lung, which is termed Pulmonary TB (PTB). However, it can likewise influence other parts of the body such as the brain, bones and lymph nodes. Hence, it is also referred to as Extra Pulmonary TB (EPTB). TB has normal symptoms, so without proper testing, it is hard to detect if a patient has TB or not. In this paper, an accurate and novel system for diagnosing TB (PTB and EPTB) has been designed using image processing and AI-based classification techniques. The designed system is comprised of two phases. Firstly, the X-Ray image is processed using preprocessing, segmentation and features extraction and then, three different AI-based techniques are applied for classification. For image processing, 'Histogram Filter' and 'Median Filter' are applied with the CLAHE process to retrieve the segmented image. Then, classification based on AI techniques is done. The designed system produces the accuracy of 98%, 83%, and 89% for Decision Tree, SVM, and Naïve Bayes Classifier, respectively and has been validated by the doctors of the Jalandhar, India.

FSS-2019-nCov: A deep learning architecture for semi-supervised few-shot segmentation of COVID-19 infection

The newly discovered coronavirus (COVID-19) pneumonia is providing major challenges to research in terms of diagnosis and disease quantification. Deep-learning (DL) techniques allow extremely precise image segmentation; yet, they necessitate huge volumes of manually labeled data to be trained in a supervised manner. Few-Shot Learning (FSL) paradigms tackle this issue by learning a novel category from a small number of annotated instances. We present an innovative semi-supervised few-shot segmentation (FSS) approach for efficient segmentation of 2019-nCov infection (FSS-2019-nCov) from only a few amounts of annotated lung CT scans. The key challenge of this study is to provide accurate segmentation of COVID-19 infection from a limited number of annotated instances. For that purpose, we propose a novel dual-path deep-learning architecture for FSS. Every path contains encoder-decoder (E-D) architecture to extract high-level information while maintaining the channel information of COVID-19 CT slices. The E-D architecture primarily consists of three main modules: a feature encoder module, a context enrichment (CE) module, and a feature decoder module. We utilize the pre-trained ResNet34 as an encoder backbone for feature extraction. The CE module is designated by a newly introduced proposed Smoothed Atrous Convolution (SAC) block and Multi-scale Pyramid Pooling (MPP) block. The conditioner path takes the pairs of CT images and their labels as input and produces a relevant knowledge representation that is transferred to the segmentation path to be used to segment the new images. To enable effective collaboration between both paths, we propose an adaptive recombination and recalibration (RR) module that permits intensive knowledge exchange between paths with a trivial increase in computational complexity. The model is extended to multi-class labeling for various types of lung infections. This contribution overcomes the limitation of the lack of large numbers of COVID-19 CT scans. It also provides a general framework for lung disease diagnosis in limited data situations.

DeepH-DTA: Deep Learning for Predicting Drug-Target Interactions: A Case Study of COVID-19 Drug Repurposing

The rapid spread of novel coronavirus pneumonia (COVID-19) has led to a dramatically increased mortality rate worldwide. Despite many efforts, the rapid development of an effective vaccine for this novel virus will take considerable time and relies on the identification of drug-target (DT) interactions utilizing commercially available medication to identify potential inhibitors. Motivated by this, we propose a new framework, called DeepH-DTA, for predicting DT binding affinities for heterogeneous drugs. We propose a heterogeneous graph attention (HGAT) model to learn topological information of compound molecules and bidirectional ConvLSTM layers for modeling spatio-sequential information in simplified molecular-input line-entry system (SMILES) sequences of drug data. For protein sequences, we propose a squeezed-excited dense convolutional network for learning hidden representations within amino acid sequences; while utilizing advanced embedding techniques for encoding both kinds of input sequences. The performance of DeepH-DTA is evaluated through extensive experiments against cutting-edge approaches utilising two public datasets (Davis, and KIBA) which comprise eclectic samples of the kinase protein family and the pertinent inhibitors. DeepH-DTA attains the highest Concordance Index (CI) of 0.924 and 0.927 and also achieved a mean square error (MSE) of 0.195 and 0.111 on the Davis and KIBA datasets respectively. Moreover, a study using FDA-approved drugs from the Drug Bank database is performed using DeepH-DTA to predict the affinity scores of drugs against SARS-CoV-2 amino acid sequences, and the results show that that the model can predict some of the SARS-Cov-2 inhibitors that have been recently approved in many clinical studies.