A cluster-based SMOTE both-sampling (CSBBoost) ensemble algorithm for classifying imbalanced data.

In this paper, a Cluster-based Synthetic minority oversampling technique (SMOTE) Both-sampling (CSBBoost) ensemble algorithm is proposed for classifying imbalanced data. In this algorithm, a combination of over-sampling, under-sampling, and different ensemble algorithms, including Extreme Gradient Boosting (XGBoost), random forest, and bagging, is employed in order to achieve a balanced dataset and address the issues including redundancy of data after over-sampling, information loss in under-sampling, and random sample selection for sampling and sample generation. The performance of the proposed algorithm is evaluated and compared to different state-of-the-art competing algorithms based on 20 benchmark imbalanced datasets in terms of the harmonic mean of precision and recall (F1) and area under the receiver operating characteristics curve (AUC) measures. Based on the results, the proposed CSBBoost algorithm performs significantly better than the competing algorithms. In addition, a real-world dataset is used to demonstrate the applicability of the proposed algorithm.

A decision support framework to evaluate the sustainability performance of urban road transportation.

This study proposes a decision support framework (DSF) based on two data envelopment analysis (DEA) models in order to evaluate the urban road transportation of countries for sustainable performance management during different years. The first model considers different years independently while the second model, which is a type of network model, takes into account all the years integrated. A multi-objective programming model under two types of uncertainties is then developed to solve the proposed DEA models based on a revised multi-choice goal programming (GP) approach. The efficiency scores are measured based on the data related to several major European countries and the factors including the level of freight and passenger transportation, level of greenhouse gas emissions, level of energy consumption, and road accidents which are addressed as the main evaluation factors. Eventually, the two proposed models are compared in terms of interpretation and final achievements. The results reveal that the efficiency scores of countries are different under deterministic/uncertain conditions and according to the structure of the evaluation model. Furthermore, efficiency changes are not necessarily the same as productivity changes. The high interpretability (up to 99.6%) of the models demonstrates the reliability of DSF for decision-making stakeholders in the transport sector. Furthermore, a set of managerial analyses is conducted based on different parameters of the performance evaluation measures for these countries including the productivity changes during the period under consideration, resilience of the countries, detection of the benchmark countries, ranking of different countries, and detection of the patterns for improving the transportation system.

An integrated decision support system to achieve sustainable development in transportation routes with traffic flow.

Due to the growing population and demand, transportation planning has received special importance in the context of supply chain management. One of the major challenges in transportation planning is the traffic problem. This challenge affects the safety, environmental, and efficiency factors of transportation systems. Accordingly, in this study, the routes, which are important pillars of transportation planning, are examined from the perspective of sustainability. In this regard, a novel decision support system is developed, wherein at first, some decision-making methods including Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), entropy technique, Nash equilibrium point (NEP), and data envelopment analysis (DEA) are employed to analyze and determine unstable routes. Then, a bi-level leader-follower multi-objective optimization model is developed, based on the vehicle types, to evaluate the routes at different time intervals and identify the most efficient time intervals as a traffic pattern. Finally, the proposed models are implemented in a real case study based on the freeways in Tehran. According to the main finding, it is revealed that heavier and bulkier vehicles have a greater impact on road instability.

Designing an integrated blockchain-enabled supply chain network under uncertainty.

With the development of communication infrastructure, the design of supply chains has changed significantly. Blockchain technology, as one of the most cutting-edge technologies, can promote transparency among members of the supply chain network. To the best of our knowledge, this is the first study that tries to develop a novel bi-objective optimization model to integrate the transparency resulting from the use of blockchain for designing a three-level supply chain network. The first objective function is to minimize total cost while the second objective function seeks to maximize transparency based on the application of blockchain technology. Moreover, it is worth noting that it is the first attempt to investigate the role of a blockchain model under stochastic conditions. The bi-objectiveness and stochastic nature of the proposed model are then treated using Fuzzy Goal Programming (FGP) and Chance-Constrained programming (CCP) approaches, respectively. To tackle the problem, an improved Branch and Efficiency (B&E) algorithm is developed by incorporating transparency along with cost and service. The impacts of blockchain exclusively through transparency (Case 1) or through transparency, cost, and benefits (Case 2) in Supply Chain Design (SCD) are compared. The results demonstrated that the first case has less computational complexity and better scalability, while the second case has more transparency, less congestion, and more security. As one of the main implications, supply chain managers who are focused on cost minimization as well as transparency maximization are advised to take into account the trade-off between featuring costs and benefits of blockchain technology.

Classification of truck-involved crash severity: Dealing with missing, imbalanced, and high dimensional safety data.

