Search

Mutual fund (MF) is one of the applicable and popular tools in investment market. The aim of this paper is to propose an approach for performance evaluation of mutual fund by considering internal structure and financial data uncertainty. To reach this goal, the robust network data envelopment analysis (RNDEA) is presented for extended two-stage structure. In the RNDEA method, leader–follower (non-cooperative game) and robust optimization approaches are applied in order to modeling network data envelopment analysis (NDEA) and dealing with uncertainty, respectively. The proposed RNDEA approach is implemented for performance assessment of 15 mutual funds. Illustrative results show that presented method is applicable and effective for performance evaluation and ranking of MFs in the presence of uncertain data. Also, the results reveal that the discriminatory power of robust NDEA approach is more than the discriminatory power of deterministic NDEA models. [ABSTRACT FROM AUTHOR]

Knowledge of protein-protein interactions (PPI) is essential for studying protein functions and understanding the biological processes. Previously, most of the works on PPI in the BioNLP domain rely solely on textual data. With the availability of different information (structure, sequence, gene ontology) about proteins, researchers have started to use other details with textual data to predict PPI more accurately. This paper reports the first attempt in integrating gene ontology(GO)-based information with the features extracted from other two modalities of proteins namely 3D structure and existing textual information. Existing two popular text-based benchmark PPI corpora, i.e., BioInfer and HRPD50 are first extended to integrate with the structure and GO-based information. Finally, some deep learning-based techniques are employed to extract features from three modalities and those are concatenated for final prediction of protein interaction. The experimentation on generated multi-modal datasets illustrates that the proposed deep multi-modal framework outperforms the baselines (uni-modal, bi-modal and multi-modal) and state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Concomitant with people beginning to understand their legal rights or entitlement to complain, complaints of offensive odors and smell pollution have increased significantly. Consequently, monitoring gases and identifying their types and causes in real time has become a critical issue in the modern world. In particular, toxic gases that may be generated at industrial sites or odors in daily life consist of hybrid gases made up of various chemicals. Understanding the types and characteristics of these mixed gases is an important issue in many areas. However, mixed gas classification is challenging because the gas sensor arrays for mixed gases must process complex nonlinear high-dimensional data. In addition, obtaining sufficient training data is expensive. To overcome these challenges, this paper proposes a novel method for mixed gas classification based on analogous image representations with multiple sensor-specific channels and a convolutional neural network (CNN) classifier. The proposed method maps a gas sensor array into a multichannel image with data augmentation, and then utilizes a CNN for feature extraction from such images. The proposed method was validated using public mixture gas data from the UCI machine learning repository and real laboratory experiments. The experimental results indicate that it outperforms the existing classification approaches in terms of the balanced accuracy and weighted F1 scores. Additionally, we evaluated the performance of the proposed method in various experimental settings in terms of data representation, data augmentation, and parameter initialization, so that practitioners can easily apply it to artificial olfactory systems. [ABSTRACT FROM AUTHOR]

In this paper, by using a lifecycle perspective, four stages related to the extraction, refining and processing of copper were identified. The different behaviors of countries in the import/export networks at the four stages synthetically reflect their position in the global network of copper production and consumption. The trade flows of four commodities related to the extraction, refining and processing of copper of 142 nations with population above 2 millions based on the UN Comtrade website (https://comtrade.un.org/data/), in five years from 2017 to 2021, were considered. The observed trade flows in each year have been modelled as a directed multilayer network. Then the countries have been grouped according to their structural equivalence in the international copper flow by using a Multilayer Stochastic Block Model. To put further insight in the obtained community structure of the countries, a deep learning model based on adapting the node2vec to a multilayer setting has been used to embed the countries in an Euclidean plane. To identify groups of nations that play the same role across time, some distances between the parameters obtained in consecutive years were introduced. We observe that 97 countries out of 142 consistently occupy the same position in the copper supply chain throughout the five years, while the other 45 move through different roles in the copper supply chain. [ABSTRACT FROM AUTHOR]

