Analysis of Critical Infrastructures: Risk Identification and Reduction

#### **Chapter 2**

## Resilience of Infrastructures and Systems to Multiple Hazardous Events: Application Cases and Future Perspectives

*Clemente Fuggini, Miltiadis Kontogeorgos, Saimir Osmani and Fabio Bolletta*

#### **Abstract**

Nowadays, Critical Infrastructure and Systems are getting more and more interconnected, while facing increasing and more intensive hazards: from man-made to natural ones, including those exacerbated by effects of the climate change. The demand for their robustness and resiliency against all these threats is finding ground to organizations' or states' ambitions and policies. The paper focuses on a review from an engineering perspective of past efforts and more importantly provides evidence of application cases the authors have developed in the past years. Finally, an outlook on future perspectives and potentials in the application of resilience is provided.

**Keywords:** resilience, hazards, complex systems, applications, impact assessment

#### **1. Introduction**

In today's landscape and emerging world, the significance of the Infrastructures and Systems (from Energy to Transportation ones) is becoming more and more critical for the well-tempered function of the states and communities. The increased demands in cities' energy consumption, the ever-expanding Transportation and Energy grids, the interconnection of these Systems and Infrastructures are some exemplary issues of this high criticality. In addition to these, the natural hazards due to climate change are appearing of higher magnitude and are causing more severe damages, estimating to billions of dollars worldwide annually [1], while future projections are predicting an increase of these costs the forthcoming years and decades [2–4]. Although the macroeconomic costs of the impacts due to climate change are highly uncertain, it is very likely to threaten development in many countries [5]. In addition, the man-made threats are always present for the global community, even expanding, due to terrorism, cyber-crime, and wars.

Under this prism, states and communities have already started to develop and set in force frameworks for robustness of their Infrastructures and Systems, both for their internal cohesion and uninterrupted continuity and for the more efficient co-operation among them in the international level. In the global sphere, cornerstones of global frameworks can be considered the Paris agreement [6] and the Sendai framework [7]. The Paris agreement is aiming to avoid more extreme natural phenomena and dangerous consequences due to climate change by taking measures in favor of global average temperature reduction, but also by enhancing societies' and states' capability to reduce the impacts of climate change. The Sendai framework has set up the priorities for actions for an effective disaster risk reduction and a resilient approach to these common threats, by understanding the disaster risk management to the enhancement of the disaster preparedness for effective response [7]. Furthermore, states or unions of countries (e.g., EU) have also developed their own frameworks [8–11], aligning simultaneously with the international agreements and goals, and showing special providence to cyber-resilience [12, 13].

The scientific and research community has faced the challenges and the demands for empowering the resiliency of Infrastructures and Systems. This was achieved by investigating many aspects of the resiliency planning against various hazards and developing in a scientific manner respective assessment and enhancement frameworks and tools. After the resilience conceptualization in a qualitative form, various quantitative metrics and approaches are suggested. Quantitative methods for assessing the resilience of Infrastructure Systems were proposed from many authors [14–16], also considering the interdependency of the Infrastructure Systems [17, 18] and expanding the field of study to the level of the communities [19]. Novel methodologies for analyzing Critical Infrastructure resilience were presented, with pilot implementation cases included as experimental part also [20]. As it was expected, studies were conducted also for examining the resilience capacity against specific hazards such as earthquake [21] or hurricane [22]. A special interest was shown for the resilience enhancement of Transport and Energy Infrastructures, due to their critical and multilevel meaning for the states' vitality and function.

The resilience of the Urban Transportation System was of interest for many researchers, and so many assessment methods were proposed [23, 24], including also multi-dimensional approach [25] and individual vertex-based and edge-based failure models [26]. The research has expanded beyond the Urban Transportation Systems, including also Railway [27, 28] and common Road and Transportation Systems [29, 30]. The factor of security has been highlighted, especially toward terrorism [31]. The Transportation Infrastructures and Systems have been tested also against various hazards, such as earthquake [32], extreme climatic and weather events [33, 34], and tsunami [35].

The vulnerability of Energy Infrastructures and Power Systems mainly to natural disasters due to climate change, but also due to manmade hazards, led the scientific community to develop assessment methods and solutions to increase the resilience capacity of these Systems. Frameworks and methods for the characterization of the resilience level of various types of Energy Infrastructures systems were proposed, such as for Nuclear Plants [36] and Hydrogen Systems [37]. Energy and Power Systems have been tested for their resilience capacity against various types of hazards such as hurricane [38], earthquake [39], or flooding [40]. In recent years, the resilience of the Energy Grids in the operational level [41], toward natural [42] or cyber [43] hazards, is being investigated thoroughly.

Although the multi-step progression in the definition of resiliency frameworks and the development of robustness' methods, Infrastructures and Systems are presenting a partial lack of efficient toolkits against multiple extreme events. Moreover, the cyber hazards are becoming more numerous and dangerous within the operational phase of the Systems, and the convergence of safety and cybersecurity has not been incorporated yet within the policies and the frameworks, which these Systems are following for their protection.

#### **2. Methodology**

A crucial element for the design of an efficient resiliency planning for Infrastructures and Systems, aiming at their protection toward multiple hazardous events, is the adopted methodology. Due to the flexibility that demands in order to be feasible for implementation to various assets and against various hazards, the methodology is setting some standard steps toward the resilience design, and then it is focusing partially on the specific asset and threat for every case. The procedure followed is being described in the **Figure 1**.

The initial step refers to the description and the characterization of the entire System (i.e., Energy System, Healthcare system), including all the necessary information, which is needed in the resiliency analysis. These can be the placement of the System toward the external environment and the interdependency between the System's elements or assets.

The next phase contains the asset characterization. This step refers more to technical information for the asset of our interest. For example, if the resilience planning aims at Energy Infrastructures, the characterization is needed to include the type of the Infrastructure (i.e., gas transmission pipeline, refineries, power plants) and then the design details and the qualifying characteristics. In case of a bridge asset, it needed the type of bridge (i.e., suspended) and the design details of the project (i.e., type of foundation, reinforcement blueprints), and so forth, regarding every asset under resilience evaluation.

The third step is about the threat characterization. The identification of potential threats and hazards is carried out, after the evaluation of the disruptive event's magnitude and criticality, and the definition of relevant hazard scenarios is taking place. According to the examined hazards, a modeling of them is following, respecting the literature database and the national codes and frameworks. In this step are included calculations such as the creation of probabilistic seismic curves due to site characteristics or the probabilistic scenarios for landslides and floodings based on site and meteorological data available.

The first three steps were decided more as part of constructing the profile of the interested Infrastructure or System and then defining the problem. The fourth and fifth steps are closer to final goal of the methodology, and they are the core of the followed philosophy. So, the fourth step is devoted to the risk and vulnerability assessment, along with the impact analysis. A quantitative expression based on failure probabilities for the examined Infrastructure or System is conducted, under the form of a vulnerability curve. This step is connecting the existing situation and behavior of the asset toward multiple scenarios of the hazards' magnitude, by defining the failure limits of the asset. Furthermore, regarding the magnitude and the intensity of an event beyond the capacity's limits, an impact analysis is conducted. The impact analysis takes into account the economic, social, environmental, and human losses aspects of a disruptive event and determines the total direct and indirect losses caused to the stakeholders (i.e., energy operators, civil protection, state) and the society. Again, the analysis contains some general criteria and aspects covered, but is delving also into specific impact details and information, respectively, to the type of the asset.

*Critical Infrastructure – Modern Approach and New Developments*

**Figure 1.** *Schematic representation of the methodology's philosophy.*

*Resilience of Infrastructures and Systems to Multiple Hazardous Events: Application Cases… DOI: http://dx.doi.org/10.5772/intechopen.108548*

**Figure 2.** *Depiction of the resilience dimensions, in terms of space, time, and type of resilience analysis.*

The last and final step is including the desirable resilience assessment. After the processing of all the necessary information toward the asset and the investigated threats, the level of service and the resilience capacity are defined. These are expressed in a quantitative form, as for this step, a set of various resilience indicators and a resilience matrix have been developed and exploited. The resilience matrix contains the robustness, rapidity, resourcefulness, redundancy sections, and after the evaluation on these specific domains, a grading of the behavior and response of the asset or the system is calculated toward a unique or multiple hazardous events.

The range of applications for a methodological approach of this type, in terms of time, space, and aspects of resilience assessment, is depicted briefly in the **Figure 2**. The analysis can be targeted to examine the operational phase of an Infrastructure or System, but also the emergency and post-recovery phase after a disruptive event. The space covered and examined in the resilience analysis can begin from a single building/ structure (i.e., cultural heritage building toward seismic hazard) and extend to a whole province or country (i.e., the national gas transmission or transport network). Finally, the aspects of resilience analysis that will contribute to the calculation of the total resilience capacity can include the social or the environmental factor, except those for the technical level of the System and the economic depiction of a disruptive event.

#### **3. Application cases**

The applications that are presented by the authors are selected in order to cover both Transportation Systems' and Energy Infrastructures' fields, by implementing the before-mentioned methodology approach and describing promising technologies for the increase of the resilience. The applications are derived or inspired from EU-funded projects. More specifically, the chosen application as an exemplary case for the Transportation Systems is describing the FORESEE project. This application is focusing on short- and long-term resilience schemes for rail and road corridors and logistics terminals. The Energy Infrastructure case is depicting the SecureGas project, which is dealing with the strengthening of the security and resilience of the European Gas Network, regarding the physical and cyber threats. Finally, the presentation of the INSPIRE project serves as an introduction to the beneficial use of potential metamaterials concepts within Infrastructures (especially Energy) and Systems, regarding their protection toward dynamic-nature hazards.

#### **3.1 Transportation system**

The main goal of the FORESEE application was to provide road authorities and managers, responsible for the rail and road corridors and logistics terminals, with a solution to anticipate, absorb, adapt, and rapidly recover from a potentially disruptive hazard or extreme event during the entire lifecycle of the transport infrastructure: planning, design, construction, operation, and maintenance [44]. In order to achieve this goal, the proposed methodology is implemented in all its five steps, and a toolkit, which is capable of collecting data for predicting the magnitude and the potential damage of various hazards to the asset of our interest, has been developed for that reason. The whole structure of the toolkit, which is aligned with the authors' methodological approach, is presented schematically in **Figure 3**.

#### *3.1.1 System characterization*

The first step is the system characterization. A whole Transportation System's network description is being conducted, also including its key elements, which are the bridges and the tunnels. Moreover, demand data (i.e., N° of vehicles/hour, traffic flow intensity, driving directions) are being collected from every available source and are being exploited as input data for the toolkit.

#### *3.1.2 Asset characterization*

In this phase, the asset characterization is focused on the network components (here bridges, tunnels, and road). The components' description is including the asset's

#### **Figure 3.**

*Schematic representation of the toolkit's design philosophy, aligning with the followed methodology's steps.*

#### *Resilience of Infrastructures and Systems to Multiple Hazardous Events: Application Cases… DOI: http://dx.doi.org/10.5772/intechopen.108548*

main properties, mainly the technical details and the design data of the structures (e.g., N° of bridges' piers, deck length). The site conditions (e.g., soil properties, slopes characteristics) are taken also into consideration as they are very important data for the behavior of the asset and the hazard characterization.

#### *3.1.3 Threat characterization*

The third step is the hazard definition and evaluation. For this step, every resource available in literature, web sources, or data shared from the infrastructure's managers are being exploited. And one of the novelties is that the toolkit can integrate satellite and terrestrial data in the analysis and the assessment of the hazards. This way, the desirable data-driven diagnostic framework is strengthened sufficiently from the accuracy of the data input. It follows a definition of relevant hazard scenarios and relevant hazard modeling (e.g., seismic and rainfall curves) in the area considered.

#### *3.1.4 Risk, vulnerability, and impact*

The before-mentioned step is necessary for the calculation of the fragility (**Figure 4**) and restoration functions, along with the vulnerability curves, toward the investigated hazard. The fragility functions can be derived from methodologies, which are found to the existing databases or literature, or from a more targeted and accurate analysis, from the Finite Element modeling. The followed impact analysis consists of the operativity loss, risk quantification, loss curves, and expected annual loss. In this step, it is important, especially for the System's operators, the calculation of the Expected Annual Loss (EAL), as part of the impact analysis. It provides the annual loss of the asset, as percentage of the repair cost, for a given hazard. These losses are generated by the repair costs applied to the asset after a possible hazard occurrence.

