Identifying and evaluating COVID-19 effects on short-term statistics

The economic downturn due to lockdown measures at the beginning of the COVID-19 crisis raised the question whether any adaptations to the short-term statistics (STS) were needed to ensure accurate and relevant output. We limit ourselves to STS on turnover and related variables like volume of production. We looked into the different stages of the production process - from data collection to output - and anticipated a number of potential lockdown effects. With respect to output relevance, there was an increased interest in faster and specific output. With respect to the output accuracy, we took measures to check whether the anticipated effects really occurred and measures to mitigate the consequences. Examples of such measures are the calculation of an additional editing score function, alternative imputations and extensions of the regular analysis step. In this paper we give an overview of the anticipated effects, the subsequent measures that we took, we evaluate to what extent the anticipated effects occurred in practice and we mention some unforeseen effects. We end this paper by discussing to what extent the developed measures are also useful to keep after the economy has recovered. © 2023 - IOS Press. All rights reserved.

Global Commodity Chains and Supply Bottlenecks in Austria with Medical Devices in the COVID-19 Pandemic

The COVID-19 Pandemic led to a strong increase in demand for medical products. At the same time, supply problems in international supply chains kicked in due to health policy interventions (e.g., lockdowns) and economic policy measures (e.g., export controls). Combined, both resulted in temporary shortages and triggered a controversial discussion about the advantages and disadvantages of globalised production structures, which led to strong dependencies on a few, primarily Asian, locations and producers. Against this background and based on case studies for Austria, the article deals with the question which factors determine the robustness of global commodity chains for respirators, protective gloves and respiratory equipment and which national and European policies could be suitable for increasing resilience in the supply of medical products. © 2022 Austrian Geographical Society. All rights reserved.

Towards Lifting the Trade-off between Accuracy and Adversarial Robustness of Deep Neural Networks with Application on COVID 19 CT Image Classification and Medical Image Segmentation

Deep neural networks (DNNs) are vulnerable to adversarial noises. Adversarial training is a general strategy to improve DNN robustness. But training a DNN model with adversarial noises may result in a much lower accuracy on clean data, which is termed the trade-off between accuracy and adversarial robustness. Towards lifting this trade-off, we propose an adversarial training method that generates optimal adversarial training samples. We evaluate our methods on PathMNIST and COVID-19 CT image classification tasks, where the DNN model is ResNet-18, and Heart MRI and Prostate MRI image segmentation tasks, where the DNN model is nnUnet. All these four datasets are publicly available. The experiment results show that our method has the best robustness against adversarial noises and has the least accuracy degradation compared to the other defense methods. © 2023 SPIE.

Longitudinal robustness of a thoracic radiographic AI model for COVID-19 severity prediction

The 'ging' of artificial intelligence/machine learning (AI/ML) models after initial development and evaluation is known to frequently occur and can pose substantial problems. When there are changes in population, disease characteristics, imaging equipment, or protocols, model performance may start to deteriorate, and the performance predicted in a research setting may no longer hold after deployment (either in a clinical setting or in further research). This data shift phenomenon is a common problem in AI/ML. We trained and evaluated a previously in-house developed AI/ML model for COVID severity prediction using two COVID-19-positive consecutive adult patient cohorts from a single institution. The first cohort was from the time that the Delta strain was dominant accounting for <95% of cases (June 24-December 11, 2021, 820 patients, 1331 chest radiographs (CXRs)) and the second cohort was from the time that the Omicron variant was dominant (Jan 1-21, 2022, 656 patients, 970 CXRs). Inclusion criteria were COVID-positivity and the availability of CXR imaging exams, in general for patients not admitted to ICU and prior to ICU admission for those patients admitted to ICU as part of their treatment. Exclusion criteria were image acquisition in ICU or the presence of mechanical ventilation. Our image-based AI/ML model was trained to predict, based on each frontal CXR from a COVID-positive patient, whether this patient would be admitted to ICU within a 24, 48, 72, or 96-hour window. The model was evaluated 1) in a cross-sectional test when trained on a subset/tested on an independent subset of the Delta cohort, 2) similarly for the Omicron cohort, and 3) in a longitudinal test when trained on the Delta cohort/tested on the Omicron cohort. Cohorts were similar in ICU admission rate and fraction of portable CXRs, while immunization rate was higher for the Omicron cohort. The model did not demonstrate signs of aging with performances in the longitudinal test being very similar to those within the Delta cohort, e.g., an area under the ROC curve in the task of predicting ICU admission within 24 hours of 0.76 [0.68;0.84] when trained/tested within the Delta cohort and 0.77 [0.73;0.80] for the longitudinal test (p>0.05). The performance within the Omicron cohort was similar as well, at 0.76 [0.66;0.84]. Our AI/ML model for COVID-severity prediction did not demonstrate signs of aging in a longitudinal test when trained on the Delta cohort and applied as-is to the Omicron cohort. © COPYRIGHT SPIE. Downloading of the is permitted for personal use only.