Rating prediction is a crucial element of business analytics as it enables decision-makers to assess service performance based on expressive customer feedback. Enhancing rating score predictions and demand forecasting through incorporating performance features from verbatim text fields, particularly in service quality measurement and customer satisfaction modelling is a key objective in various areas of analytics. A range of methods has been identified in the literature for improving the predictability of customer feedback, including simple bag-of-words-based approaches and advanced supervised machine learning models, which are designed to work with response variables such as Likert-based rating scores. This paper presents a dynamic model that incorporates values from topic membership, an outcome variable from Latent Dirichlet Allocation, with sentiment analysis in an Extreme Gradient Boosting (XGBoost) model used for rating prediction. The results show that, by incorporating features from simple unsupervised machine learning approaches (LDA-based), an 86% prediction accuracy (AUC based) can be achieved on objective rating values. At the same time, a combination of polarity and single-topic membership can yield an even higher accuracy when compared with sentiment text detection tasks both at the document and sentence levels. This study carries significant practical implications since sentiment analysis tasks often require dictionary coverage and domain-specific adjustments depending on the task at hand. To further investigate this result, we used Shapley Additive Values to determine the additive predictability of topic membership values in combination with sentiment-based methods using a dataset of customer reviews from food delivery services. [ABSTRACT FROM AUTHOR]

Deep neural network (DNN) architectures such as recurrent neural networks and transformers display outstanding performance in modeling sequential unstructured data. However, little is known about their merit to model customer churn with time-varying data. The paper provides a comprehensive evaluation of the ability of recurrent neural networks and transformers for customer churn prediction (CCP) using time-varying behavioral features in the form of recency, frequency, and monetary value (RFM). RFM variables are the backbone of CCP and, more generally, customer behavior forecasting. We examine alternative strategies for integrating time-varying and non-variant customer features in one network architecture. In this scope, we also assess hybrid approaches that incorporate the outputs of DNNs in conventional CCP models. Using a comprehensive panel data set from a large financial services company, we find recurrent neural networks to outperform transformer architectures when focusing on time-varying RFM features. This finding is confirmed when time-invariant customer features are included, independent of the specific form of feature integration. Finally, we find no statistical evidence that hybrid approaches (based on regularized logistic regression and extreme gradient boosting) improve predictive performance—highlighting that DNNs and especially recurrent neural networks are suitable standalone classifiers for CCP using time-varying RFM measures. [ABSTRACT FROM AUTHOR]

Faced with dynamic and increasingly diversified public transport requirements, bus operators are urged to propose operational innovations to sustain their competitiveness. In particular, ordinary bus operations are heavily constrained by well-established route options, and it is challenging to accommodate dynamic passenger flows effectively and with a good level of resource utilization performance. Inspired by the philosophy of sharing economy, many of the available transport resources on the road, such as minibuses and private vehicles, can offer opportunities for improvement if they can be effectively incorporated and exploited. In this regard, this paper proposes a metric learning-based prediction algorithm which can effectively capture the demand pattern and designs a route planning optimizer to help bus operators effectively deploy fixed routing and cooperative buses with traffic dynamics. Through extensive numerical studies, the performance of our proposed metric learning-based Generative Adversarial Network (GAN) prediction model outperforms existing ways. The effectiveness and robustness of the prediction-supported routing planner are well demonstrated for a real-time case. Further, managerial insights with regard to travel time, bus fleet size, and customer service levels are revealed by various sensitivity analysis. [ABSTRACT FROM AUTHOR]

In some production processes, the effort associated with a certain activity for its execution can vary over time. In this case, the amount of work per time unit devoted to each activity, so as its duration, is not univocally determined. This kind of problem can be represented by an activity project network with the so-called feeding precedence relations, and activity variable execution intensity. In this paper, we propose a forward recursion algorithm able to find the earliest start and finish times of each activity, in O (m log n) time, with n and m being the number of activities and the number of precedence relations, respectively. In particular, this requires the calculation of the (optimal) execution intensity profile, for each activity, that warrants the earliest start schedule and the minimum completion time of the project. [ABSTRACT FROM AUTHOR]

Transportation is a broad multidisciplinary science concerned with the study of transportation network design, vehicle application engineering, as well as transportation operations and management. Due to the breadth of the topic and the increasing complexity of the decision processes involved in managing transportation operations, transportation has become a fertile field of study for operations research. Inspired by this, we have compiled this special issue of the Annals of Operations Research comprising of 24 new research contributions and two systematic literature reviews supporting and categorizing the current state of transportation research. [ABSTRACT FROM AUTHOR]