#### *3.1.5 Resilience assessment*

The integration of all this information is leading to the creation of the multiscenarios in which the Transportation System will be simulated, and the final resilience assessment will be conducted, under a specific framework with indicators and

**Figure 4.** *Typical fragility function (e.g., for earthquake).*

#### **Figure 5.**

*Typical results presented for a resilience assessment of an asset (e.g., bridge).*

different assumptions (i.e., deterministic or indeterministic approaches). The results are expressed in terms of operativity and time for the various limit states of the asset (**Figure 5**). The final goal of the resilience assessment is the creation of the necessary input for decision support tool, which is an instrument offered to Infrastructure managers about disruptive hazards impacting effects on their assets, and it enhances the overall operational phase of the Systems and Infrastructures.

#### **3.2 Energy infrastructure**

The application of the SecureGas was aiming more to a resilience design and management (**Figure 6**), rather than to a resilience assessment or risk management, of a gas transmission network. The methodology is adjusted respectively to the outer goal and this way, the desired adaptability is being justified in practice. The first three steps of the methodological approach are the same, and the subsequent upgrade of the overall resilient behavior was considered granted.

#### *3.2.1 System characterization*

In this initial step, the system characterization follows this of a typical gas network and plants, which means that special focus was given to the location (site characteristics), geo-politics, and climate. The first step was closely connected to the threat characterization, as the hazards were taken into consideration after the study of the relevance literature for the respective system characterization for a gas transmission network.

#### *3.2.2 Asset characterization*

Every technical and design details were collected in this phase, especially those referring to the safety and the security of the gas plants and networks operational phase (e.g., maximum gas pressure in the pipelines). Every step on the gas value chain was taken into consideration (production, storage, transmission, distribution), and the safety protocols were followed in detail.

*Resilience of Infrastructures and Systems to Multiple Hazardous Events: Application Cases… DOI: http://dx.doi.org/10.5772/intechopen.108548*

#### **Figure 6.**

*A typical flowchart of resilience management process, which is aligned with the proposed methodology, as it has incorporated its basic steps and the final resilience assessment/quantification, also including the risk management of an asset and the monitoring of the resilience behavior.*

#### *3.2.3 Threat characterization*

The threat characterization was conducted based on the literature, searching for the most frequently classified threats, and the user requirements, which have been set with by the end-users, in order to align better with their needs in the real-life circle of the assets' operational function [45]. Among them are the external interference or third-party activity (including political/geo-political interference), corrosion, construction defect and mechanical or material failure, natural hazards, operational error, and cyber-attacks [46].

#### *3.2.4 Resilience management*

The fourth and fifth steps here can be considered as the resilient design, which is shifting to the operational management of the Infrastructure, and not only the assessment phase. So, within SecureGas project, a toolkit based on High-Level Architecture [46] and the respective Conceptual Model followed was developed, aiming at the prevention, detection, response, and mitigation of combined physical and cyber threats to gas transmission grid network. Following this philosophy as an expansion of the proposed methodology, the resilience management of the asset (**Figure 7**) is enhancing and has become more robust toward any potential threat and hazards identified in the second step.

The main goal and the novelty of this toolkit is the convergence between physical and cyber threats or the so-called safety-security convergence. A central and undivided platform was designed, which covers the user requirements and where all the threats (cyber, natural, and man-made) to the gas transmission network or the plant can be addressed and recorded. The input data are derived from the sensors placed to the network and the plant, the UAV inspecting the facilities and the software for the cyber-protection of the System's operation.

#### *Critical Infrastructure – Modern Approach and New Developments*

**Figure 7.**

*Brief representation of the resilience management process across the life cycle of an infrastructure followed from RINA, where the proposed methodology is contributing to the design and evaluate and plan phase (source: RINA).*

This way, the surveillance and the control of the asset in the operational level are becoming more efficient, and the grid is enhancing its safety against multiple hazards. The real-time monitoring of the grid's condition is securing the high level of the situational awareness, and the early detection of disruptive event is leading to faster restoration of potential damage and a more targeted emergency management. The decision support system is based on the data acquisition and the threat evaluation, while the feature for the information sharing with the public is securing the safety of the communities.

#### **3.3 Meta-materials and energy infrastructures**

The notion of "meta-materials" refers to natural or artificial materials or structures, which exhibit extraordinary properties for inhibiting or conditioning wave propagation in all spatial directions over broad frequency bands [47], and this way, protecting the underlying structures toward dynamic-nature hazards. Due to the periodic structure of the meta-materials, the so-called band-gaps are being created, which are considered to be mitigation zones for specific frequency ranges of the transmitted waves (**Figure 8**). This way, the potential damage from dynamic-nature phenomena such as blast or seismic is mitigated, or in some cases, the structures are becoming isolated toward this hazard.

Meta-materials are relatively recent to the civil engineering design practices, but concepts based on this design philosophy for the protection of Energy Infrastructures have been already developed. Examples of them are the seismic protection of fuel storage tanks [48] and of nuclear plants [49] via the concept of a meta-foundation. Also, a meta-material concept for the blast protection of gas transmission pipelines has been proposed [50], and it is shown in **Figure 9**. It is worth noted that the already existing meta-material concepts such as the meta-concrete [51] or the meta-barriers [52] can be implemented for the increase of the resilience capacity of Energy Infrastructures and Transportation Systems, but the beneficial implementation of them has not yet been evaluated. Finally, meta-materials concepts can be exploited for the enhancing the resilience level of already constructed Energy Infrastructures, as they can be placed around the structure.

The comparative advantage of the meta-materials is the upgrade of the resilience capacity from the design phase of a civil engineering project. Although, there are

*Resilience of Infrastructures and Systems to Multiple Hazardous Events: Application Cases… DOI: http://dx.doi.org/10.5772/intechopen.108548*

#### **Figure 8.**

*A typical example of a band-gap, expressed in terms of frequency and length wavenumber, where the shaded regions are considered the mitigation zones for transmitting waves of these frequency ranges.*

#### **Figure 9.**

*The proposed meta-material concept for the protection of gas transmission pipelines, in a) exploded view and b) cross section [50]. The successive layers of the two different materials are leading to the creation of the band-gaps and the zones of energy mitigation.*

solutions of this philosophy (e.g., meta-barriers) that can be exploited in order to reinforce the behavior of the System or a specific Infrastructure toward specific hazards. For both cases, the methodological approach can still be implemented and lead to a comparative study of the resilience level between solutions, which are including or not the existence of meta-materials in their scope.

#### *3.3.1 System characterization*

Regarding the type of the Energy Infrastructures, the system characterization will follow the standard procedure of collecting the necessary data and information.

#### *3.3.2 Asset characterization*

Regarding the type of the Energy Infrastructures, the system characterization will follow the standard procedure of collecting the necessary data and information.

#### *3.3.3 Threat characterization*

In the third step, except for the standard procedure regarding the type of the Energy Infrastructure, there will be included also the necessary information for the respective design of a meta-material concept, which means the frequency spectra of the dynamic-nature hazard.

#### *3.3.4 Risk, vulnerability, and impact*

Toward the fourth and most crucial step, the calculation of the vulnerability and risk assessment for the specific threat will take place and the results for the solution based on the meta-materials will reveal the beneficial presence of these concepts to the respective damage mitigation and risk reduction.

#### *3.3.5 Resilience assessment*

The subsequent enhancement of the resilience capacity will be verified, in the fifth and last step, following the before-mentioned resiliency framework and metrics.

In this case, the methodological approach is contributing, especially via the comparative calculation of the vulnerabilities and the impact analyses, to the highlighting of these meta-material-based solutions for the scope of the Infrastructures' and Systems' resilience. Under a more general prism, a way is being paved for respective advanced methods of design, which are derived from the latest research achievements, to be transferred in the real-world projects, serving the goal of resilience.

#### **4. Future perspectives**

The further expansion of the current knowledge is crucial, and it is needed to be oriented in the future demands and landscape of Infrastructures and Systems. The proposed resilience methodology and the respective application cases that were presented are future-oriented, but their future exploitation is not limited to the so far produced results. For this reason, the authors are giving directions and are suggesting potential concepts, based on the investigated fields of interest.

#### **4.1 Transportation systems**

The methodological resilience assessment process, which was followed in the Transportation System's application, and the toolkit, which led to a decision support system, are needed to be expanded in other types of Infrastructures, such as these in the sector of Energy. Also, the current range of applications can include the resilience assessment of Transportation Systems during war or the so-called war resilience assessment. The authors' suggestions are being presented in **Table 1**.

#### **4.2 Energy infrastructures**

The Gas Energy Infrastructure's toolkit has spotlighted the significance of the convergence between physical and cyber security and the subsequent upgrade of the resilience capacity for the gas network grid. It is needed also to expand its feasibility for tackling hybrid threats and warfare. Furthermore, the whole function and the

*Resilience of Infrastructures and Systems to Multiple Hazardous Events: Application Cases… DOI: http://dx.doi.org/10.5772/intechopen.108548*


**Table 1.**

*Authors' suggestions for future exploitation of Transportation System's application.*


**Table 2.**

*Authors' suggestion for future exploitation of the Energy Infrastructure's application.*


**Table 3.**

*Authors' suggestion for future exploitation of meta-materials concepts for the resilience upgrade of Energy Infrastructures.*

capabilities of the specific solution (indicated by the proposed methodology) can inspire respective toolkits for other types of energy plants or transmission grids. The authors' suggestions are being presented in **Table 2**.

In the field of advanced materials' exploitation for the scope of the resilience enhancement, the future perspectives and the range of applications are more. Various meta-material concepts can be exploited for the purposes of upgrading the resilience profile of existing or new Energy Infrastructures. They can also be implemented to numerous types of Energy Infrastructures such as electricity plants or underwater pipeline grid, against various types of hazards such as blast or seismic. The authors' suggestions are being presented in **Table 3**.

#### **5. Conclusion**

The aim of this study was to demonstrate novel approaches for the enhancement of the resilience for the Infrastructures and Systems, via a specific methodological approach, which has been followed within three applications, mainly cases referred to Transport Systems and Energy Infrastructures. The promising and efficient methodological approach is clearly presented in its general structure and then was specified for every project. The application cases were chosen in order to spotlight the need for upgrade in the design philosophy for the resiliency planning and the robustness methods, regarding the current and future demands. A powerful toolkit is developed in order the methodological approach to be followed in the technical level, in scope of Transport Systems' resilience assessment. The convergence of safety and cyber-security is of high importance for the Infrastructures and Systems resilience management, and it is needed to be considered in every approach for the robustness of a resilience planning. Advanced technologies such as meta-materials can upgrade the resilience capacity of various projects (e.g., Energy Infrastructures) even from the design phase, and it paved a way in order the research-based technical solution to be integrated in resiliency frameworks. The authors have also described the future perspectives of the methodology in the studied sections and suggested specific concepts and directions for the further exploitation.

#### **Acknowledgements**

The authors acknowledge the financial support from the European Commission, as the INSPIRE project has received funding from the European Union's Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No. 813424, the SecureGas project has received funding from the European Union's Horizon 2020 research and innovation program under the grant agreement No. 833017, and the FORESEE project has received funding from the European Union's Horizon 2020 research and innovation program under the grant agreement No. 769373.

### **Conflict of interest**

The authors declare no conflict of interest.

#### **Notes**

Due to confidentiality reasons, it was not allowed to share specific results and values in some of the application presented. Whoever will be interested into more details about the applications can contact the author for further questions.

*Resilience of Infrastructures and Systems to Multiple Hazardous Events: Application Cases… DOI: http://dx.doi.org/10.5772/intechopen.108548*

#### **Author details**

Clemente Fuggini1 , Miltiadis Kontogeorgos1,2\*, Saimir Osmani1 and Fabio Bolletta1

1 Rina Consulting S.p.A., Rozzano, Italy

2 School of Civil Engineering, National Technical University of Athens, Athens, Greece

\*Address all correspondence to: miltiadis.kontogeorgos@rina.org

© 2022 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

### **References**

[1] Hallegatte S, Rentschler J, Julie R. Lifelines: The Resilient Infrastructure Opportunity. Washington: World Bank; 2019

[2] Neumann JE, Price J, Chinowsky P, Wright L, Ludwig L, Streeter R, et al. Climate change risks to US infrastructure: Impacts on roads, bridges, coastal development, and urban drainage. Climatic Change. 2015;**131**(1):97-109

[3] Shiklomanov N, Streletskiy AD, Suter L. Assessment of the cost of climate change impacts on critical infrastructure in the circumpolar Arctic. Polar Geography 2019;**42**(12):1-20

[4] Twerefou KD, Chinowsky P, Adjei-Mantey K, Strzepek LK. The economic impact of climate change on road infrastructure in Ghana. Sustainability. 2015;**7**(9):11949-11966

[5] Abeygunawardena VP, Knill Y, Foy P, Harrold T, Steele M, Tanner P, et al. Poverty and Climate Change: Reducing the Vulnerability of the Poor through Adaptation. Washington: World Bank; 2009

[6] Paris Agreement 2015 of United Nations. 2015. Available from: https:// unfccc.int/sites/default/files/english\_ paris\_agreement.pdf [Accessed: August 14, 2022]

[7] Sendai Framework for Disaster Risk Reduction 2015-2030. 2015. Available from: https://www.preventionweb.net/ files/44983\_sendaiframeworkchart.pdf [Accessed: August 14, 2022]

[8] Disaster Resilience Framework: Principles for Analyzing Federal Efforts to Facilitate and Promote Resilience to Natural Disasters [Internet]. 2019.