Analyzing the effects of COVID-19 pandemic on supplier performance evaluation factors via fuzzy cognitive map with a hybrid learning system

PurposeThe COVID-19 pandemic has proven that how supply chain management (SCM) can become a crucial process for sustainability of the world's production/service. The global supply chain crisis during pandemic has affected most of the sectors. Home and personal care products manufacturers are among them. In this study (1) the problems at SCM of personal and home care products manufacturers during pandemic are discussed with the help of medium-size manufacturer and (2) the factors affecting suppliers' performance for the relevant sector during COVID-19 are analyzed comprehensively.Design/methodology/approachThe importance of the factors is evaluated using fuzzy cognitive maps that can help to reveal hidden casual relationships with the help of expert knowledge. In order to eliminate subjectivity due to usage of expert knowledge, the maps are trained with a hybrid learning approach that consists of Non-linear Learning and Extended Great Deluge Algorithms to increase robustness of the analysis.FindingsThe findings of the study indicate that the factors such as general quality level of products/services, compliance to delivery time, communication skills and total production capacity of suppliers have been crucial factors during pandemic.Originality/valueWhile the implementation of the hybrid learning approach on supply chain can fill the gap in the relevant literature, the promising results of the study can prove the convenience of the methodology to model the of complex systems like supply chain processes.

A NEW FLOW-BASED MODEL OF END-TO-END INTEGRATION IN PREMIUM PRODUCT SUPPLY CHAINS

Gedryf deur die totale koste van eienaarskap, handel en tegnologiemededinging tussen die Verenigde State van Amerika en China, en die COVID-19-pandemie, ondergaan wereldwye voorsieningskettings 'n groot herstrukturering wat binnekort die besigheid en ekonomie oor die hele wereld sal transformeer. Onlangs het voorsieningskettings met end-totend-integrasie vir premium landbouvoedselprodukte as 'n nuwe sakemodel na vore gekom. Hierdie artikel ondersoek hoe hulle moet funksioneer, en identifiseer die voorsieningskettingstruktuur / - produksie / - besigheids toestande wat nodig is vir hul ontwikkeling. Ons bestudeer 'n premium voorsieningsketting wat bestaan uit baie klein plase wat piesangs van topgehalte produseer, een integreerfirma en duisende kleinhandelwinkels. Ons gebruik industrie- en besigheidsdata om 'n meervoudige roete-vloei-gebaseerde model te kalibreer van plase tot integreerder tot kleinhandelaars/markte. Ons gebruik dan sensitiwiteitsanalise om die belanghebbendes se besluitgedrag te analiseer, en identifiseer en bespreek drie hoofbesluitkwessies: kontrakboerdery, kapasiteitstrategie en besigheidsrobuustheid. Vir kontrakspesifikasie is kontraktering op prys, eerder as hoeveelheid, bevorderlik om die belange van die belanghebbendes te koördineer. Vir die kapasiteitstrategie moet die integreerder rou produkte van baie klein plase verkry eerder as minder groot plase. Vir besigheid se robuustheid kan die integreerder steeds robuuste winste verseker deur sy produkaanbod te reguleer wanneer nuwe mededingers ontstaan of vraag verander. Hierdie resultate word onder verskeie scenario's getoets om die impak van insetparameters of voorsieningskettingstruktuur te bepaal, en word geverifieer met 'n bedryfspraktisyn wat ondervinding het met veelvuldige premium agri-voedselprodukte. Die resultate, tesame met die vloeimodel en sy berekeningsprosedure, kan deur voorsieningskettingbeplanners gebruik word om nuwe besighede te begin of om kleinhandelaars se premium produkaanbiedinge in mededingende besigheidsomgewings te onderskei.Alternate :Driven by the total cost of ownership, US-China trade and technology competition, and the COVID-19 pandemic, global supply chains are undergoing a major restructuring that will soon transform business and economics all over the world. Recently, supply chains with end-to-end integration for premium agri-food products have emerged as a new business model. This paper examines how they should function, and identifies the supply chain structure/production/business conditions necessary for their development. We study a premium supply chain consisting of many small farms that produce top-quality bananas, one integrator firm, and thousands of retail stores. We use industry and business data to calibrate a multiple-route flow-based model from farms to integrator to retailers/markets. We then use sensitivity analysis to illuminate the stakeholders' decision behaviour, and identify and discuss three main decision issues: contract farming, capacity strategy, and business robustness. For contract specification, contracting on price rather than quantity is conducive to coordinating the interests of the stakeholders. For the capacity strategy, the integrator should source raw products from many small farms rather than fewer large farms. For business robustness, the integrator could still ensure robust profits by regulating its product supply when new competitors arise or demand changes. These results are tested under various scenarios to determine the impact of input parameters or supply chain structure, and are verified with an industry practitioner who has experience with multiple premium agri-food products. The results, along with the flow model and its computation procedure, could be used by supply chain planners to start new businesses or to differentiate retailers' premium product offerings in competitive business environments.