While the cost of road traffic fatalities in the U.S. surpasses $240 billion a year, the availability of high-resolution datasets allows meticulous investigation of the contributing factors to crash severity. In this paper, the dataset for Trucks Involved in Fatal Accidents in 2010 (TIFA 2010) is utilized to classify the truck-involved crash severity where there exist different issues including missing values, imbalanced classes, and high dimensionality. First, a decision tree-based algorithm, the Synthetic Minority Oversampling Technique (SMOTE), and the Random Forest (RF) feature importance approach are employed for missing value imputation, minority class oversampling, and dimensionality reduction, respectively. Afterward, a variety of classification algorithms, including RF, K-Nearest Neighbors (KNN), Multi-Layer Perceptron (MLP), Gradient-Boosted Decision Trees (GBDT), and Support Vector Machine (SVM) are developed to reveal the influence of the introduced data preprocessing framework on the output quality of ML classifiers. The results show that the GBDT model outperforms all the other competing algorithms for the non-preprocessed crash data based on the G-mean performance measure, but the RF makes the most accurate prediction for the treated dataset. This finding indicates that after the feature selection is conducted to alleviate the computational cost of the machine learning algorithms, bagging (bootstrap aggregating) of decision trees in RF leads to a better model rather than boosting them via GBDT. Besides, the adopted feature importance approach decreases the overall accuracy by only up to 5% in most of the estimated models. Moreover, the worst class recall value of the RF algorithm without prior oversampling is only 34.4% compared to the corresponding value of 90.3% in the up-sampled model which validates the proposed multi-step preprocessing scheme. This study also identifies the temporal and spatial (roadway) attributes, as well as crash characteristics, and Emergency Medical Service (EMS) as the most critical factors in truck crash severity.

A branch and efficiency algorithm to design a sustainable two-echelon supply chain network considering traffic congestion and uncertainty.

This work aims to design a sustainable two-echelon supply chain not only based on the widely used cost perspective, but also based on the efficient use and preservation of limited resources. For this purpose, a branch and efficiency (B&E) algorithm is developed, which includes an optimization model and an evaluation model. The proposed tri-objective optimization model simultaneously minimizes the total cost of the supply chain, maximizes the sustainability score, and minimizes inequity among customers. The solutions obtained from the optimization model are then evaluated by extended data envelopment analysis (EDEA) models based on common criteria (i.e., cost and service) and traffic congestion criterion. To take into account real-world conditions, parameters related to labor and demand are assumed under uncertainty. Since the presented models consist of more than one objective function, fuzzy goal programming (FGP) method is utilized to tread the multi-objectiveness. The obtained results from tackling a case study problem demonstrate that considering sustainability issues can positively affect both the economic and social aspects of the problem. Furthermore, the developed B&E algorithm is able to reduce costs in each iteration; this is what supply chain managers are interested in. On the other hand, this algorithm can provide more services to applicants compared to one of the competing algorithms.

Microbiology and antimicrobial sensitivity of ventriculo-peritoneal shunt infections in a referral paediatric neurosurgery ward during a period of 7 years.

OBJECTIVES: This study aims to evaluate the aetiology and antibiotic susceptibility pattern in children with ventriculo-peritoneal (VP) shunt infections. METHODS: A retrospective investigation was conducted from March 2011 to March 2018 among 163 children, who were admitted because of VP shunt infections in a referral educational hospital in Isfahan, Iran. RESULTS: Coagulase negative staphylococcus (CoNS) was the most common organism (49.0%), followed by Acinetobacter baumannii (10.4%) and Staphylococcus aureus (8.9%). Susceptibility of Gram-positive organisms to a different class of antibiotics was as follows: glycopeptides (82, 96%), carbapenems (78.38%), aminoglycosides (57.81%), fluoroquinolones (50.00%) and trimethoprim-sulfamethoxazole (50%). Gram-negative isolates were more susceptible to aminoglycosides (56.52%) and fluoroquinolones (52.78%), respectively. Sensitivity of Gram-negative bacteria to fluoroquinolones was significantly higher in infections that occurred in the first month of the shunt implantations compared to infections that were detected afterward (78.95% vs. 23.53%, respectively, P = 0.001). CONCLUSION: The highest sensitivity of Gram-positive bacteria was to glycopeptides and carbapenems. In addition, Gram-negative organisms had the best susceptibility to aminoglycosides and fluoroquinolones. A high level of resistance to studied antibiotics among Gram-negative isolates underscores the necessity of including new antibiotics such as colistin, fosfomycin, ceftazidime/avibactam, ceftolozane/tazobactam and tigecycline in susceptibility testing and empiric antibiotic therapy for VP shunt infections.

A Seven-year Study on the Effects of Intravenous Antibiotic Therapy on Infection of Ventriculoperitoneal Shunts in Children.

BACKGROUND: Infection is one of the most common complications of ventriculoperitoneal (VP) shunts. Its optimal management is controversial. We aimed to report current practice within a large pediatric neurosurgical center. METHODS: We retrospectively studied a cohort of children <15 years with VP shunt infection, who underwent treatment from March 2011 to March 2018 in the main referral children hospital in Isfahan, Iran. Patients with distal shunt infection or associated wound infection were not included. The initial protocol was empiric intravenous antibiotic therapy and repeated evaluation of cerebrospinal fluid (CSF) analysis and culture. After obtaining the culture results, antibiotic regimen was adjusted according to the sensitivity results. Response to protocol was defined as negative culture results after 72 hours of appropriate antibiotic, which persisted at least for 1 month after discontinuation of therapy. In patients who failed to respond to protocol, the shunts were replaced. RESULTS: Totally, 148 cases with mean (SD) age of 21.2 (30.1) months included to the study. Of which, 56.1% were boys and 68.2% had responded to the protocol. This response was significantly prominent in patients who infected with Gram-negative bacteria (82.9%), especially with Acinetobacter spp. (100%) and Pseudomonas aeruginosa (100%). The response was significantly higher in patients with CSF glucose of greater than 40 mg/dl (83% versus 58.1%, respectively, P = 0.004). CONCLUSION: We found that using only intravenous antibiotics is sufficiently enough for treating many children with VP shunt infections, especially in those infected by Gram-negative organisms and CSF glucose level of greater than 40 mg/dl.