Police patrol is an effective crime prevention tool and boosts public confidence in urban security. Many interesting decision making problems appear in route design, resource allocation and jurisdiction planning. Many cities across the world have adopted a structured and intelligent method of police patrol due to the presence of a variety of operational and resource constraints. In this paper, we present a comprehensive review of the state-of-the-art in this domain, especially from the practice of operations research (OR) point of view. This is the first-of-its-kind review on police patrol presenting a classification scheme based on the type of problem, objective and modelling approach. In this novel scheme, one can track any paper almost readily to find the specific contribution. The applicability of OR in this domain is set to grow significantly as the governments formulate policies related to smart city planning and urban security. This study reveals many practical challenges in police patrolling for future research. [ABSTRACT FROM AUTHOR]

Supply chain management aims to designing, managing and coordinating material/product, information and financial flows to fulfill the customer requirements at low costs and thereby increasing supply chain profitability. In the last decades, data envelopment analysis has become the main topic of interest as a mathematical tool to evaluate supply chain management. While, various data envelopment analysis models have been suggested to measure and evaluate the supply chain management, there is a lack of research regarding to systematic literature review and classification of study in this field. Regarding this, some major databases including Web of Science and Scopus have been nominated and systematic and meta-analysis method which called “PRISMA” has been proposed. Accordingly, a review of 75 published articles appearing in 35 scholarly international journals and conferences between 1996 and 2016 have been attained to reach a comprehensive review of data envelopment analysis models in evaluation supply chain management. Consequently, the selected published articles have been categorized by author name, the year of publication, technique, application area, country, scope, data envelopment analysis purpose, study purpose, research gap and contribution, results and outcome, and journals and conferences in which they appeared. The results of this study indicated that areas of supplier selection, supply chain efficiency and sustainable supply chain have had the highest frequently than other areas. In addition, results of this review paper indicated that data envelopment analysis showed great promise to be a good evaluative tool for future evaluation on supply chain management, where the production function between the inputs and outputs was virtually absent or extremely difficult to acquire. The facility of multiple inputs and multiple outputs of the data envelopment analysis model was definitely an attractive one to most researchers and, therefore, the data envelopment analysis procedure had found many applications beyond supply chain management into organization and industry. [ABSTRACT FROM AUTHOR]

I Seddigh et al. i in their work I "Approaching towards sustainable supply chain under the spotlight of business intelligence" i develop a theoretical model to explain how business intelligence capabilities influence the company's supply chain sustainability. They apply SmartPLS 3.0, which uses partial least squares structural equation modelling (PLS-SEM), to test the correlation between supply chain risks and supply chain performance, investigate the mediating effect of supply chain risk management on the correlation between supply chain risks and supply chain performance. The COVID-19 pandemic has created a lot of disturbances in supply chains around the globe leading to disturbances in the availability of logistics infrastructure (e.g., ports closure, boarders' control, etc.) and workforce absence. They propose a novel robust-heuristic optimization approach which support the supply chain managers in decision-making in selecting of carbon emission policies in supply chains with different vehicle types, demand and economic uncertainty in large scale-problems. [Extracted from the article]

Disaster management (DM), within a broad field of operations management, is becoming an emergent area of interest for academics and practitioners. This study examines the literature evolving on disaster management linked to application of simulation modelling. Many literature review studies on disaster management are offered by numerous authors. However, broad analysis of various applications of simulation based modelling within a DM context are not adequate. As a result, the proposed research carries out a review of 100 papers published in numerous peer-reviewed academic journals during the period 2000 and January 2016. In particular, the paper presents a classification of publications based on the simulation technique(s) used; examines the disaster issues addressed by the technique (s) used; analyzes trends and identifies the impact of published research on simulation modeling pertaining to DM. This study would enable researchers to understand the significance of various simulation modelling techniques in resolving a variety of disaster related challenges. [ABSTRACT FROM AUTHOR]

The field of supply chain management has been extensively studied, while its role in disaster relief operations has received limited contributions. Recent dedicated special issues on Humanitarian Operations and Supply Chain Management (HOSCM) and a dedicated Journal focusing on humanitarian logistics and supply chain management clearly indicate the growing popularity of the HOSCM literature. The purpose of our current study is to undertake an extensive review of extant literature published in operations and supply chain management journals as well as popular interdisciplinary journals. A review of 362 papers published between 2011 and 2017 provides a thematic outline of the study. The study pivots around nine key themes, which have gained prominent attention from HOSCM scholars, and draws a roadmap from reviewing earlier review papers to performance evaluation of HOSCM related studies. Some of the key themes include humanitarian logistics, theory focused research, case studies, mathematical models, humanitarian supply chain properties and resources needed for efficient and effective management of humanitarian operations. Finally, our study offers several research directions which may take the existing debates to a next level. [ABSTRACT FROM AUTHOR]