Available from: https://www.gao.gov/ assets/gao-20-100sp.pdf [Accessed: August 14, 2022]

[9] Resilience Framework of U.S. Department of Homeland Security. 2018. Available from: https://www.dhs.gov/ sites/default/files/publications/dhs\_ resilience\_framework\_july\_2018\_508.pdf [Accessed: August 14, 2022]

[10] Proposal for a DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL on the resilience of critical entities [Internet]. 2020. Available from: https://eurlex.europa.eu/legal-content/EN/ TXT/?uri=CELEX%3A52020PC0829 [Accessed: August 14, 2022]

[11] National Committee on Sustainable Development Goals. UAE and the 2030 agenda for Sustainable Development. Dubai: Federal Competitiveness & Statistics Authority Secretariat for UAE National Committee on SDGs; 2017

[12] Directive (EU) 2016/1148 of the European Parliament and of the Council of 6 July 2016 concerning measures for a high common level of security of network and information systems across the Union [Internet]. 2016. Available from: https:// eur-lex.europa.eu/eli/dir/2016/1148/oj [Accessed: August 14, 2022]

[13] Framework for Improving Critical Infrastructure Cybersecurity from National Institute of Standards and Technology [Internet]. 2018. Available from: https://nvlpubs.nist.gov/nistpubs/ CSWP/NIST.CSWP.04162018.pdf [Accessed: August 14, 2022]

[14] Didier M, Broccardo M, Esposito S, Stojadinovic B. A compositional demand/ supply framework to quantify the

*Resilience of Infrastructures and Systems to Multiple Hazardous Events: Application Cases… DOI: http://dx.doi.org/10.5772/intechopen.108548*

resilience of civil infrastructure systems (Re-CoDeS). Sustainable and Resilient Infrastructure. 2018;**3**(2):86-102

[15] Nan C, Sansavini G, Kröger W, Heinimann H. Quantitative methods for assessing the resilience of infrastructure systems. In: Probabilistic Safety Assessment and Management (PSAM '12). Honolulu. Hawaii; June 2014

[16] Ongkowijoyo CS, Doloi H. Riskbased resilience assessment model focusing on urban infrastructure system restoration. Procedia Engineering. 2018;**212**:1115-1122

[17] Kong J, Simonovic SP, Zhang C. Resilience assessment of interdependent infrastructure systems: A case study based on different response strategies. Sustainability. 2019;**11**(23):6552

[18] Liu X, Prodan I, Zio E. Resilience analysis of interconnected systems by a set-theoretic approach. In: Congres Lambda Mu 19. Dijon, France; 2014

[19] Blagojevic N, Didier M, Hefti F, Henken J, Stojadinovic B. Quantifying disaster resilience of a community with interdependent civil infrastructure systems. Structure and Infrastructure Engineering. 2022. Available from: https://www. tandfonline.com/action/showCitFor mats?doi=10.1080%2F15732479.2022. 2052912&area=0000000000000001. DOI: 10.1080/15732479.2022.2052912

[20] Storesund K, Reitan KN, Sjöström J, Rød B, Guay F, Almeida R, et al. Novel methodologies for analysing critical infrastructure resilience. In: Safety and Reliability – Safe Societies in a Changing World. Trondheim, Norway; 2014

[21] Shafieezadeh A, Burden I.V. Scenariobased resilience assessment framework for critical infrastructure systems: Case

study for seismic resilience of seaports. Reliability Engineering & System Safety. 2014; 132: 207-219

[22] Tokgoz BE, Gheorghe AV. Resilience quantification and its application to a residential building subject to hurricane winds. International Journal of Disaster Risk Science. 2013;**4**(3):105-114

[23] Leobons CM, Gouvêa Campos VB, Bandeira M. Assessing urban transportation systems resilience: A proposal of indicators. Transportation Research Procedia. 2019;**37**:322-329

[24] Gonçalves JPAL, Riberio P. Resilience of urban transportation systems. Concept, characteristics, and methods. Journal of Transport Geography. 2020;**85**(11):102727

[25] Khagani F, Jazizadeh F. mD-resilience: A multi-dimensional approach for resilience-based performance assessment in urban transportation. Sustainability. 2020;**12**(12):4879

[26] Wang J, Gao Y. A resilience assessment framework for urban transportation systems. International Journal of Production Research. 2020;**59**(2):1-16

[27] Enache FM, Al-Janabi D, Letia T. Conceiving of resilient railway systems. In: International Conference on Automation, Quality and Testing, Robotics (AQTR). Cluj-Napoca, Romania; 2020

[28] Regt de A, Siegel WA, Schraagen MJ. Toward quantifying metrics for railsystem resilience: Identification and analysis of performance weak resilience signals. Cognition Technology and Work. 2016;**18**(2):319-331

[29] Leviäkangas P, Aapaoja A. Resilience of transport infrastructure systems. CSID Journal of Infrastructure Development. 2015;**1**(1):80-90

[30] Do M, Jung H. Enhancing road network resilience by considering the performance loss and asset value. Sustainability. 2018;**10**(11):4188

[31] Cox A, Prager F, Rose A. Transportation security and the role of resilience: A Foundation for Operational Metrics. Transport Policy. 2011;**18**(2): 307-317

[32] Shafei B, Alipour A. Seismic resilience of transportation networks with deteriorating components. Journal of Structural Engineering. 2016;**142**(8):C4015015

[33] Testa AC, Furtado MN, Alipour A. Resilience of coastal transportation networks faced with extreme climatic events. Journal of the Transportation Research Board. 2015;**2532**(1)

[34] Valenzuela YB, Rosas SR, Mazari M, Risse M, Rodriguez-Nikl T. Resilience of road infrastructure in response to extreme weather events. In: International Conference on Sustainable Infrastructure. New York, USA; 2017

[35] van Dijk M. Tsunami Resiliency of Transport Systems [thesis]. Delft: Delft University of Technology; 2018

[36] Labaka L, Hernantes J, Sarriegi JM. Resilience framework for critical infrastructures: An empirical study in a nuclear plant. Reliability Engineering & System Safety. 2015;**141**:92-105

[37] Afgan N, Veziroglu A. Sustainable resilience of hydrogen energy system. International Journal of Hydrogen Energy. 2012;**37**:5461-5467

[38] Ouyang M, Dueñas-Osorio L. Multi-dimensional hurricane resilience assessment of electric power

systems. Journal of Structural Safety. 2012;**48**:15-24

[39] Sun L, Stojadinovic B, Sansavini G. Resilience evaluation framework for integrated civil infrastructurecommunity systems under seismic hazard. In: 16th World Conference on Earthquake. Santiago, Chile; 2017

[40] Souto L, Yip J, Wu WJ, Austgen B, Kutanoglu E, Hasenbein J, et al. Power system resilience to floods: Modeling, impact assessment, and mid-term mitigation strategies. International Journal of Electrical Power & Energy Systems. 2022;**135**(3):107545. DOI: 10.1016/j.ijepes.2021.107545

[41] Cotilla-Sanchez E, Ekisheva S, Papic M. A risk-based approach to assess the operational resilience of transmission grids. Applied Sciences. 2020;**10**(14):4761

[42] Maccioni M, Bragatto T, Cresta M, Cortesi F, Gatta F, Geri A, et al. Assessment and possible solution to increase resilience: Flooding threats in Terni distribution grid. Energies. 2019;**12**(4):4761

[43] Nguyen T, Wang S, Alhazmi M, Nazemi M, Estebsari A, Dehghanian P. Electric power grid resilience to cyber adversaries: State of the art. IEEE Access. 2020;**2020**:8

[44] About the FORESEE project [Internet]. 2018. Available from: https:// foreseeproject.eu/about-us/ [Accessed: August 14, 2022]

[45] Agrafioti E, Chalkidoy A, Papadakis G, Gazi A, Gkotsi I, Stritecku V, et al. The SecureGas Key Performance Indicators. In: Soldatos J, Praça I, Jovanovic A, editors. Cyber-Physical Threat Intelligence for Critical Infrastructures Security: Securing Critical Infrastructures in Air Transport, Water, Gas, Healthcare, Finance and Industry.

*Resilience of Infrastructures and Systems to Multiple Hazardous Events: Application Cases… DOI: http://dx.doi.org/10.5772/intechopen.108548*

Boston-Delft: Now Publishers Inc; 2021. pp. 342-360

[46] SecureGas deliverable 2.2. 2021

[47] About the INSPIRE project [Internet]. 2018. Available from: https:// itn-inspire.eu/about/ [Accessed: August 14, 2022]

[48] Bursi SO, Basone F, Wenzel M, Fossetti M. Finite locally resonant Metafoundations for the seismic protection of fuel storage tanks. Earthquake Engineering & Structural Dynamics. 2018;**48**(2):232-252. DOI: 10.1002/ eqe.3134

[49] Mo YL, Witarto W, Tang Y, Wang SJ, Yang CY, Chang KC, et al. Three-dimensional periodic materials as seismic base isolator for nuclear infrastructure. AIP Advances. 2019;**9**(4):045014

[50] Kontogeorgos M, Fuggini C. Meta-material layout for the protection of buried steel pipes against surface explosion. Geotechnics. 2022;**2**(2):427-440

[51] Pandolfi A, Ortiz M, Bricolla D. Experimental validation of Metaconcrete blast mitigation properties. Journal of Applied Mechanics. 2019;**84**:031001

[52] Palermo A, Krödel S, Marzani A, Daraio C. Engineered metabarrier as shield from seismic surface waves. Scientific Reports. 2016;**6**(1):39356

#### **Chapter 3**

## An Open-Data-Based Methodology for the Creation of a Graph of Critical Infrastructure Dependencies at an Urban Scale

*Antonio Di Pietro, Alessandro Calabrese, Antonio De Nicola, Daniele Ferneti, Luisa Franchina, Josè Martì and Tommaso Ruocco*

#### **Abstract**

This paper presents the MARIS (*Modeling infrAstructuRe dependencIes at an urban Scale*) methodology, allowing the generalization of one of the possible graphs modeling Critical Infrastructure (CI, hereafter) interdependencies at an urban scale starting from uncertain data. This leverages a set of known interdependencies at the system level, topological open data of local services and Points of Interest collected at an urban scale, and some heuristics. Indeed, interdependencies at an urban scale are usually not known to decision makers (e.g., CI operators, emergency planners) due to, for example, a lack of integration of knowledge held by different critical infrastructure operators and privacy restrictions. Here, these interdependencies are determined through geographic-based strategies. The resulting graph can be a valuable input to simulate emergency scenarios of CIs in the area of interest and, thus, plan proper countermeasures.

**Keywords:** interdependencies, open-data, critical infrastructures, graph, GIS

#### **1. Introduction**

Nowadays, Critical Infrastructures (CI) (roads and railways, electrical and telecommunication networks, gas and water pipelines, etc.), supplying primary services to citizens, are mutually connected as they provide their services not only to final users but also to other infrastructures. In this aspect, infrastructures are said to be dependent on each other and" interdependent." The concept of dependency stands on the fact that an infrastructure (e.g., the electrical network) provides service to another (e.g., the telecommunication network), and, thus, the latter is said to be dependent on the former. CI interdependencies can involve different abstraction layers. For instance, a general statement about the dependency between a railway and a power station involves a different layer related to a railway and a power station located in a given city.

For the sake of simplicity, in the following, we refer to the former as *system dependency* and to the latter as *urban dependency*.