Robustness and resilience of supply chains during the COVID-19 pandemic.

Using a unique firm-level data set from Asia, this study examines what determined the robustness and resilience of supply chain links, that is, the ability of maintaining links and recovering disrupted links by substitution, respectively, when firms faced economic shocks due to the spread of the coronavirus disease (COVID-19). We find that a supply chain link was likely to be robust if the link was between a foreign-owned firm and a firm located in the foreign-owned firm's home country, implying that homophily on a certain dimension generates strong ties and thus supply chain robustness. We also find that firms with geographic diversity of customers and suppliers tended to increase their transaction volume with one partner while decreasing the volume with others. This evidence shows that firms with diversified customers and suppliers are resilient, mitigating the damage from supply chain disruption through the substitution of partners. Furthermore, the robustness and resilience of supply chains are found to have led to higher performance.

Assessing the robustness of COVID-19 vaccine efficacy trials: systematic review and meta-analysis, January 2023.

BackgroundVaccines play a crucial role in the response to COVID-19 and their efficacy is thus of great importance.AimTo assess the robustness of COVID-19 vaccine efficacy (VE) trial results using the fragility index (FI) and fragility quotient (FQ) methodology.MethodsWe conducted a Cochrane and PRISMA-compliant systematic review and meta-analysis of COVID-19 VE trials published worldwide until 22 January 2023. We calculated the FI and FQ for all included studies and assessed their associations with selected trial characteristics using Wilcoxon rank sum tests and Kruskal-Wallis H tests. Spearman correlation coefficients and scatter plots were used to quantify the strength of correlation of FIs and FQs with trial characteristics.ResultsOf 6,032 screened records, we included 40 trials with 54 primary outcomes, comprising 909,404 participants with a median sample size per outcome of 13,993 (interquartile range (IQR): 8,534-25,519). The median FI and FQ was 62 (IQR: 22-123) and 0.50% (IQR: 0.24-0.92), respectively. FIs were positively associated with sample size (p < 0.001), and FQs were positively associated with type of blinding (p = 0.023). The Spearman correlation coefficient for FI with sample size was moderately strong (0.607), and weakly positive for FI and FQ with VE (0.138 and 0.161, respectively).ConclusionsThis was the largest study on trial robustness to date. Robustness of COVID-19 VE trials increased with sample size and varied considerably across several other important trial characteristics. The FI and FQ are valuable complementary parameters for the interpretation of trial results and should be reported alongside established trial outcome measures.

Digital Supply Chain Capabilities: Mitigating Disruptions and Leveraging Competitive Advantage Under COVID-19

Researchers and practitioners have highlighted the importance of supply chain analytic capabilities in managing risk while maintaining a competitive advantage (COA). However, the importance of digital supply chain capabilities (DSCCs) in improving resilience, agility, and robustness practices to foster the implementation of sustainable supply chain practices and any resulting COA remains unclear. Based on the dynamic capabilities view, we propose a research model for achieving a COA in contexts of uncertainty, such as the COVID-19 pandemic. A survey of Indian small and medium-sized enterprises in the original equipment manufacturing industry, comprising 310 respondents, was administered. Using structural equation modeling, we examine the proposed model. The findings show a significant positive effect of DSCCs on supply chain resilience and agile practices. The findings also indicate that supply chain resilience, robustness, and agile practices positively affect sustainable supply chain practices. Moreover, sustainable supply chain practices positively influence COA. Furthermore, the study reveals that the effect of DSCCs on sustainable supply chain practices is mediated by supply chain resilience, robustness, and agile practices. Managers concerned with investment in sustainable supply chain practices can obtain a COA through the successful implementation of supply chain resilience, robustness, and agile practices.

