Data analytics and decision-making systems: Implications of the global outbreaks

1. Introduction

The COVID-19 pandemic has disrupted the goals and functions of technology-based systems around the world [16]. From a health perspective, the World Health Organization has reported that over six million have died of the disease. Government efforts to control the spread of COVID-19 have had drastic impact on global supply chains. The world's health systems have been heavily taxed. Travel is difficult anywhere, and more difficult internationally. Economies are still in downturns [13], as many workers have been taken out of the economy for indefinite periods of time, and the structure of work has radically changed, with many working remotely [11]. This in turn has led to vacant offices, as well as a migration from inner cities to suburbs in some countries. The impacts to society and human quality of life have been severe [3].

There are a number of important critical decisions when coping with and managing pandemics. First, there is a need to identify threats at early stages. Then means to cope with the pandemic are critical. The first system to see impact is typically the medical system [5]. The COVID-19 pandemic stressed hospital and care-giver capacities to the limit. Transportation networks also experienced extreme stress [7]. Businesses suffered from the removal of workers and customers, and governments struggled with policies to balanced control of the spread with the need for economies to provide their people with food, shelter, and transportation [4,10,12].

The capacity of pandemic preparedness to confront these threats needed strengthening [8]. The call for papers for this Special Issue stated a need for research in the effectiveness of preparedness systems, for epidemic monitoring, and for means to stabilize economic activity and reduce systematic risks. It called for research on high-performance decision support systems to track verified events and their impact on public health and financial markets, as well as for means to suggest proper and efficient reactions to pandemic threats.

Data analytics and artificial intelligence-based decision support technologies have been applied to deal with epidemic diseases in the past, to include preemption, prevention, and combating threats of infectious disease epidemics [15]. Also important is better understanding of health-seeking behaviors and control of public emotions during pandemics [9].

2. Decision support system research

The Decision Support Systems journal has a tradition of publishing articles applying artificial intelligence and analytic models to support decisions in a variety of decision-making contexts. Recent articles related to disaster and risk management issues include Chaudhuri and Bose [2] who applied artificial intelligence in the form of neural networks in a disaster management context, providing tools for emergency response planners to more effectively plan response. Geo-tagged images from regions hit by earthquakes provided image data from smart infrastructures. Deep learning models classified images to aid in identification of survivors among debris.

Soon after the onset of COVID-19, Lin et al. [6] addressed the business value of information technology to Chinese agricultural firms. The impact of e-commerce capabilities on business agility in agribusinesses was sampled. It was found that e-commerce capabilities enabled market capitalizing agility as well as operational adjustment capability. Their research also found that environmental complexity positively moderated the effects of e-commerce capability on both forms of agility, while environmental dynamism did not.

Cardonha et al. [1] studied the impact of COVID-19 on university scheduling and delivery of teaching. They presented a decision support system to aid in assigning teaching modalities and rooms at the University of Connecticut, incorporating safety standards to include reduction of room seating capacities, while maximizing opportunities for in-person instruction. Two mixed integer programming models were presented and evaluated in terms of dealing with the complexity of the assignment problem involving hundreds of rooms and thousands of classes.

Wu et al. [14] analyzed financial risk forecasting in the Chinese banking system in light of the uncertainty imposed by COVID-19. Unexpected rapid growth over 2021 demonstrated high levels of uncertainty in real investment decisions. A multi-layer perceptron model applying neural network technology was compared with the traditional Altman Z-score model. These methods were combined to develop a new hybrid enterprise crisis warning model capable of providing early warning signals of a company's deteriorating financial situation.

3. Special issue contents

This Special Issue continues the Decision Support System journal focus on models to aid analysis of decision problems. Six papers covering a variety of topics related to problems created by COVID-19 are presented.

Kang et al. studied travel restrictions imposed during the Chinese Lunar New Year celebrations following COVID-19. Scheduled trains on China's railway system were either heavily adjusted or canceled. A mixed-integer linear programming model and a two-step solution algorithm were developed to handle large-scale adjustments to these schedules. The aims were to obtain flexible time windows for each operation line and locomotive traction operations while minimizing the number of locomotives and total idle time. The uncertainty created by COVID-19 led to development of two tailored approaches. Kang et al. present a case study of the Beijing-Tianjin intercity railway from the start of the COVID-19 outbreak to the recovery of operations.

Ertem et al. studied the COVID-19 pandemic impact as a public health problem. They present an efficient decision analytic approach for assessment of the effectiveness of early social distancing measures in communities with different population characteristics. An empirical estimate of reproduction numbers for two different states was obtained. The study developed an age-structured compartmental simulation model for the disease spread to demonstrate the variation in the observed outbreak. Computational results were presented and analyzed, finding that early triggering of social distancing strategies would result in smaller death tolls. The impact of relatively larger second waves was identified. These second waves led to late trigger social distancing strategies resulting in higher initial death tolls but relatively smaller subsequent waves. Ertem et al. demonstrated that decision analytic tools can help policy makers simulate different social distancing scenarios in the early stages of a global outbreak.

Kazemi Matin et al. looked at blood supply chains, which play a strategic and crucial role in healthcare systems, especially in unexpected situations such as earthquakes and pandemic outbreaks. Measuring the sustainability and resilience of blood supply chains is a major challenge for many decision-makers in healthcare systems. This paper presented an advanced network data envelopment analysis method to evaluate the sustainability and resilience of blood supply chains. A new directional distance function was developed for evaluating both the overall and stage efficiency scores. A case study demonstrated the usefulness of the proposed model.