Whereas repositories of system dependencies already exist1 , these are not sufficient to build urban dependencies due to several factors including lack of integration of knowledge held by different critical infrastructure operators and privacy restrictions. Furthermore, these data are constantly changing and difficult to collect because different stakeholders keep them. In order to build simulation models, the unavailability of information on real interdependencies is generally overcome using literature data or through survey data analysis addressed to critical infrastructure experts [1]. However, interdependency data would allow more reliable simulation for risk assessment and crisis management in case of natural events, such as earthquakes, or other events, such as cyber-attacks.

In this context, this paper aims to define a methodology to generate one of the possible graphs modeling CI interdependencies at an urban scale starting from uncertain data. The proposed methodology, named MARIS (Modeling infrAstructuRe dependencIes at an urban Scale), allows the discover of possible hidden dependencies using Geographical Information System (GIS) Open data of CIs and Points of Interest (e.g., the location of substations of an electric distribution network serving an urban area) and to apply proximity criteria to model dependencies when these are not known.

In addition, the paper presents the results of a survey on CIs conducted by several experts from Italy, aiming at quantifying the dependencies between critical infrastructures. This can be a valuable input for applying the MARIS methodology to produce realistic graphs of interdependencies in an urban context.

The rest of the paper is organized as it follows. Section 2 gives an overview of the present work on CI interdependencies. Section 3 presents the MARIS methodology. Section 4 presents a survey analysis of experts on the dependencies between Critical Infrastructures. Section 5 describes a case study related to Rome in Italy. Finally, Section 6 concludes the paper.

#### **2. Related work**

Modeling of CI interdependencies has been considered as a relevant problem since the seminal paper of [2].

In Chiara et al. [3], the authors proposed the Mixed Holistic Reductionist (MHR) methodology based on the interaction of three layers: (i) a holistic layer where CIs are seen as singular entities with defined boundaries and functional properties; (ii) a reductionist layer modeling the behavior of individual CI components; and (iii) a service layer that describes the functional relationships between components and the infrastructure at different levels of granularity.

An ontology design pattern to model CI interdependencies was proposed by Ref. [4] as part of the TERMINUS ontology. Crisis management [5] and risk assessment [6, 7] are two possible applications of ontologies modeling critical infrastructures [8].

In Rosato et al. [9], the authors focus on modeling cyber dependencies of a set of infrastructures in an urban context. In particular, a dependency matrix [10] was used to reveal the potential vulnerability of a given node to the unavailability, corruption, or disclosure of data from an interdependent node regardless of the current state of the shared data infrastructure.

<sup>1</sup> https://websites.fraunhofer.de/CIPedia/index.php/Interdependency.

*An Open-Data-Based Methodology for the Creation of a Graph of Critical Infrastructure… DOI: http://dx.doi.org/10.5772/intechopen.113045*

The work of Michel et al. [11] allowed us to define a dependency matrix generated based on the analysis of CI disruptions gathered from public media in the Netherlands from 2004 to 2010. In particular, a specific set of news sources containing impact keywords were acquired by Google News and further organized to save information including the affected CI sector and service, the initiating event (if any), and its dependency on another affected CI service.

In Franchina et al. [12], the authors present a methodology able to classify critical infrastructures starting from citizenship basic needs and foresee possible cascading effects.

In the MARIS methodology, we address all types of dependencies (logical, cyber, and physical) that are modeled by a specific system dependency layer and acquire open data of CIs and points of interest in order to create an urban dependency layer that can be valuable for further CI dependency analysis.

#### **3. The MARIS methodology**

#### **3.1 Overview**

Dependencies and interdependencies increase the vulnerability of infrastructures as they allow failures and perturbations to propagate from one system to another with the consequence that an infrastructure, which is not directly affected by some event that undermines its functioning, can be perturbed by the lack of functionality coming from another infrastructure which is turn hit by the event. Whereas from a conceptual side, the phenomenon is clear, on the practical side, dependencies and interdependencies are responsible for producing complex phenomena as, for instance, propagation time-scales can be quite different depending on the first perturbed infrastructure and on the infrastructure where perturbation will flow. For example, whereas perturbation on an electrical line may affect a very short time scale (seconds or less), an electrical perturbation to the traffic system or the water distribution network may take hours to produce sizeable effects. In that, if one would consider a" dependency matrix" where row and columns represent infrastructures and the i-j element their interaction, such a matrix is highly nonsymmetric (the ij element may be largely different from the ji one) as some infrastructure might perturb (severely and rapidly) other infrastructures which, in turn, have a very poor effect on the former when, in turn, perturbed. In some cases (i.e. when perturbation originates from the electrical system), the" coupling" strength is often very strong (i.e. many systems depend primarily on the electrical power), and the resulting reduction (or loss) of function established in a very short time scale. The presence of such strong coupling between infrastructures leads the system to be a" unique" system that cannot be treated and approximated as a system of independent (or nearly independent) infrastructure but as a" system of systems" that cannot be linearized A further element which leads the interdependency problem even more complex to be treated originates from the possibility of closed loops involving more than a couple of infrastructures. Perturbation from a first infrastructure might flow on several infrastructures before returning to the first one, providing negative feedback.

All these issues lead to an operational approach to interdependency that is extremely complex due primarily to an incomplete description of system's dependencies, time scales, and latencies. All dependency and interdependency data can be

achieved through direct and indirect methods [12], but a complete theory of interdependency is still lacking.

Despite all that, it is possible to empirically deal with the problem of describing a number of dependent and interdependent systems with the aim of providing some type of decision support systems (DSS) enable to support decision makers (e.g., CI and Civil Protection managers) in the risk management process.

The MARIS methodology makes it possible to transform the layer of interdependencies known at the system level into a layer of interdependencies at the level of the physical components that lie in an urban context, as shown in **Figure 1**.

In the following, first, we give some definitions of the main concepts involved in the MARIS methodology, and then, we give an overview of its main steps.

#### **3.2 Definitions**

The MARIS methodology is based on seven fundamental concepts pertaining to the two above-mentioned layers. The ones dealing with the system dependency layer are the following:


Those pertaining to the urban infrastructure layer are:

**Figure 1.** *Georeferenced dependency upscaling.*

*An Open-Data-Based Methodology for the Creation of a Graph of Critical Infrastructure… DOI: http://dx.doi.org/10.5772/intechopen.113045*


#### **3.3 Main steps**

**Figure 2** describes the main steps of the MARIS methodology. First, dependencies at system scale are collected through a survey eliciting knowledge on the dependencies between critical infrastructures. As mentioned, these dependencies concern subsystems. Information on entities are retrieved from a GIS (Geographical Information System) such as OpenStreetMap (OSM). Then, similar to the approach presented in Ref. [14], subsystems are matched with entity types. The resulting pairs are included in an annotation table. Finally, all the above information, i.e., subsystems dependencies, entities, and the annotation table, are used to infer urban dependencies and create an interdependency graph.

#### **4. Survey analysis on the dependencies between critical infrastructures**

In this section, we present the results of a survey on Cis conducted with experts from Italy coming from various economic and social sectors.

This activity, in addition to providing a set of real interdependencies useful for the application of the MARIS methodology, was used as input to the DOMINO simulation model [12], developed by Tesseract Srl company, aiming at studying and quantifying the consequences of negative, unexpected, and disruptive events to the Supply and Value Chain systems of a country. The purpose of this model is to reproduce the impacts that would occur on such systems if relevant functions of the Chains were disrupted by natural or malicious events. This DOMINO approach leverages

**Figure 2.** *Main steps of the MARIS methodology.*

the concept of an *item* (as defined in Section 3.1) based on the NACE2 and ATECO3 classifications. The survey was conducted by involving industry experts to detail the direct consequences produced by their own organizations due to the loss (or degradation) of a specific item. In particular, for each of the 117 items considered (**Table A1** in the Appendix section), a set of the impacted items was collected. In addition, data regarding the duration of the propagation (in terms of hours, days, months, and years) were also collected in order to perform impact analysis through the DOMINO model.

The overall model will thus consider all the interdependencies between the various sectors (including Cis) that make up the Country Supply Chain System, based on the data reported by the various experts.

In the following, two examples of dependency trees are discussed. **Figures 3** and **4** show the direct dependencies of the *Drinking water* and *Bank* items respectively. At the top of each tree a timeline is displayed to show the "falling time" of each ITEM, starting from the "time zero," corresponding to the root node.

In the first dependency tree, it can be noted that, following the disruption of the *Drinking Water* item, two items, i.e., *Education* and *Research,* are impacted in the range of only 4 hours. On day 1, 45 items are impacted. Most of the dependencies represented are directly linked to the root node, but few of them are activated due to second (or higher) order effects (i.e., Maintenance services, Agriculture, and products).

In the second dependency tree, a few items are impacted in the order of a few minutes. However, most of the items are impacted after 3 days from the initiating failure.

**Figure 3.** *DOMINO model from tesseract Srl: Drinking water dependency tree.*

<sup>2</sup> https://www.en.wikipedia.org/wiki/Statistical\_Classification\_of\_Economic\_Activities\_in\_the\_European\_ Community.

<sup>3</sup> https://www.istat.it/it/archivio/17888.

*An Open-Data-Based Methodology for the Creation of a Graph of Critical Infrastructure… DOI: http://dx.doi.org/10.5772/intechopen.113045*

#### **Figure 4.**

*DOMINO model from tesseract Srl: Bank dependency tree (items with impact higher than 30 days have been removed for the sake of space).*

#### **5. Case study**

The case study concerns an area of the city of Rome characterized by a high concentration of various business activities and power centers, as well as the presence of CIs, which are essential to maintaining vital societal functions. As shown in **Table 1**, the mentioned area includes = 10 systems and 11 subsystems.

#### **5.1 System dependency layer**

In this step, we imported the subset of subsystem dependencies that were collected by means of the survey described in Section 4. It is worth noting that these dependencies expand those more generic related to the system level, which are addressed, for instance, in Rinaldi et al. [2]. **Figure 5** shows an example of how system dependencies are extended to subsystem dependencies.

#### **5.2 Territorial open-data acquisition**

The development of the local dataset in the case study is based on the integration of open data collected and provided by OSM. The aim of this platform is to provide free geospatial information of world features. OSM uses open tagging mechanism to add meaning to geographic objects, so that any OSM user can add a new tag to them. For instance, some keys can be used to classify OSM entities into classes (e.g., highway, building, and amenity) while other keys play the role of attributes (e.g., name and maxspeed). OSM also allows to set attribute values, while values for classes are used to classify class members into categories (e.g., residential, hotel, monument).

#### *Critical Infrastructure – Modern Approach and New Developments*


#### **Table 1.**

*Dataset of infrastructure types in this study, categorized under 11 subsystems and 10 systems in the area of interest.*

#### **Figure 5.**

*(a) Example of system dependency layer; (b) example of subsystem dependency layer obtained from open data in the area of interest; (c) a fragment of the entity layer.*

*An Open-Data-Based Methodology for the Creation of a Graph of Critical Infrastructure… DOI: http://dx.doi.org/10.5772/intechopen.113045*

In order to facilitate the understanding of OSM data and reduce the number of incorrect entity-classes associations [15], we used the Taginfo platform4 to gather tag statistics (e.g., providing the tags are actually in the database, the number of users choosing those tags) and to semantically enrich the entities. For example, when analyzing the railway subsystem, by setting the tag *railway* = *station* in Taginfo, we were able to select the railway station entities located in the area of interest and discard those combinations (e.g., railway = crossing, railway = radio) that were associated with different concepts.

Then, we used *Overpass turbo*<sup>5</sup> , a web-based data mining tool for OSM, to acquire the OSM data of interest representing entity type occurrences of **Table A1** (e.g., university buildings and water towers) according to a geoJSON data format.

Therefore, the GIS open data acquired for CIs and Points of Interest were classified according to the specific subsystem.

#### **5.3 Urban infrastructure layer**

Regarding the dependencies to be associated to the entity layer, we considered those at the subsystem level and scaled them to the entity level. In other words, given that, according to the system dependency layer, the health system depends on the energy system, when scaling to the entity layer, we adopted the criterion that the specific hospital X depends on the energy supplied by the electrical substation Y (in particular, the one closest to the hospital).

**Figure 5** shows the result of the application of the MARIS workflow, i.e., a fragment of the dependency at an urban level graph. It can be noticed that a geographical proximity criterion was applied to set up dependencies between CI components.

#### **6. Conclusion**

The presented MARIS methodology allows the generation of a graph at urban scale that can capture the known dependencies of CIs and Points of interest and model those dependencies that are unknown (because of partial, uncertain, or sensitive data) through the application of proximity criteria.