Untargeted white-box adversarial attack to break into deep leaning based COVID-19 monitoring face mask detection system.

The face mask detection system has been a valuable tool to combat COVID-19 by preventing its rapid transmission. This article demonstrated that the present deep learning-based face mask detection systems are vulnerable to adversarial attacks. We proposed a framework for a robust face mask detection system that is resistant to adversarial attacks. We first developed a face mask detection system by fine-tuning the MobileNetv2 model and training it on the custom-built dataset. The model performed exceptionally well, achieving 95.83% of accuracy on test data. Then, the model's performance is assessed using adversarial images calculated by the fast gradient sign method (FGSM). The FGSM attack reduced the model's classification accuracy from 95.83% to 14.53%, indicating that the adversarial attack on the proposed model severely damaged its performance. Finally, we illustrated that the proposed robust framework enhanced the model's resistance to adversarial attacks. Although there was a notable drop in the accuracy of the robust model on unseen clean data from 95.83% to 92.79%, the model performed exceptionally well, improving the accuracy from 14.53% to 92% on adversarial data. We expect our research to heighten awareness of adversarial attacks on COVID-19 monitoring systems and inspire others to protect healthcare systems from similar attacks.

Meatpacking working conditions and the spread of COVID-19

This paper explores how working conditions in meatpacking plants contributed to the spread of the COVID-19 virus. Data from the Occupational Information Network was used to construct a set of industry-level working condition variables and compare meatpacking to the sample of other manufacturing industries in our comparison group. This novel approach showed that proximity to others in the meatpacking industry is likely the main factor influencing the spread of COVID-19, more than three standard deviations higher in meatpacking than our comparison sample of other manufacturing industries. Subsequently, we performed a county-level analysis on COVID-19 spread, comparing rural counties with a large share of meatpacking workers to nonmetropolitan counties that were similarly dependent on other single manufacturing industries, using the time frame of mid-March to the end of 2020. In mid-April 2020, COVID-19 cases in meatpacking-dependent rural counties rose to more than 12 times compared to rural counties dependent on other single manufacturing industries. This difference disappeared completely by mid-July and held steady throughout the year. We demonstrate that our results are robust to a battery of robustness checks ruling out the set of plausible alternative hypotheses, including examining data on COVID-19 spread among meatpacking workers directly.

Secure NIfTI Image Authentication Scheme for Modern Healthcare System

Advances in digital neuroimaging technologies, i.e., MRI and CT scan technology, have radically changed illness diagnosis in the global healthcare system. Digital imaging technologies produce NIfTI images after scanning the patient's body. COVID-19 spared on a worldwide effort to detect the lung infection. CT scans have been performed on billions of COVID-19 patients in recent years, resulting in a massive amount of NIfTI images being produced and communicated over the internet for diagnosis. The dissemination of these medical photographs over the internet has resulted in a significant problem for the healthcare system to maintain its integrity, protect its intellectual property rights, and address other ethical considerations. Another significant issue is how radiologists recognize tempered medical images, sometimes leading to the wrong diagnosis. Thus, the healthcare system requires a robust and reliable watermarking method for these images. Several image watermarking approaches for .jpg, .dcm, .png, .bmp, and other image formats have been developed, but no substantial contribution to NIfTI images (.nii format) has been made. This research suggests a hybrid watermarking method for NIfTI images that employs Slantlet Transform (SLT), Lifting Wavelet Transform (LWT), and Arnold Cat Map. The suggested technique performed well against various attacks. Compared to earlier approaches, the results show that this method is more robust and invisible.

Do the macro and global economic factors drive the nonperforming loans in GCC economies?

