Simultaneous forecasting of several natural disasters - conception

ACADEMIA Letters Simultaneous forecasting of several natural disasters conception Sania Kalitska 1. Justification of the feasibility of this program The article presents the innovative research program’s conception of the simultaneous forecasting of various natural disasters: earthquakes, volcanic eruptions, fires, floods, hurricanes, storms, long-term droughts, severe frosts and blizzards, heavy rains, avalanches, epidemics, and others. The area of interest may also be human activity as a source of natural hazards and / or the intensification of their impact on the environment. In this case, scientific research, called forensic investigation, aims to provide an objective and evidence-based description and interpretation of an event in order to prevent similar events in the future or to improve disaster management and contribute to building resilience to threats. The term “forensics” is not intended to impose legal responsibility. Slowing climate changes is an urgent task. Our world is getting closer to an even more widespread crisis than the global COVID-19 pandemic. Every catastrophe makes the situation sharply aggravated, accelerating the onset of this crisis. There are many studies examining single disasters, or rather their consequences, the main goal of which is to work out a response to them and prepare as best as possible for the same phenomenon to occur again in the same place, it is generally unknown when. There are four factors supporting the fact that today the problem of predicting several disasters at the same time is solvable: Academia Letters, June 2022 ©2022 by the author — Open Access — Distributed under CC BY 4.0 Corresponding Author: Sania Kalitska, skalitska@gmail.com Citation: Kalitska, S. (2022). Simultaneous forecasting of several natural disasters - conception. Academia Letters, Article 5562. 1 • a huge amount of information is collected and systematized about the history of catastrophes on the Earth; • a lot of sound environmental research (terrain, soil moisture and temperature, air quality, snow / ice depth, yields, etc.) has been done and performed by international teams that typically include China, US, EU countries, Australia etc. [1]; • the use of more and more accurate measurement systems, located on satellites, airplanes, drones, etc., significantly expanded the area of research; for example, terrain measurements were made taking into account all possible parameters that contribute to the accuracy of the measurements: vegetation, incidence of measurement signals, soil moisture, the influence of weather (light, sunny, dark, rain, snow, …) etc.; • the use of AI methods and the ability to manage big data in AI Center of Excellence. There is literature that describes the history of catastrophes on Earth from time immemorial. For example, the work of Emmer [2] is an excellent overview of 580,000 studies on various types of natural hazards around the world published between 1900 and 2017. Very subtle changes in environmental parameters at the level of “white noise” are always – it is normal state. The possibility of any of several disasters may be indicated by the deviation of the relevant environmental parameters from the normal state. The more parameters apparently deviate from the norm, the higher the probability of relevant catastrophe(s). In practice, we deal with a huge amount of recorded data about environmental parameters through a huge number of measurement systems. In such a situation, the problem can be solved only by AI methods and modern methods of big data management, which is why the application of the AI Center of Excellence was proposed here, which will be used to build AI Networks of Excellence centers around the globe. 2. The conception of forecasting natural disasters The disaster forecasting conception represents two main problems: • how can you build an effective AI Network of Excellence; • how can you manage large amounts of data and perform accurate forecasting in AI Center of Excellence. Academia Letters, June 2022 ©2022 by the author — Open Access — Distributed under CC BY 4.0 Corresponding Author: Sania Kalitska, skalitska@gmail.com Citation: Kalitska, S. (2022). Simultaneous forecasting of several natural disasters - conception. Academia Letters, Article 5562. 2 2.1. Principles of building an AI Network of Excellence Below is a proposal on how to build a network of cooperating AI Centers of Excellence. The process can be represented as follows. 1. All analyzed catastrophes are divided into groups with numbers i = 1,…, I. Every group contains types of catastrophes, correlated with each other through the examined parameters, e.g., group 1: (flood, prolonged drought, fire). Each group has its own AI Centers of Excellence (i, j), j =1,…Ji, where Ji.- number of centers in the group i. These centers are located on Earth depending on the agreed criteria: the frequency of disasters, areas covered by AI Centers of Excellence (i, j), the lowest research costs, minimizing the risk of errors, maximizing the reliability of disaster forecasting, etc. 2. Creating AI Network of Excellence (i), i =1,. I from all centers AI Center of Excellence (i,j) forecasting disasters of groups i = 1,…, I on Earth. 