Safety, environmental and risk management related to Covid-19.

The year 2020 will be remembered for the start of the Covid-19 global pandemic and will emerge as a great source of change in the safety community. In fact, in March 2020, the World Health Organization (WHO) declared a pandemic state and the Covid-19 pandemic continues across the globe, with, at the time of writing, over 414 million cases and with over 5.8 million fatalities reported across all countries and territories (WHO, 2022). Alongside this, developing advanced techniques to enhance process safety, minimize environmental impacts and minimize risk under pandemic crises attracted significant attention, especially considering the still continuing emergence in various countries and concern about subsequent waves or even continual Covid-19, notwithstanding vaccine development. The evolution of the current pandemic is still affected by an extreme uncertainty connected to many unknown unknowns, allowing future challenges in research covering many topics, such as risk assessment of the impacts of a pandemic on the safe operation of process plants, including risks of shutdown and start-up under pandemic conditions, re-purposing process plants for responses to pandemics, cross-fertilization from different industrial sectors and trans-disciplinary approaches to personnel and process safety. Relevant research issues also include potential risks to the environment due to changes in monitoring, enforcement measures and activities as well as novel technologies and approaches for the mitigation of environmental impacts of Covid-19.

This special issue focuses on environmental, safety and risk analysis and management under different Covid-19 waves and demonstrates the need for novel approaches to re-engineering process safety including a critical and balanced application of novel developments in data science and digital technologies with fundamental science and engineering principles (Pasman and Fabiano, 2021). Readers will therefore not be surprised to discover that six papers in this special issue primarily focus on investigating the epidemiological characteristics of Covid-19 and minimizing impact and damage on human health relying on different Artificial Intelligence (AI) approaches, thus trying to capture meaningful signals about pandemic changes modifications in the dynamic real-world scenario. Efforts were dedicated to how epidemiological and engineering approaches can be put together for pandemic risk assessment and management. In this regard, the pandemic offers many learning opportunities to improve engineering risk management practices. By drawing a parallel between the two domains, we believe that the lessons learned from the Covid-19 pandemic will immensely benefit engineering safety personnel and healthcare experts in efficient policymaking. Regarding the main perspective of the contributions, the first five papers of this issue rely on concepts, methodologies derived and applied in process safety, whilst others are at the forefront of basic research and reflect new thinking, or new challenges in risk management and environmental protection, whose applicability can go beyond the Covid-19 pandemic.

Within the first theme of this issue, the paper by Alauddin et al. (2021) is related to the wide topic of management of pandemic risk by the implementation of engineering safety principles and the development of hierarchical risk reduction strategies. Several areas of similarities were identified where process safety and epidemiology could benefit from each other, including early fault detection vs early case detection, identification of effective control mechanism, the fast response of public health vs operator response, effective resource allocation and mobilization, identification of the most vulnerable elements, and application of expertise from similar outbreaks in the past vs use of historical process data. The study proposes the implementation of specific measures and distinct risk-reducing strategies, realized due to natural evolution, government interventions, societal responses, and individual practices are integrated with the event tree analysis.

The paper by Brown et al. (2021) shows the application of bow-tie analysis to the Covid-19 pandemic, including the explicit application of inherently safer design (ISD) principles as a novel contribution to the state of the art. The prioritizing of risks for analysis on a qualitative basis is completed in the study by using the judgement of the personnel performing the analysis. The proposed approach may allow further improvement by completing a bow tie workshop, through a partnership with healthcare experts specialized in infection prevention and control.

The third paper, authored by Seiti et al. (2022) proposes a causal mathematical model to estimate variability and risk of interactive factors while considering different scenarios in a causal chain of various factors. It uses data gathered from experts and the outputs are easily interpretable and explicable by decision-makers.

The contribution by Wood et al. (2021) addresses the risks entailed within intensive care units by the elevated use of medical oxygen during the pandemic and recommends safety management strategies for prevention and emergency preparedness, based on a lesson learned approach by the analysis of more than fifty oxygen-related fires across scientific literature and media reports. Results outline the need to adopt more rigorous risk management approaches in hospitals, including borrowing from strategies developed for chemical process safety to manage explosive atmospheres.

The purpose of the last paper of this group, by Nnaji et al. (2022) is to describe the fieldworker’s perspective of Covid-related safety measures used on construction projects, whether they are satisfied with the preventive measures, and whether there is a difference in perception based on demographic characteristics and organizational factors. The findings on the perceptions of the effectiveness of different prevention methods suggest differences between the perceptions of the front line workers and the management, highlighting the importance of proper management commitment and safety and health policy enforcement.

As already stated, the main body of this issue features papers integrating epidemiological and engineering approaches for Covid-19 management, according to the ongoing evolution in the safety domain towards more precise and continuously updated predictions based on data science and AI. The first paper of this group by Li et al. (2022) proposes a new prediction model based on gradient-based optimizer variational mode decomposition (GVMD), extreme learning machine (ELM), and autoregressive integrated moving average (ARIMA), named GVMD-ELM-ARIMA. The proposed method for the prediction of cumulative Covid-19 diagnostic data was applied to the latest epidemic data of the United States, India and Russia to carry out multi-step prediction evidencing a fairly good prediction accuracy and indicating as well the need of further refinement to extend the reliable prediction extension.