PurposeThe current study aims to investigate the determinants of nonperforming loans (NPLs) in the GCC economies during the period spanning 2000 to 2018. It also examines whether the worldwide financial crisis of 2007–2008, which brought the issue of non–performing loans to the greater attention of academics and policymakers, had a substantial impact on NPLs in this region.Design/methodology/approachThe sample consists of 53 conventional banks from GCC countries, and the basic data for the study is obtained from various sources such as Bankscope, IMF World Economic Outlook, World Bank and Chicago Board of Options Exchange Market Volatility Index. The estimations were done by dynamic panel data regression modeling using system generalized methods of moments.FindingsThe findings reveal that both, the non-oil real GDP growth rate and inflation have favorable effects on NPLs. On the other hand, domestic credit to the private sector and the volatility index have an adverse effect on NPLs. Furthermore, the period-wise analysis shows that the relevance and significance of the determinants of NPLs vary between the precrisis and postcrisis periods. It is also reflected through the intercept dummy, which is found to be significant, indicating that the financial crisis, as a global economic factor, had a significant impact on NPLs. A number of robustness tests are applied, which indicate that the results are mostly robust and consistent in terms of the significance of the explanatory variables and the direction of their relationship with the dependent variable.Practical implicationsPolicymakers and bank authorities must strive to maintain a healthy economy and implement macroprudential policies to improve the financial stability of banks and reduce credit risk.Originality/valueTo the best of the authors' knowledge, this is likely the first study that empirically investigates the influence of the financial crisis on NPLs in the context of GCC economies. In addition, the research spans 19 years to produce more conclusive results.

Blockchain-Based Trusted Federated Learning with Pre-Trained Models for COVID-19 Detection

COVID-19 is a serious epidemic that not only endangers human health, but also wreaks havoc on the development of society. Recently, there has been research on using artificial intelligence (AI) techniques for COVID-19 detection. As AI has entered the era of big models, deep learning methods based on pre-trained models (PTMs) have become a focus of industrial applications. Federated learning (FL) enables the union of geographically isolated data, which can address the demands of big data for PTMs. However, the incompleteness of the healthcare system and the untrusted distribution of medical data make FL participants unreliable, and medical data also has strong privacy protection requirements. Our research aims to improve training efficiency and global model accuracy using PTMs for training in FL, reducing computation and communication. Meanwhile, we provide a secure aggregation rule using differential privacy and fully homomorphic encryption to achieve a privacy-preserving Byzantine robust federal learning scheme. In addition, we use blockchain to record the training process and we integrate a Byzantine fault tolerance consensus to further improve robustness. Finally, we conduct experiments on a publicly available dataset, and the experimental results show that our scheme is effective with privacy-preserving and robustness. The final trained models achieve better performance on the positive prediction and severe prediction tasks, with an accuracy of 85.00% and 85.06%, respectively. Thus, this indicates that our study is able to provide reliable results for COVID-19 detection.

Predicting the effects of supply chain resilience and robustness on COVID-19 impacts and performance: Empirical investigation through resources orchestration perspective.

This study investigates the effects of supply chain (SCRE) and robustness (SCRO) on COVID-19 super disruption impacts and firm's financial performance by mobilizing the resources orchestration theory (ROT) as the main theoretical framework. We adopt structural equation modeling analysis of data collected from 289 French companies. The findings reveal the significantly positive influence of resources orchestration on SCRE and SCRO and the role of the latter in mitigating the pandemic disruption impacts. Notwithstanding, depending on whether the measures are objective or subjective, the effects of SCRE and SCRO on financial performance vary. Overall, this paper presents empirical evidence of the influence of both of SCRE and SCRO on pandemic disruption impacts and financial performance. Furthermore, this research provides insights to guide practitioners and decision makers regarding resources orchestration and the deployment of SCRE and SCRO.

A comparative analysis of instructional delivery methods in a higher education institution in the United Arab Emirates

This comparative quantitative study aims at investigating whether instructional delivery methods, such as online, hybrid, blended learning and face-to-face delivery methods, had an effect on students' grades when teaching mathematics to English language learners in a higher education Institution in the United Arab Emirates. Final course grades, in GPA format, of 574 students were collected over the course of three academic years. Assumptions of analysis of variance (ANOVA), post hoc tests, effect sizes Cohen's d were examined. The statistically significant difference across four different instructional delivery methods showed effect sizes that grew from medium to large to very large when increasing the amount of online instruction. These findings surpass other studies and suggest a high practical significance. The consequences of COVID19 on instructional delivery methods had a favourable effect on students' grades. Higher education institutions in the UAE may use these findings for future planning, even beyond the pandemic. Recommendations are made for further research to include more variables and other disciplines.