The use of open software and data (Taginfo, Overpass turbo, OSM) and the knowledge of CI interdependencies data at the system level were used to estimate the interdependencies at an urban level and to produce a realistic graph of interdependencies.

The methodology was applied to the city of Rome to create an interdependency graph characterized by 6025 entities representing real CIs and Points of interest. Future developments will concern the application of dynamic simulation models to the interdependency graph, obtained through the MARIS methodology, in order to reproduce the impacts that would occur on such systems when relevant functions of the supply chain of an area of interest were disrupted by natural or anthropic events.

<sup>4</sup> http://taginfo.openstreetmap.org.

<sup>5</sup> Overpass-turbo.eu.

### **Appendix A**


*List of items considered in the DOMINO model*

*An Open-Data-Based Methodology for the Creation of a Graph of Critical Infrastructure… DOI: http://dx.doi.org/10.5772/intechopen.113045*



**Table A1.**

*List of items.*

*An Open-Data-Based Methodology for the Creation of a Graph of Critical Infrastructure… DOI: http://dx.doi.org/10.5772/intechopen.113045*

#### **Author details**

Antonio Di Pietro1 \*, Alessandro Calabrese2 , Antonio De Nicola1 , Daniele Ferneti3 , Luisa Franchina2 , Josè Martì4 and Tommaso Ruocco2

1 ENEA, Casaccia Research Centre, Rome, Italy

2 Tesseract srl, Rome, Italy

3 Sapienza University of Rome, Roma, RM, Italy

4 Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, B.C., Canada

\*Address all correspondence to: antonio.dipietro@enea.it

© 2023 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

### **References**

[1] Laugé A, Hernantes J, Sarriegi JM. Critical infrastructure dependencies: A holistic, dynamic and quantitative approach. International Journal of Critical Infrastructure Protection. 2015;**8**:16-23. DOI: 10.1016/j. ijcip.2014.12.004. Available from: https:// www.sciencedirect.com/science/article/ pii/S187454821400081X

[2] Rinaldi SM, Peerenboom JP, Kelly TK. Identifying, understanding, and analyzing critical infrastructure interdependencies. IEEE Control Systems Magazine. 2001;**21**(6):11-25. DOI: 10.1109/37.969131

[3] Foglietta C, Panzieri S, Pascucci F. Algorithms and tools for risk/impact evaluation in critical infrastructures. International Journal of Critical Infrastructures. 2015;**565**:227-238. DOI: 10.1007/978-3-662-44160-2\_8

[4] Coletti A, De Nicola A, Vicoli G, Villani ML. Semantic modeling of cascading risks in interoperable socio-technical systems. In: Popplewell K, Thoben K-D, Knothe T, Poler R, editors. Enterprise Interoperability viii. Cham: Springer International Publishing; 2019. pp. 119-129

[5] Nicola D, Antonio MM, Villani ML. Creative design of emergency management scenarios driven by semantics: An application to smart cities. Information Systems. 2019;**81**:21-48. DOI: 10.1016/j.is.2018.10.005. Available from: https://www.sciencedirect.com/ science/article/pii/S0306437918304277

[6] Coletti A, De Nicola A, Villani ML. Building climate change into risk assessments. Natural Hazards. 2016;**84**(2):1307-1325. DOI: 10.1007/ s11069-016-2487-6

[7] Coletti A, De Nicola A, Di Pietro A, La Porta L, Pollino M, Rosato V, et al. A comprehensive system for semantic spatiotemporal assessment of risk in urban areas. Journal of Contingencies and Crisis Management. 2020;**28**(3):178-193. DOI: 10.1111/1468-5973.12309. Available from: https://onlinelibrary.wiley.com

[8] Antonio DN, Villani ML. Smart city ontologies and their applications: A systematic literature review. Sustainability. 2021;**13**(10):20-60. DOI: 10.3390/ su13105578. Available from: https://www. mdpi.com/2071-1050/13/10/5578

[9] Rosato V, Di Pietro A, Kotzanikolaou P, Stergiopoulos G, Smedile G. Integrating resilience in time-based dependency analysis: A large-scale case study for urban critical infrastructures. In: Rosato V, Di Pietro A, editors. Issues on Risk Analysis for Critical Infrastructure Protection, Chap. 5. Rijeka: IntechOpen; 2021. DOI: 10.5772/intechopen.97809

[10] Macaulay T. Critical Infrastructure: Understanding its Component Parts, Vulnerabilities, Operating Risks, and Interdependencies. Taylor & Francis; 2008

[11] Eeten M, Nieuwenhuijs A, Luiijf E, Klaver M, Cruz E. The state and the threat of cascading failure across critical infrastructures: The implications of empirical evidence from media incident reports. Public Administration. 2011;**89**:381. DOI: 10.1111/j.1467-9299.2011.01926.x

[12] Franchina L, Carbonelli M, Crisci M, Gratta L, Perucchini D. An effective approach for cascading effects prevision in critical infrastructures. Studies in Computational Intelligence. 2011a;**7**: 73-90. DOI: 10.1504/IJCIS.2011.038958

*An Open-Data-Based Methodology for the Creation of a Graph of Critical Infrastructure… DOI: http://dx.doi.org/10.5772/intechopen.113045*

[13] Nirandjan S, Koks EE, Ward PJ, Aerts JCJH. A spatially-explicit harmonized global dataset of critical infrastructure. Scientific Data. 2022;**9**(1):150. DOI: 10.1038/ s41597-022-01218-4

[14] Minardi R, Villani ML, De Nicola A. Semantic reasoning for geolocalized assessment of crime risk in smart cities. Smart Cities. 2023;**6**(1):179-195. DOI: 10.3390/smartcities6010010. Available from: https://www.mdpi. com/2624-6511/6/1/10

[15] Ali AL, Schmid F. Data quality assurance for volunteered geographic information. In: Duckham M, Pebesma E, Stewart K, Frank AU, editors. Geographic Information Science. Cham: Springer International Publishing; 2014. pp. 126-141

#### **Chapter 4**

## Science Gateways and AI/ML: How Can Gateway Concepts and Solutions Meet the Needs in Data Science?

*Sandra Gesing, Marlon Pierce, Suresh Marru, Michael Zentner, Kathryn Huff, Shannon Bradley, Sean B. Cleveland, Steven R. Brandt, Rajiv Ramnath, Kerk Kee, Maytal Dahan, Braulio M. Villegas Martínez, Wilmer Contreras Sepulveda and José J. Sánchez Mondragón*

#### **Abstract**

Science gateways are a crucial component of critical infrastructure as they provide the means for users to focus on their topics and methods instead of the technical details of the infrastructure. They are defined as end-to-end solutions for accessing data, software, computing services, sensors, and equipment specific to the needs of a science or engineering discipline and their goal is to hide the complexity of the underlying infrastructure. Science gateways are often called Virtual Research Environments in Europe and Virtual Labs in Australasia; we consider these two terms to be synonymous with science gateways. Over the past decade, artificial intelligence (AI) and machine learning (ML) have found applications in many different fields in private industry, and private industry has reaped the benefits. Likewise, in the academic realm, large-scale data science applications have also learned to apply public high-performance computing resources to make use of this technology. However, academic and research science gateways have yet to fully adopt the tools of AI. There is an opportunity in the gateways space, both to increase the visibility and accessibility to AI/ML applications and to enable researchers and developers to advance the field of science gateway cyberinfrastructure itself. Harnessing AI/ML is recognized as a high priority by the science gateway community. It is, therefore, critical for the next generation of science gateways to adapt to support the AI/ML that is already transforming many scientific fields. The goal is to increase collaborations between the two fields and to ensure that gateway services are used and are valuable to the AI/ML community. This chapter presents state-of-the-art examples and areas of opportunity for the science gateways community to pursue in relation to AI/ML and some vision of where these new capabilities might impact science gateways and support scientific research.

**Keywords:** science gateways, virtual research environments, artificial intelligence, machine learning, collaboration

#### **1. Introduction**

Science gateways are end-to-end solutions for accessing data, software, computing services, and equipment specific to the needs of a science or engineering discipline. The goal of science gateways is to hide the complexity of the underlying research infrastructure and to enable scientists and educators to focus on their research and teaching—science gateways form one of the building blocks for critical infrastructure in research. Science gateways are often called Virtual Research Environments (VREs) in Europe and Virtual Labs (VLs) in Australasia [1]; we consider these two terms to be synonymous with science gateways. While quite a few research domains such as the life sciences, chemistry, and geospatial sciences have adapted the use of science gateways, there is still the need for a larger uptake in those domains and for broadening participation to further domains and user groups such as high-school students [2].

Some artificial intelligence/machine learning (AI/ML) research is supported by science gateways. Usually, this in the form of "software-as-a-service" for developed and trained AI/ML applications, for example, AI4Mars [3] on Zooniverse [4] that offers science gateways as a service for citizen sciences (see **Figure 1**).

Even though AI/ML and science gateways are both well-anchored in the highperformance computing (HPC) community, the field of science gateways still has low visibility to AI/ML application developers and users. A reason for the small degree of overlap of the AI community with the science gateway community includes that the AI/ML community has focused more on developer tools and languages than on intuitive and graphical interfaces. The trajectory of new concepts in academia is often first to develop effective methods, second to increase their efficiency and then, finally, to open them up to a wider community via considering usability. In the case of AI/ML it means to enhance the visibility via a set of capabilities that support AI/ML development and improve the uptake of science gateways in the AI/ML community. Furthermore, there is a need to advance the field of science gateway cyberinfrastructure itself.

#### **Figure 1.**

*The project AI4Mars uses citizen sciences to teach Mars rovers how to classify martian terrain.*

*Science Gateways and AI/ML: How Can Gateway Concepts and Solutions Meet the Needs in Data… DOI: http://dx.doi.org/10.5772/intechopen.110144*

These topics are critical for the next generation of science gateways and its community as AI/ML is already transforming many scientific fields.

One goal of science gateways is to facilitate collaborations between highly technical practitioners and less computer-savvy researchers, and thereby expand the reach and impact of science. In this case, it means to make AI/ML services accessible to the community. This chapter is part of the work to investigate the target groups in AI/ML research and to identify the opportunities and activities needed to achieve this goal and contribute significantly to critical infrastructure in research, teaching, and beyond.

For promoting collaborations between the science gateway community and the AI/ML community, it is important to target the three major groups: academic communities, funding agencies, and industry. Important actors in the science gateway community for increasing the collaboration are providers and developers of mature science gateway frameworks such as HUBzero [5], Apache Airavata [6], and Tapis [7]. Thus, in the area of academic communities, they can organize special tracks at Gateways conferences [8], which allow more conversations and exchanges of ideas in various science gateway communities including users of science gateways. Another possibility for widening the outreach is to hold webinars, panels, birds-of-a-feather sessions, and similar outreach efforts at conferences such as Practice & Experience in Advanced Research Computing (PEARC) [9], eScience [10], and Supercomputing [11] to promote the use of science gateways to research computing and interdisciplinary research communities, including AI/ML experts attracted by these conferences.

Another way to reach AI researchers is at their domain-specific conferences. Presentations at appropriate AI conferences would raise awareness of the concept of science gateways and would elucidate the opportunities to address pain points regarding usability and integration with complex computing and data infrastructures, for example, the International Conference on Machine Learning, Optimization and Data Science [12]. Papers could be "crowd-sourced" as previous papers to International Workshop on Science Gateways (IWSG) [13] or this manuscript.

Funding agencies and funded institutes/large projects around AI form another important target group for gathering requirements on usability and accessibility of AI/ML methods.

Examples of outreach to these groups include the promotion of science gateways to AI/ML award winners, especially at the various AI institutes [14] funded by National Science Foundation (NSF), and contacting cognizant program officers of AI/ML research-supporting programs to discuss roles for science gateways. While science gateways are well known in NSF directorates such as Office of Advanced Cyberinfrastructure (OAC) [15], program officers in other NSF directorates or other federal agencies might be less familiar with or not yet know about science gateways. In the 2019 update to the NSF National Artificial Intelligence Research and Development Strategic Plan [16], one of the major topics was the recognition of the gap between the capabilities of AI algorithms and the usability of AI systems by humans. The report states "Human-aware intelligent systems are needed that can interact intuitively with users and enable seamless machine–human collaborations." This is exactly the goal of science gateways and the fact that they are not mentioned as a potential solution emphasizes the point that science gateways are not yet well known in the AI community.