Under the same theme, Marzouk et al. (2021) present the application of two deep learning models, namely the long short-term memory network (LSTM) and convolutional neural network (CNN), to predict the spread of the Covid-19 outbreak in Egypt. Results evidence that the LSTM model outperformed the other deep learning and machine learning models for one week and one month-ahead forecasting, due to its ability to propagate data in the backward pass and to capture the nonlinear time pattern in the input data.

The paper by Jin et al. (2021) deals with models and data analysis for spatial-temporal characterization on Covid-19 pandemic in China, moreover, the focus is also devoted to provinces and cities to provide evidence of the effectiveness of the epidemic prevention measures adopted by the governments at all levels. Based on the experimental results of curve-fitting and computational time cost authors suggest the polynomial fitting as the best approach for the analysis of the daily-change tendency of the number of confirmed cases.

The paper by Al-qaness et al. (2021) compares the prediction capabilities of four artificial intelligence-based models for forecasting the total Covid-19 cases in two hotspot countries, Russia and Brazil, starting from WHO reported cases. The results of an improved marine predators algorithm, chaotic MPA (CMPA) outperformed the other algorithms in the experiment measures demonstrating its ability to balance between exploration and exploitation phases, fast convergence and an acceptable computation time.

In the last paper within this theme authored by Elsheikh et al. (2021), long short-term memory (LSTM) network is introduced as a robust deep learning model to forecast the number of total confirmed cases, total recovered cases and total deaths in Saudi Arabia. The model was trained using the official data reported by Saudi Ministry of Health and the optimal values of the model's parameters that maximize the forecasting accuracy were determined.

The contribution by Kumar et al. (2021) represents the logical trait d′union with the last broad theme of this issue, as it focuses on the development of an AI-based automated solution for sorting Covid-19 related medical waste streams from other waste, allowing as well reliable data-driven decisions for recycling in the context of circular economy. Their findings indicate that the classification done using a decision level fusion scheme and machine learning techniques is significantly superior to the single best performance feature scheme for the correct separation of waste in recycling categories—four trash categories, i.e. glass, metal, Covid-19 waste, and plastic.

As anticipated, the last four papers of this issue focus on potential risks to the environment, short and long-term environmental implications of the Covid-19 crisis, as well as on novel technologies and approaches to mitigation of environmental impacts of Covid-19. The first one, by Elsaid et al. (2021) reports a web-based critical review on heating, ventilation, and air conditioning (HVAC) systems in relation to their impact on the spread of the hybrid SARS-CoV-2 pandemic, deriving practical recommendations addressing contact prevention and protection.

The paper by Zoran et al. (2021) explores the relation between the change in weather-related factors and the Covid-19 incidence and severity in the Madrid region during the first, second and third Covid-19 pandemic waves, based on daily in-situ and geospatial time series data statistical analysis and Spearman rank correlation tests. It is presented how different synoptic atmospheric circulation patterns, affecting turbulent processes and wind circulations at regional and local scales, may be related to the three Covid-19 waves start-up and regional evolution in the Madrid metropolitan region. Accurate estimation of the local and regional seasonality of the environmental and epidemiological conditions can provide useful information on the seasonality characteristics of the effects of the future Covid-19 disease impact.

The review paper by Nippes et al. (2021) reports about the presence of the drugs chloroquine, hydroxychloroquine, azithromycin, ivermectin, dexamethasone, remdesivir, favipiravir and antivirals for HIV in environmental matrices. Even though advanced theoretical investigation and experimental is required, the survey showed the presence of these drugs in different regions, with concentration data that may be adopted as a benchmark for comparative studies, considering the possible increase of the environmental exposure to these drugs also due to future pandemic waves.

The last paper of the collection, by Iqbal et al. (2021) is focused on exploring the relationship between NCOV-2019 cases and deaths with the exposure to air pollutants (PM2.5, SO2, PM10, NO2, and O3) in China. Empirical outcomes of the negative binomial regression model analysis show the interdisciplinary research challenges for a comprehensive analysis of the relationship between air pollution and indirect interaction with viral vectors and possible negative impact on the country’s public health.

Conclusions

This special issue was designed to widen and deepen the current discussion on Covid 19, relying on the experience gained during the various waves of the pandemic emergency, hopefully to provide a foundation for researchers, professionals and policymakers interested in developing resilience and adaptive capacity at the organizational, or societal level.

As illustrated by the papers in this issue, notwithstanding noteworthy advancements, inherent safety and resilience are still common sense rather than common practice, so risk mitigation techniques and modelling remain up-to-date research topics in the broad field of process safety and environmental protection. The pandemic crisis represents a perfect field for transdisciplinary and multi-cultural evaluation of real complex systems and hopefully open up a new vista and perspective on methodological improvements, necessary in the ever-increasing complexity of safe processing manufacturing and distribution in a competitive world.

The special issue shows that the experience of process safety in the management of risk of complex systems has shown that the transfer of experience, e.g. Bow-Tie analysis, and learning lessons from incidents, to other realms of risk management is a valuable knowledge sharing opportunity.