Generalisable long COVID subtypes: Findings from the NIH N3C and RECOVER programmes

Background Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterised by a wide range of manifestations that are difficult to analyse computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested.Methods We present a method for computationally modelling PASC phenotype data based on electronic healthcare records (EHRs) and for assessing pairwise phenotypic similarity between patients using semantic similarity. Our approach defines a nonlinear similarity function that maps from a feature space of phenotypic abnormalities to a matrix of pairwise patient similarity that can be clustered using unsupervised machine learning.Findings We found six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centres to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems. Interpretation Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC.

Do talent management practices improve organizational resilience? An empirical study within Tunisian companies

Purpose: The purpose of this research is to study the impact of talent management practices on organizational resilience in Tunisian firms in times of the sanitary crises due to COVID-19. Design/methodology/approach: A hypothetico-deductive approach is adopted. First, it is hypothesized that four talent management practices positively affect organizational resilience. Then, the hypotheses were tested by using quantitative methods. Data were collected through questionnaires and analyzed with PLS-SEM techniques. Findings: Results show that talent identification positively affects organizational resilience operationalized through the three dimensions of agility, integrity and robustness. Talent development and talent succession planning positively influence the firms' agility only, whilst talent retention had no effect on the three organizational resilience dimensions. Practical implications: The findings of this research may be helpful for human resources managers to recognize among talent management practices those that are mostly associated with organizational resilience and its dimensions. This could help them revise some talent management practices and implement those that are lacking to ensure strong and resilient firms, especially in a context characterized by the occurrence of crises of different natures. Originality/value: The literature review showed that talent management practices and organizational resilience relationships are understudied. This research empirically highlights the relevance of the linkage between them. It contributes to the rare existent works by identifying a significant effect of talent identification on all organizational resilience dimensions and a positive effect of talent development and succession planning on agility. © 2023, Emerald Publishing Limited.

Resilience Improvement and Risk Management of Multimodal Transport Logistics in the Post–COVID-19 Era: The Case of TIR-Based Sea–Road Multimodal Transport Logistics

The COVID-19 pandemic has severely impacted international economics and trade, including cargo transportation. As a result, enhancing the resilience of transport and logistics in the post–COVID-19 era has become a general trend. Multimodal transport, with its advantages of speed, large volume and multiple modes, has increasingly gained attention from countries worldwide. However, multimodal transport logistics is a complex and systematic process. Its smooth flow depends not only on the transport itself, but also on the efficient supervision of customs and other government departments at ports. This study employs the theory and method of a super-network to establish a model of multimodal transport logistics, which includes TIR-based sea–road multimodal transport and customs supervision relationships. Structural and resilience-related characteristics of the super-network are analyzed, and performance parameters of the super-network are proposed. A simulation analysis is conducted, and based on the results, countermeasures to improve the resilience and promote risk management of multimodal transport logistics in the post–COVID-19 era are suggested. The findings of this study provide an exploration of more effective ways to ensure the smoothness of multimodal transport logistics and improve system resilience. The study concludes with theoretical and managerial implications. © 2023 by the authors.

Robustness of Support Vector Regression and Random Forest Models: A Simulation Study

Classical statistics are usually based on parametric models, where the performance depends heavily on assumptions and is not robust in the presence of outliers in the data. Due to the COVID-19 pandemic, our daily lives have changed significantly, including slowing economic growth. These extreme changes can manifest as an outlier in time series studies and adversely affect the results of data analysis. Many classical methods of official statistics are prone to outliers. In this work, we evaluate machine learning methods: Support Vector Regression (SVR) and Random Forest (RF) and compare it with ARIMA to determine the robustness through simulation studies. Robustness is measured by the sensitivity of the SVR and Random Forest hyperparameter and the model's error in the presence of outliers. Simulations show that more outliers lead to higher RMSE values, and conversely, more samples lead to lower RMSE values. The type of outliers significantly impacts the RMSE value of the ARIMA model, where additional outliers (AO) have a worse impact than temporary change (TC). Consecutive outliers produce a smaller RMSE mean than non-consecutive outliers. Based on the sensitivity of hyperparameters, SVR and Random Forest models are relatively robust to the presence of outliers in the data. Based on the simulation results of 100 iterations, we find that SVR is more robust than ARIMA and Random Forest in modeling time series data with outliers. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.