As a large producer and user of AI/ML concepts and technologies, industry is an important target group, especially projects that foster collaboration of funding agencies and companies, that is, a collaboration between Amazon and NSF that funds ten

#### **Figure 2.**

*The Permafrost Discovery Gateway uses AI to provide access to pan-Arctic permafrost knowledge and information about the globe regarding pan-Arctic change.*

research projects on fairness of AI [17]. The uptake of science gateways in industry is one of the goals some science gateway providers pursue for widening their community. Such adoption is also a measure of the sustainability achieved by science gateway frameworks. Meaningful connections include to introduce science gateways to technology providers, especially those in the space of cloud services such as Omnibond Systems who are already part of the science gateways community.

While there is interest especially from science gateway providers to form collaborations with academic communities, industry, and funding agencies, it is important to carefully select content to be presented to the target audiences. In order to achieve a wide outreach, we aim at answering an overarching question: how can Gateway concepts and solutions meet the needs of data science? In order to answer this question, the chapter is laid out with the following sections: first, we provide a brief background on AI, ML, and data-intensive computing in general. Second, we explain the terminologies, especially the difference between AI, ML, and data-intensive computing. Third, we explain what science gateways can do and describe their general capabilities. In this section, we also provide example gateways across different disciplines. Fourth, we discuss several science gateway opportunities for AI/ML research. Finally, we wrap up the chapter with the future outlook (**Figure 2**).

#### **2. Background**

Since its introduction [18], interest in AI has gone through several peaks and troughs. Generally, peaks have been driven by algorithmic progress coupled with the availability of appropriate computing resources and the troughs by their lack. This in particular has been true of brute force algorithms and shortcut modifications to those algorithms that prune brute force exploration. Contemporary with the notion of AI was the publication of the foundations of neural networks [19]. Although first appearing at a similar point in time, neural networks only began receiving widespread attention as a means for ML when software libraries and products became available to allow nonexperts to test the application of neural networks within their own research

*Science Gateways and AI/ML: How Can Gateway Concepts and Solutions Meet the Needs in Data… DOI: http://dx.doi.org/10.5772/intechopen.110144*

domains, especially during the 1980s. After a great wave of progress, limitations of existing neural network topologies and training algorithms were realized, spawning decades of research into more effective ways to represent and compute upon neural networks.

One of the confounding factors for early progress on ML was the need for ML algorithms to be fed by large amounts of training data. In many domains, such training sets did not exist. However, one of the significant drivers of progress in this area has been the emergence of the Internet. The past three decades has seen the creation of massive training sets in the form of user actions on the Internet, immense corpi of content authored by people participating in the Internet, large libraries of images and video becoming widely available, and so forth. Another key force in creating training sets has been the advancement of massively parallel supercomputer resources (e.g., XSEDE [20], OSG [21]) that can compute large amounts of data using physics-based models, which subsequently can be used to produce ML-based approximate models to rapidly compute these same outputs. These decades have also created a paired interest in ML involving both scientific and commercial interests. Although this pairing creates a variety of potential ethical issues, it has driven progress in this field at a rate faster than most fields that lack this symbiosis. It has even seen commercial organizations making ML algorithms available to the public (e.g., see Abadi et al. [22]).

Based on this history, today AI, ML, and (more generally) data-intensive computing have been identified as high priorities for federally funded research. In both 2016 [23] and 2019 [16, 24], reports by the National Science & Technology Council, spanning two different presidential administrations. In response, comprehensive programs and funding priorities have been put forward by federal agencies including the NSF [25], DARPA [26], NIH [27], and DOE [28]. AI/ML methods are also high priorities for mission-driven science agencies such as National Aeronautics and Space Administration (NASA). National Institute and Standards and Technology (NIST) is leading efforts related to safety and benchmarking of AI/ML applications. Many government agencies are releasing large, curated data sets that can be used for training.

The applications of AI/ML research are scientifically promising, of strategic importance to national defense, and important for economic competitiveness. Research utilizing AI/ML is expanding in multiple specific domains, including big data and high-energy physics, astronomy, animal husbandry, agriculture (Ag), food security, climate change, and city infrastructure. The AI100 Project [29] and the Computing Community Consortium [30], as well as the National Science & Technology Council Reports, provide long-range overviews of AI/ML research challenges and opportunities, including their likely impacts on society as a whole.

Science gateways are widely known to bring advanced scientific capabilities to researchers in the form of data sets, HPC resources, and instruments such that researchers do not need to be experts in accessing those resources [31]. Today, we face another evolution in AI/ML, where it can be further democratized and advanced through the use of science gateways. This overview examines the opportunities for integrating AI/ML research with science gateway cyberinfrastructure, based on the extensive background information surveyed above.

#### **3. Terminology**

Following Stone et al. [30], we will distinguish AI, ML, and data-intensive computing as follows:


This overview focuses on the requirements of ML methods that are being applied to a wide range of scientific data in diverse scientific fields in support of scientific research. ML holds the promise for scalably extracting information and knowledge (including scientific insights) from the large amounts of data generated by both experiments at all scales as well as scientific simulations, and fits well with the capabilities of science gateways today.

#### **4. Science gateway infrastructure**

Science gateways in general are noted for their ability to provide the following capabilities to support scientific research:


In other words, we can consider science gateways as cyberinfrastructure environments to support Findable, Accessible, Interoperable, and Reusable (FAIR) research [32]. Many of the concerns and opportunities identified in [16, 23, 24] for the use of AI/ML research are FAIR challenges. The FAIR principles have created significant momentum in the research community recognizing a need to improve the quality of research by establishing common standards. However, bridging the principles with the research infrastructure remains a challenging task due to its diversity and domain*Science Gateways and AI/ML: How Can Gateway Concepts and Solutions Meet the Needs in Data… DOI: http://dx.doi.org/10.5772/intechopen.110144*

specific nature of tasks. Science gateways provide an excellent opportunity to achieve FAIRness of research data and software.

#### **4.1 Use cases in artificial intelligence**

#### *4.1.1 Physics*

International collaboration is often basic for our scientific development, and these efforts are at the core of its infrastructure. Researchers are partnering with NSF and DOE to study how AI Frameworks can be leveraged in physics research. One such project, ML and FPGA Computing for Real-Time Application in Big-Data Physics Experiments as a science driver investigates the creation of a FAIR framework for AI [33]. Example is a project focusing on Inspired Artificial Intelligence in High-Energy Physics which builds on the successes of the last years with the Large Hadron Collider (LHC) and the combination of the Laser Interferometer Gravitational—wave Observatory (LIGO) and the Large Synoptic Survey Telescope (LSST) for Multi-Messenger Astrophysics by making artificial AI models and data more accessible and reusable with the goal to accelerate research and outperform current approaches [34]. Broad international collaborations as the Event Horizon Telescope (EHT) are excellent opportunities to introduce science gateways. This single-event global array of eight ground-based radio telescopes aimed at obtaining the image a black hole and its shadow could become a n international mainstay [35].

#### *4.1.2 Photonics and quantum optics*

AI is already taking part in the development of future technologies, as is the case of the quantum technologies. In this case, the integration of two communities, AI and photonics, have become complementary during the last decades, where ML protocols are being matched to photonic platforms giving rise to photonic neural network architecture [36]. The applications of AI, especially neural networks and ML in the field of quantum optics have also become prevalent for experimental setups used for classification and identification of light sources and quantum states by using these two approaches [37–39]. Such examples with computational properties have lowcomplexity and low-cost implementations promising quantum architecture that could apply underlying cyberinfrastructure enabling users to create their own workflows to run simulations codes. Those applications designed by the quantum community would generate tools with unprecedented capabilities available to researchers unfamiliar with the world of quantum optics and photonics.

#### *4.1.3 Astronomy*

The goal of the AGNet [40] project is to leverage AI to develop a novel interdisciplinary approach combining astronomy big data with ML tools to build a deep learning algorithm to estimate the masses of super-massive black holes. Measuring the masses *via* traditional methods is very expensive and such a new algorithm could transform the field of cosmology.

#### *4.1.4 Agriculture*

The connection between AI and Ag seems obvious. One applies big-data analytics, ML, and deep learning algorithms in geospatial information systems (GIS), satellite

data, lidar information, sensor data, and other tools and technologies to improve crops [41]. However, the connection is not as clear when discussing animal husbandry. An AI tool Project [42], Solving Dairy Cattle Genetic Improvement Challenges using Deep Learning, will use AI "to identify cattle that have the highest genetic potential for milk production and health status and make simplistic assumptions about the relationship between phenotypes and genotypes." Another interesting connection between AI and Ag is in the area of Food Security [43]. The Alan Turing Institute [18] is using ML and AI to leverage data and models of plant development, plant pathology, crop yields, and climate science to form a cohesive national crop modeling framework for the United Kingdom.

#### *4.1.5 Climate change*

The impact of climate change on the planet is much discussed in the news, but the ability to understand the true influence is limited by the ability to quantify how multiple factors work together to impact this change. The Permafrost Discovery Gateway [44] is using AI to assist with the management of ingesting large amounts of remote sensing data into machine and deep learning models which will ultimately provide "access to pan-Arctic permafrost knowledge, which can immediately inform the economy, security, and resilience of the Nation, the Arctic region, and the globe with respect to pan-Arctic change."

#### *4.1.6 Urban planning*

What do roads, sidewalks, parks, access to food and medical care, and even tree canopies providing shade on roads and sidewalks have to do with AI? Many studies have been done on food deserts and transit deserts, but by leveraging AI, researchers are able to look at all different types of neighborhood-scale infrastructure [45] (see **Figure 3**). By identifying infrastructure deserts, communities' ability to deal with them will be strengthened in the long term, particularly in low-income communities.

**Figure 3.** *The figure shows the different infrastructure deficiencies dependent on income in neighborhoods in Dallas.*

*Science Gateways and AI/ML: How Can Gateway Concepts and Solutions Meet the Needs in Data… DOI: http://dx.doi.org/10.5772/intechopen.110144*

#### *4.1.7 Biology*

Concerned with the detection of Regulatory Elements using GRO-seq data, the dREG science gateway [46] identifies the location of DNA sequence regions known as transcript regulation elements including promoters and enhancers—the critical components of the genetic regulatory programs of all organisms. The dREG computational code itself uses a support vector machines-based model trained by large-scale data. This science gateway democratizes the use of these sophisticated ML techniques to a wider community. The gateway interfaces with XSEDE compute infrastructures for seamlessly enabling access to compute intensive training and prediction phases. On the front-end of these ML models, user-friendly data visualization interfaces enable a wider community to interact with bigWig data, dREG signal predictions, and genome coordinates of peaks of transcriptions.

#### *4.1.8 Humanities*

Snow Vision uses image classification methods to identify pottery sherds created by Native Americans of the US Southeast. The Snow Vision science gateway [47] enables the humanities community to utilize ML-based matching algorithms to compare user uploaded sherd images to identify the original stamped designs from which their fragments descend. The gateway makes available the matching algorithms implemented by a deep-learning Point Cloud Library (PCL) for generating depth maps from 3D sherd image files and Caffe [48] deep learning toolkit.

#### **5. Science gateways, quantum computing, and artificial intelligence**

New paradigms in computing open up novel areas for exploring the potential of science gateways and AI/ML. A future direction is the emerging paradigm of quantum computing which uses the fundamental properties of quantum mechanics, such as superposition, entanglement, and interference. This domain strongly differs from classical computing, where one data qubit is equivalent to two classical bits of information. This feature is a promising solution for higher computational power within shorter calculation time, which is very useful for artificial neural networks and ML. At this stage, the latest frontier of computation relies on the hybrid development of these two areas in quantum computing and how both could support each other in the evolution of classification and clustering of big classical-to-quantum data. While the topic of AI and ML is well established in computer sciences, both are quite novel approaches on the science and technology side and not fully adapted yet. We analyzed where ML has developed better among those different fields and where it has moved to Quantum computation, more specifically, in the Quantum Machine Learning (QML) frontier. We have used Scopus data to show (after a normalization) in which of the sciences have become more active. A Scopus search on ML will provide us with the following fields list:


**Figure 4** provides those answers by displaying in blue the above fields list and in orange those in QML. The striking disparity on Physics and Astronomy (5), Materials Science (8) and Chemistry (13) can be explained from their expected larger processing demands. Besides, the actual growth of QML has happened in a much shorter timespan but pursues the very active growth shown in ML. Therefore, larger uptake of ML and QML is a fundamental need and can be improved by introducing science gateways.