Davazdahemami et al. examined the time required to identify and validate pandemic risk factors. Disease impact analysis traditionally involves numerous clinical trials that may take several years, during which strict preventive measures must be in place to control the outbreak and reduce deaths. Davazdahemami et al. proposed advanced data analytics techniques to guide and speed this process. Evolutionary search algorithms were combined with deep learning and advanced model interpretation methods to develop a holistic exploratory-predictive-explanatory machine learning framework to assist clinical decision-makers in reacting to the challenges of a pandemic in a timely manner. The proposed framework was demonstrated with emergency department (ED) readmissions of COVID-19 patients using ED visits from a real-world electronic health records database. After an exploratory feature selection phase using a genetic algorithm, a deep artificial neural network was applied to predict early (i.e., 7-day) readmissions (AUC = 0.883). Then a model was formulated to estimate additive Shapley values (i.e., importance scores) of the features and to interpret the magnitude and direction of their effects.

The global outbreak of COVID-19 has significantly changed firms' strategic decision-making landscapes and created complex effects of IT-business alignment and big data analytics capability on strategic decision-making. Chen et al. contextualized event criticality and event disruption in the context of COVID-19 and examined their contingent roles in the effects of IT-business alignment and big data analytics capability on strategic decision-making. Two rounds of surveys of respondents from 175 Chinese firms were studied to elucidate the differential moderating roles of event criticality and disruption of COVID-19. Results indicated that the event criticality of COVID-19 strengthened the effects of IT-business alignment on decision speed and quality but weakened the influence of big data analytics capabilities on decision quality.

Dhar and Bose studied COVID-19 pandemic lockdown on organizations across the globe. They proposed that organizations needed to strategize their crisis responses and communicate with stakeholders to reduce the threat to reputational capital and manage stakeholder reactions in the pandemic. Twitter communications during the COVID-19 crisis were studied through the lens of situational crisis communication theory. Dhar and Bose analyzed 325,627 tweets collected from the Twitter pages of 464 organizations belonging to the Fortune 500 list. The Twitter data reflected organizational COVID-19 crisis response strategies and demonstrated organizational use of Twitter for crisis communication. Lexicon-based emotion mining was applied to identify and measure emotions. Topic mining was used to measure crisis response scores from this multi-organization dataset. Path analysis was used to test the research model. The analysis found that instructing and adjusting information can minimize threats to organizational reputation in a victim crisis and manage stakeholder reactions. Positive emotions had a stronger association with behavioral outcomes. Emotion neutral tweets generated more favorable stakeholder reactions.

4. Conclusions

COVID-19 has had a major impact on the world, impacting health, government policy, economics, and supply chain operations. The scientific community is still working through analyses of these impacts. Analytics and artificial intelligence models offer tools that can aid humans in dealing with many issues. There has been a massive stream of published research on COVID-19 impact. This Special Issue continues the tradition of the Decision Support Systems journal to publish modeling efforts to deal with disaster and risk decision-making.

Biographies

Desheng Wu is with the School of Economics and Management, University of Chinese Academy of Sciences, Beijing, China. He has authored or coauthored more than 150 papers in refereed journals such as Production and Operations Management, JMIS, Decision Sciences, Risk Analysis, IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, etc. His research interests include enterprise risk management in operations, data intelligence, and decision sciences. Dr. Wu has been an Associate Editor/Guest Editor for more than 10 reputable journals such as IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, Risk Analysis, Omega, etc. He is elected member of Academia Europaea and elected member of European Academy of Sciences and Arts.

David L. Olson is the James & H.K. Stuart Professor and Chancellor's Professor at the University of Nebraska. He has published research in over 200 refereed journal articles, primarily on the topic of multiple objective decision-making, information technology, supply chain risk management, and data mining. He has authored over 40 books, to include Decision Aids for Selection Problems, Introduction to Information Systems Project Management, Managerial Issues of Enterprise Resource Planning Systems, Supply Chain Risk Management, and Supply Chain Information Technology. He has served as associate editor of Service Business, Decision Support Systems, and Decision Sciences and co-editor in chief of International Journal of Services Sciences. He has made over 200 presentations at international and national conferences on research topics. He is a member of the Decision Sciences Institute, the Institute for Operations Research and Management Sciences, and the Multiple Criteria Decision Making Society. He was a Lowry Mays endowed Professor at Texas A&M University from 1999 to 2001. He was named the Raymond E. Miles Distinguished Scholar award for 2002, and was a James C. and Rhonda Seacrest Fellow from 2005 to 2006. He was named Best Enterprise Information Systems Educator by IFIP in 2006. He is a Fellow of the Decision Sciences Institute.

James H. Lambert is a Professor of Engineering Systems & Environment at the University of Virginia, USA. He is a Site Director of the US National Science Foundation Center for Hardware & Embedded Systems Security & Trust. He is the Director of the Center for Risk Management of Engineering Systems. He is a Fellow of each of the AAAS, IEEE, ASCE, and Society for Risk Analysis. His areas of interest are systems engineering and risk analysis, with a focus on the disruption of priorities. He was Chair of the Eighth Annual Symposium on Engineering Systems, Chair of the Fifth World Congress on Risk, and Chair of the 2015 Annual Meeting of the Society for Risk Analysis. He is a past President (2015–2016) of the worldwide Society for Risk Analysis. He is Editor-in-Chief of the Springer journal Environment Systems & Decisions, and an Area Editor of the Wiley journal Risk Analysis. He is a recipient of best paper awards from the IEEE, American Chemical Society, and others. He is a member of the Technical Advisory Council of the Commonwealth Center for Advanced Logistics Systems. He served on the Standing Committee on Health Threats and Workforce Resilience of the US National Academies of Sciences, Engineering and Medicine.