*Science Gateways and AI/ML: How Can Gateway Concepts and Solutions Meet the Needs in Data… DOI: http://dx.doi.org/10.5772/intechopen.110144*

#### **Figure 4.**

*Distributions of QML (orange bar chart) and ML (blue bar chart) into the different fields of Science and Technology. The uptake of QML on ML is described by normalization ratio of QML/ML of 139.6.*

#### **6. The theoretical framework for science gateways in AI/ML research**

AI/ML research faces following challenges that can be addressed by a theoretical framework for integration with science gateways. Each challenge is part of the research question how science gateways can improve AI/ML research and add aspects that are not well considered as part of the AI/ML research yet.

#### **6.1 Availability**

Providing software-as-a-service is a common mode of operation for science gateways. Many gateways are already providing trained ML methods to their user communities. In this sense, an ML application is just another piece of software that a gateway can provide, simplifying access and helping to ensure the software is up to date, is used in the correct way, and is installed on adequate resources.

High-performance AI/ML is a strategic priority, as it will be necessary for these methods to scale to support enormous data sets. Providing simplified access to HPC and other specialized resources is a strength of science gateways, particularly as the landscape of those resources evolves into Graphics Processing Unit (GPU), Field Programmable Gate Arrays (FPGA), and other environments tailored for efficient AI/ML. Various federal initiatives are promoting improved AI standards, access to benchmarked, open-source applications, and access to public data sets usable for training. Science gateways will provide access to these tools and data.

In science gateways, we typically consider AI/ML-based applications as having a target audience beyond that of the developer; that is, a developer has created an application and wants it to reach a larger user community. There is also an important scenario in which a researcher develops an AI/ML application to further their own research; the application itself may have no (or at least no perceived) broader use by others, at least in the initial phase. It is still essential that science gateways are available to support such research. Even if the software itself is never used again beyond

the original scope, its results may be published and thus must be auditable and reproducible by reviewers and future readers.

#### **6.2 Validity, reproducibility, transparency**

Software based on ML methods is very different from traditional scientific software in that the ML-based applications must be trained on data sets. Thus, one must separately validate any results obtained from an ML-based application. Changes in training data sets will give different results, so it is important to track not only the versions of a particular software used but also the version of the trained model and the data sets used in training, including any processes for cleaning or otherwise filtering the data.

Gateways, in their role as supporting scientific software-as-a-service, are already well-positioned to at least support the collection of version metadata needed to support reproducibility or at least the provenance of how a particular result was achieved.

More broadly, many ML applications should be understood as dataflow programs that combine well-known algorithms into specific applications. Experiments to improve the workflow and the validation of the AI methods are currently performed outside most gateways and are considered publication quality research in such journals as *Nature Methods*. Capturing these processes is an important enhancement that could easily be supported by science gateways.

Minimally, a provider of an ML-based application could at least publish the metadata about how a particular application was developed, trained, and validated, but gateways themselves can also support these processes directly. This would enable users to reproduce and inspect the application itself and the training data. Gateways could furthermore track the development of alternative pipelines and trainings by both the code authors and the interested members of the community. This would provide direct support and supplemental information to the methods publications by the original authors.

#### **6.3 Privacy**

Data privacy is the obverse aspect of trustworthiness, since many ML applications may work with sensitive data such as personal health information and proprietary data. Science gateways can be used to support privacy for data sets by limiting access to data through controlled and auditable user and programming interfaces. These gateways can operate within privacy protected environments that support The Health Insurance Portability and Accountability Act (HIPAA), Federal Information Security Modernization Act (FISMA), and other regulated data classifications. Further limitations, such as differential privacy, are open-research areas that can also be supported by gateways.

#### **6.4 Trustworthiness, explainability, and uncertainty quantification**

Trustworthiness, explainability, and uncertainty quantification are larger open problems in AI/ML research. Trustworthiness of scientific results obtained from AI/ML methods in scientific applications can be increased through the ability of science gateways to support reproducibility, auditability, and transparency in tracking how a particular application was developed, trained, and validated. Gateways can serve as a focal point where experimental results, computational results, and AI/MLbased models can be cross compared and validated. Explainability is an open-research area, as the results of many current methods (most notably artificial neural networks) cannot be understood by humans, even if we have full access to the software and

*Science Gateways and AI/ML: How Can Gateway Concepts and Solutions Meet the Needs in Data… DOI: http://dx.doi.org/10.5772/intechopen.110144*

training data. However, as new, more explainable methods are developed, science gateways should be an important delivery mechanism. Science gateways may also be a strong vehicle for delivering codes that analyze AI/ML-based models for explainability. Such explainability and trustworthiness will be essential for R&D applications in regulated spaces such as clean energy systems and climate-mitigation technologies. Trustworthy and transparent workflows for AI/ML models will be necessary to apply those approaches to advanced nuclear energy and integrated energy systems technology R&D, a field which demands high standards of safety and performance. Finally, an ML-based application could be completely trustworthy and explainable and still give the wrong answer; more precisely, the answer has both known and unknown limits. Known limits are probabilities of correctness, and some AI/ML methods (such as reinforcement learning) benefit from continual training and supervision. Collecting feedback on correct versus incorrect outputs needs to be coordinated. AI/ML software deployed into gateways, as opposed to running in separate, isolated environments by each user, can track each methods'success rates.

#### **6.5 Usability and user experience**

Using gateways to enhance usability and the user (and researcher) experience has probably received the least amount of attention by the field. The application of AI/ML methods within gateways to enhance user experience (such as guided access, usage analytics, digital assistants) would enhance science gateways' capabilities and advance the field to a new level of maturity. In this space, we expect Javascript-based AI/ML solutions to become important for gateways. Already, there is a vibrant ecosystem of such frameworks, including: TensorFlow.js<sup>1</sup> , ml5.js<sup>2</sup> , Propel<sup>3</sup> , Brain.js,<sup>4</sup> ml.js,<sup>5</sup> Neuro. js,<sup>6</sup> Synaptic,<sup>7</sup> and others. Worthy of mention, also, is ConvNetJS<sup>8</sup> which has some particularly user-friendly educational web applications. We note that AI has already begun to shape a lot of human/machine interfaces. Google search can know and categorize images from text. Voice assistants have become good at parsing queries and returning data correctly. We expect AI to transform Science Gateways themselves, as well as the way they work. "The main trick here is to allow humans to stay human. For decades computers were not exciting to use as they required us to change our ways."—Heilmann.<sup>9</sup> Gateways were always trying to help the humans stay human, AI/ML should enable them to succeed in new ways.

#### **6.6 AI for gateway cyberinfrastructure**

In addition to opportunities to leverage AI/ML in research, there are also benefits to adopting these technologies in the underlying cyberinfrastructure powering the

<sup>1</sup> https://www.tensorflow.org/js

<sup>2</sup> https://ml5js.org/

<sup>3</sup> https://stackshare.io/propel

<sup>4</sup> https://github.com/BrainJS/brain.js

<sup>5</sup> https://github.com/mljs/ml

<sup>6</sup> https://neuro.js.org/

<sup>7</sup> http://caza.la/synaptic/#/

<sup>8</sup> https://cs.stanford.edu/people/karpathy/convnetjs/

<sup>9</sup> https://www.infoq.com/news/2018/11/human-interfaces-ai/

gateways. For, instance, there are opportunities for centralizing sets of services that could be leveraged for small/short computational jobs (such as classifying an item) that could be provided by frameworks such as Tapis [49], Airavata [50], HUBzero [51], or even commercial cloud services (AWS lambda etc.) with potential to support a hosted catalog of AI/ML functions that could be leveraged by existing and new gateways. Gateways could leverage these lambda-like functions for AI integrations both for AI/ML research as well as incorporating some of these tools into the way the gateway is managed and delivers functionality—recommendations, analyzing gateway data/metrics, and classification of user jobs/workflows that could make gateway operations more efficient and useful to end users/researchers. Further, there is the potential for gateways to leverage AI to enhance usability and accessibility through


*Science Gateways and AI/ML: How Can Gateway Concepts and Solutions Meet the Needs in Data… DOI: http://dx.doi.org/10.5772/intechopen.110144*

Overall, AI will be revolutionary in the way gateways can be developed/managed/ protected and how users will interact with the gateway. Investing in and developing AI-powered tools is a concrete area that will lead to improvements in gateway usability and functionality and doing so in centralized ways can push the entire research community forward.

#### **7. Cybersecurity and critical infrastructure protection**

The primary danger posed by the use of Science Gateways comes from the "community account" model on which these applications are based. Some of the science gateway frameworks create on purpose a single account through which they schedule all the jobs by a group of users of the web-facing interface, for example, [6] allowing for accessing HPC resources nationwide in the USA.

Because it is automated, such systems must submit their jobs (typically) without going through two-factor authentication systems (although the gateway infrastructure could, in principle use a two-factor system on the web-facing side). HUBzero has implemented such modules and allows for re-use of login credentials such as Google accounts only if the two-factor system is used at least once during the first login to the system.

Science gateway frameworks are differently designed in regard to security: some only run a limited set of commands related to moving files and a science code, others such as Jupyter Notebooks take code from a web-facing interface and run it on the target system.

Most of the science codes developed in research domains are often written without any kind of security in mind. AI networks themselves, depending on the details of their training, could, in principle, contain vulnerabilities that would be very hard to identify. Individual applications are likely to have numerous vulnerabilities and a clever hacker could provide input parameter files that trigger buffer overrun attacks, etc. In addition, there is the small (but nonvanishing) chance that the science gateway framework itself might be hacked in some way.

Some solutions help mitigating these issues, for example, singularity with AppArmor on the target system. AppArmor provides kernel-level protection against arbitrary sorts of unwanted access. AI is a promising solution *via* pattern recognition of bad actor behaviors supporting to identify compromised accounts or API security flaws.

#### **8. Outlook**

Outreach to different target groups will be a crucial topic to establish new science gateways in addition to the ones already used in the AI/ML community. Zooniverse is a great example of a science gateway for collaborating on AI methods. It is one of the frameworks we plan to reach out to. The academic community is well reachable at conferences, and thus, presenting about science gateways and AI at a diverse set of conferences can connect the communities. Furthermore, federal funding programs for AI/ML research and development are important influencers in this space. NSF, NASA, Advanced Research Project Agency-Energy (ARPA-E), Office of Nuclear Energy (DOE-NE) and Office of Energy Efficiency & Renewable Energy (DOE-EERE) all have solicitations associated with AI/ML applications. The last three of these agencies

are in the energy space. Existing science gateways in the AI area are good starting points to analyze the uptake of such gateways by the community and to identify existing pain points using such solutions. Partnering with industry is a promising way to accelerate the collaboration between the science gateways and the AI/ML community. The goal is not only to increase the uptake of science gateways for AI methods but also to increase the uptake of AI methods for science gateway infrastructures. Both fields can benefit from each other and accelerate science *via* combining methods for ML with usability of science gateways. Integrating science gateways into teaching will further enhance the knowledge in the community and train students on using and potentially developing science gateways and/or AI/ML concepts and methods and, hence, train the next generation of users and developers. With AI being such an important field in academia and industry, this is a crucial step for highly needed workforce development. This combination of AI/ML and science gateways creates the next generation of critical infrastructure for research and teaching.

#### **Acknowledgements**

The authors would like to acknowledge SGCI and NSF as funding body for SGCI's funding under award 1547611. B.M. Villegas Martínez would like to thank finantial support of the Mexican National Council on Science and Technology (CONACyT) through the postdoctoral scholarship CVU 549550. Wilmer Contreras Sepulveda would like to thank CONACyT for the academic scholarship to develop master studies.

### **Author details**

Sandra Gesing<sup>1</sup> \*, Marlon Pierce2 , Suresh Marru<sup>2</sup> , Michael Zentner<sup>3</sup> , Kathryn Huff<sup>4</sup> , Shannon Bradley<sup>5</sup> , Sean B. Cleveland<sup>6</sup> , Steven R. Brandt<sup>7</sup> , Rajiv Ramnath<sup>8</sup> , Kerk Kee9 , Maytal Dahan10, Braulio M. Villegas Martínez11, Wilmer Contreras Sepulveda11 and José J. Sánchez Mondragón<sup>11</sup>

1 Applied R&D, University of Illinois Systems Discovery Partners Institute, US

2 Cyberinfrastructure Integration Research Center, Indiana University, US

3 Sustainable Scientific Software Division, San Diego Supercomputer Center, US

4 Department of Nuclear, Plasma and Radiological, University of Illinois at Urbana-Champaign, US

5 Innovative Software and Data Analysis, University of Illinois at Urbana-Champaign, US

6 Information Technology Services Cyberinfrastructure, University of Hawaiʻi, US

7 Computer Science and Engineering, Louisiana State University, US

8 Computer Science and Engineering, The Ohio State University, US

9 College of Media and Communication, Texas Tech University, US

10 Advanced Computing Interfaces, Texas Advanced Computing Center, US

11 Optics Department, National Institute of Astrophysics, Optics and Electronics, Mexico

\*Address all correspondence to: sgesing@uillinois.edu

© 2023 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

### **References**

[1] RDA VRE-IG. https://www.rd-allia nce.org/groups/vre-ig.html. [Online; Accessed: 09 March 2021]

[2] Gesing S, Brandt S, Bradley S, Potkewitz M, Kee K, Whysel N, Perri M, Cleveland S, Rugg A, Smith J. A vision for science gateways: bridging the gap and broadening the outreach. In: Practice and Experience in Advanced Research Computing, PEARC '21, New York, NY, USA, 2021. Boston, MA, USA, Association for Computing Machinery

[3] AI4Mars. https://www.zooniverse. org/projects/hiro-ono/ai4mars. [Online; Accessed: 09 March 2021]

[4] Zooniverse. https://www.zooniverse. org/. [Online; Accessed: 09 March 2021]

[5] Gesing S, Zentner M, Clark S, Stirm C, Haley B. Hubzero®: novel concepts applied to established computing infrastructures to address communities' needs. In: Proceedings of the Practice and Experience in Advanced Research Computing on Rise of the Machines (Learning), PEARC '19, New York, NY, USA, 2019. Chicago, IL, USA, Association for Computing Machinery

[6] Pierce M, Marru S, Abeysinghe E, Pamidighantam S, Christie M, Wannipurage D. Supporting science gateways using apache airavata and scigap services. In: Proceedings of the Practice and Experience on Advanced Research Computing, PEARC '18, New York, NY, USA, 2018. Pittsburgh, PA, USA, Association for Computing Machinery

[7] Padhy S, Jamthe A, Cleveland SB, Smith JA, Stubbs J, Garcia C, Packard M, Terry S, Looney J, Cardone R, Dahan M, Jacobs GA. Building tapis v3 streams api support for real-time streaming data

event-driven workflows. In: Practice and Experience in Advanced Research Computing, PEARC '21, New York, NY, USA, 2021. Boston, MA, USA, Association for Computing Machinery

[8] Gateways Conferences. https://scie ncegateways.org/engage/annual-confere nce. [Online; Accessed: 09 March 2021]

[9] PEARC Conference. https://pearc. acm.org/. [Online; Accessed 09 March 2021]

[10] eScience Conference. https://escie nce2021.org/. [Online; Accessed: 09 March 2021]

[11] SC Conference Series - The International Conference for High Performance Computing, Networking, Storage, and Analysis. https://superc omputing.org/. [Online; Accessed: 09 March 2021]

[12] The 8th International Online & Onsite Conference on Machine Learning, Optimization, and Data Science. https:// lod2022.icas.cc/. [Online; Accessed: 09 March 2021]

[13] Atkinson M, Gesing S. IWSG 2018 - International Workshop on Science Gateways 2018. CEUR Workshop Proceedings: IWSG; 2019

[14] NSF AI Institutes. https://www.nsf. gov/pubs/2020/nsf20604/nsf20604. htm. [Online; Accessed: 09 March 2021]

[15] NSF OAC. https://www.nsf.gov/cise/ oac/about.jsp. [Online; Accessed 09 March 2021]

[16] The National Artificial Intelligence Research and Development Strategic Plan - 2019 update. https://www.nitrd. *Science Gateways and AI/ML: How Can Gateway Concepts and Solutions Meet the Needs in Data… DOI: http://dx.doi.org/10.5772/intechopen.110144*

gov/pubs/National-AI-RD-Strategy-2019.pdf. [Online; Accessed: 09 March 2021]

[17] 3 questions about the Amazon– National Science Foundation collaboration on fairness in AI. https:// www.amazon.science/3-questionsabout-the-amazon-national-sciencefoundation-collaboration-on-fairness-inai. [Online; Accessed: 09 March 2021]

[18] Turing AM. Computing Machinery and Intelligence. Netherlands, Dordrecht: Springer; 2009. pp. 23-65

[19] McCulloch W, Pitts W. A logical calculus of the ideas immanent in nervous activity. The Bulletin of Mathematical Biophysics. 1943;**5**(4): 115-133

[20] Towns J, Cockerill T, Dahan M, Foster I, Gaither K, Grimshaw A, et al. Xsede: accelerating scientific discovery. Computing in Science & Engineering. 2014;**16**(05):62-74

[21] Pordes R, Petravick D, Kramer B, Olson D, Livny M, Roy A, et al. The open science grid. Journal of Physics: Conference Series. 2007;**78**:012057

[22] Abadi M, Barham P, Chen J, Chen Z, Davis A, Dean J, et al. Tensorflow: a system for large-scale machine learning. In: Proceedings of the 12th USENIX Conference on Operating Systems Design and Implementation, OSDI'16, pp. 265–283, USA, 2016. Savannah, GA, USA, USENIX Association

[23] The National Artificial Intelligence Research and Development Strategic Plan 2016. https://www.nitrd.gov/PUBS/ national\_ai\_rd\_strategic\_plan.pdf. [Online; Accessed: 09 March 2021]

[24] The National Artificial Intelligence Research and Development Progress

Report. https://www.whitehouse.g ov/wp-content/uploads/2019/11/AI-Re search-and-Development-Progress-Re port-2016-2019.pdf. [Online; accessed 09-03-2021]

[25] Artificial Intelligence (AI) at NSF. https://nsf.gov/cise/ai.jsp. [Online; Accessed: 09 March 2021]

[26] DARPA AI Next Campaign. https:// www.darpa.mil/work-with-us/ai-next-ca mpaign. [Online; Accessed: 09-03-2021]

[27] NIH Artificial Intelligence, Machine Learning, and Deep Learning. https:// www.nibib.nih.gov/research-fund ing/machine-learning. [Online; Accessed: 09 March 2021]

[28] DOE Artificial Intelligence and Technology Office. https://www.energy. gov/science-innovation/artificial-inte lligence-and-technology-office. [Online; Accessed: 09 March 2021]

[29] Gil Y, Selman B. A 20-Year Community Roadmap for Artificial Intelligence Research in the US. Computing Community Consortium (CCC) and Association for the Advancement of Artificial Intelligence (AAAI). 2019. arXiv preprint arXiv: 1908.02624. Available at: https://cra.org/ ccc/resources/workshopreports/

[30] Stone P, Brooks R, Brynjolfsson E, Calo R, Etzioni O, Hager G, et al. Artificial Intelligence and Life in 2030. One hundred year study on artificial intelligence: Report of the 2015-2016 Study Panel. Stanford University, Stanford, CA, p. 2016, 2016

[31] Lawrence KA, Zentner M, Wilkins-Diehr N, Wernert JA, Pierce M, Marru S, et al. Science gateways today and tomorrow: positive perspectives of nearly 5000 members of the research community. Concurrency and

Computation: Practice and Experience. 2015;**27**(16):4252-4268

[32] Wilkinson MD, Dumontier M, Aalbersberg IJJ, Appleton G, Axton M, Baak A, et al. The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data. 2016 Mar 15;**3**:160018. DOI: 10.1038/sdata.2016.18. Erratum in: Sci Data. 2019 Mar 19;6(1):6. PMID: 26978244; PMCID: PMC4792175. Available from: https://pubmed.ncbi. nlm.nih.gov/26978244/

[33] Project at the intersecion of AI and physics. http://www.ncsa.illinois.edu/ news/story/\doe\_awards\_2.2m\_to\_ project\_at\_the\\_intersection\_of\_ai\_ and\_high\_energy\_physic. [Online; Accessed: 09 March 2021]

[34] Collaborative Research: Frameworks: Machine learning and FPGA computing for real-time applications in big-data physics experiments. https://figshare.com/ articles/poster/\Collaborative\_Research\_ Frameworks\_Machine\_Learning\_and\\_ FPGA\_computing\_for\_real-time\_applica tions\_in\_big-data\_physics\_experiments/ 11803764. [Online; Accessed: 09 March 2021]

[35] Akiyama K, Alberdi A, Alef W, Asada K, Azulay R, Baczko A-K, et al. First M87 event horizon telescope results. IV. Imaging the central supermassive black hole. The Astrophysical Journal Letters. 2019; **875**(1):L4

[36] Shi B, Calabretta N, Stabile R. Numerical simulation of an inp photonic integrated cross-connect for deep neural networks on chip. Applied Sciences. 2020;**10**(2):474. DOI: 10.3390/ app10020474

[37] Ahmed S, Sánchez Muñoz C, Nori F, Kockum AF. Classification and

reconstruction of optical quantum states with deep neural networks. Physical Review Research. 2021;**3**: 033278

[38] Bhusal N, Lohani S, You C, Hong M, Fabre J, Zhao P, et al. Spatial mode correction of single photons using machine learning. Advanced Quantum Technologies. 2021;**4**(3):2000103

[39] You C, Quiroz-Juárez MA, Lambert A, Bhusal N, Dong C, Perez-Leija A, et al. Identification of light sources using machine learning. Applied Physics Reviews. 2020;**7**(2): 021404

[40] AGNet. http://www.ncsa.illinois.ed u/about/fellows\_awardees/\agnet\_weigh ing\_black\_holes\_with\_deep\_learning. [Online; Accessed: 09 March 2021]

[41] Agriculture. https://calendars. illinois.edu/detail/6824/33401360. [Online; Accessed: 09 March 2021]

[42] Animal husbandry. https://ai.ncsa. illinois.edu/research/funded-projects/. [Online; Accessed: 09 March 2021]

[43] Food security. https://sciencecouncil.food.gov.uk/sites/default/ \files/appofadvancedanalyticsfinalre port.pdf. [Online; Accessed: 09 March 2021]

[44] Liljedahl AK, Jones BM, Brubaker M, Budden AE, Cervenec JM, Grosse G, Jones MB, Marini L, McHenry K, Moss J, Morin PJ, Nitze I, Soliman A, Wind G, Witharana C. Permafrost discovery gateway: a web platform to enable discovery and knowledge-generation of permafrost big imagery products. In: AGU Fall Meeting 2019, San Francisco, USA, December 2019. AGU

[45] Infrastructure Deserts. https:// storymaps.arcgis.com/stories/\7b8c *Science Gateways and AI/ML: How Can Gateway Concepts and Solutions Meet the Needs in Data… DOI: http://dx.doi.org/10.5772/intechopen.110144*

2029e30749449a3a692451e51ddf. [Online; Accessed: 09 March 2021]

[46] Wang Z, Christie MA, Abeysinghe E, Chu T, Marru S, Pierce M, and Danko CG. Building a science gateway for processing and modeling sequencing data via apache airavata. In: Proceedings of the Practice and Experience on Advanced Research Computing, pp. 1–7. 2018

[47] Zhou J, Smith K, Wilsbacher G, Sagona P, Reddy D, and Torkian B. Building science gateways for humanities. In: Practice and Experience in Advanced Research Computing, pp. 327–332. 2020

[48] Jia Y, Shelhamer E, Donahue J, Karayev S, Long J, Girshick R, Guadarrama S, Darrell T. Caffe: convolutional architecture for fast feature embedding. In: Proceedings of the 22nd ACM international conference on Multimedia, pp. 675–678, 2014

[49] Cleveland SB, Jamthe A, Padhy S, Stubbs J, Packard M, Looney J, et al. Tapis api development with python: best practices in scientific rest api implementation: Experience implementing a distributed stream api. In: Practice and Experience in Advanced Research Computing, PEARC '20, pp. 181–187, New York, NY, USA, 2020. Portland, OR, USA, Association for Computing Machinery

[50] Marru S, Gunathilake L, Herath C, Tangchaisin P, Pierce M, Mattmann C, et al. Apache airavata: a framework for distributed applications and computational workflows. In: Proceedings of the 2011 ACM Workshop on Gateway Computing Environments, GCE '11, pp. 21–28, New York, NY, USA, 2011. Seattle, Washington, USA, Association for Computing Machinery

[51] David Benham and Sandra Gesing. Hubzero© goes onescienceplace: The next community-driven steps for providing software-as-a-service. In: 2019 15th International Conference on eScience (eScience), pp. 642–643, 2019

### Section 3