3. Creating the AI Network of Excellence from all AI Network of Excellence (i), i = 1,…, I, to forecast disasters on Earth. This networks would change along with changes in the conception of disaster assessment, modernization of environmental parameter assessment methods, expansion of the geography of the network or disaster types, etc. 2.2. Managing the forecasting process in the AI Center of Excellence Any problem using big data needs two different, unrelated types of analysis to avoid conflict between data science groups and the research that is the focus of the work. To this end, in [3,4] was proposed a “data science bridge”: an intermediary group tasked with ensuring better communication and integration between the two groups. Data science project management should be presented as a continuous loop (Fig 1) [3]: the overall research strategy is channeled towards the ‘data science bridge’, the team that oversees all research projects in this AI Center of Excellence. Academia Letters, June 2022 ©2022 by the author — Open Access — Distributed under CC BY 4.0 Corresponding Author: Sania Kalitska, skalitska@gmail.com Citation: Kalitska, S. (2022). Simultaneous forecasting of several natural disasters - conception. Academia Letters, Article 5562. 3 The results are then looped to provide new insights into the overarching project strategy. This strategy defines what needs to be achieved and provides a high-level direction to the data science bridge. Good governance includes stimulating collaboration, developing human capital, ensuring data quality, managing a project portfolio, and ensuring the business impact of all data science activities. Let us pay attention to two basic tasks (Core tasks) related to the human factor - cooperation of various people in the big data management process. Here, interpersonal relations and cooperation are of particular importance because the team consists of people from different countries around the globe. This work would be attended by renowned high-class scientists and specialists, each of whom has their own experience and vision to solve problems in their field. Collaborative culture [5] is characterized by the degree of mutual sharing of one’s substantive achievements and having personal interests in this. By adopting a more adaptive approach to creative collaboration, each program design increases its chances of bringing new ideas to work together. Under such conditions, there must be consensus on every matter Academia Letters, June 2022 ©2022 by the author — Open Access — Distributed under CC BY 4.0 Corresponding Author: Sania Kalitska, skalitska@gmail.com Citation: Kalitska, S. (2022). Simultaneous forecasting of several natural disasters - conception. Academia Letters, Article 5562. 4 relating to the problems to be solved in the projects and thus in the entire program. Major changes in cooperation arose in connection with the need to work online during the pandemic [6]. There are two complementary principles for managing remote innovation teams: connecting online to collaborate and create contradictions. Then the drive to innovate is seen through people intentionally coming together on a one-to-one basis - in their own remote teams or upwards - to promote ideas and build support. Human capital consists of knowledge, skills, and health. It is a universal issue, even there is a special project by The Human Capital Project targeting people all over the globe to eradicate extreme poverty and create more integrated societies. If we are talking about developing human capital in the company, the creation of AI Centers of Excellence is a good example of this. One of the main goals of the center is cooperation with mutual knowledge sharing, without which it is impossible to solve the problems we are talking about. At the same time, staff of the future were trained in such an important and difficult topic that we analyzed. Solutions of analytical problems to solve an analytics problem, the marketing group needs to have a good understanding of the strategic challenge of the problem, identify the required data, and then design an appropriate analytical solution. This may mean acquiring new data, accessing new data from the same or a different measuring device, turning to other analytical methods. Importantly, this approach increases the likelihood that employees will benefit from better data sources and analytical methods that directly map their professional challenges. Therefore, the development of cyber infrastructure is needed for fast big data transfer, intuitive data visualization and overall efficient data management. Here, Human-Machine calculations are proposed, in which AI technology is combined with the intellect of researchers. This method was proposed and is actively developing at MIT [7]. In general, AI is used in situations where it can perform repetitive specific tasks or detect relationships between data flows from different sources, where the speed and quality of this execution exceeds human capabilities, however, when making AI decisions, the machine needs human judgment because it does not have the full necessary context. AI should recommend, and people should decide. 2.3. Managing the forecasting process in networks AI Network of Excellence (.) and AI Network of Excellence In each AI Network of Excellence (i), we can evaluate correlations between the results obtained at different AI Centers of Excellence (i,j) of this network: the probability of disasters, the time of their occurrence, the impact on them of parameters other than those assumed in each individual central. In-depth analysis using AI methods must have a wide range of paAcademia Letters, June 2022 ©2022 by the author — Open Access — Distributed under CC BY 4.0 Corresponding Author: Sania Kalitska, skalitska@gmail.com Citation: Kalitska, S. (2022). Simultaneous forecasting of several natural disasters - conception. Academia Letters, Article 5562. 5 rameters that may at first glance have nothing to do with a defined catastrophe. This analysis requires penetrating research of top-class specialists. An analysis of the entire AI Network of Excellence further expands the range of factors contributing to disasters forecasting. Perhaps an analysis of the deep processes around the Earth and in its depths will be needed. Only the future can show that. Academia Letters, June 2022 ©2022 by the author — Open Access — Distributed under CC BY 4.0 Corresponding Author: Sania Kalitska, skalitska@gmail.com Citation: Kalitska, S. (2022). Simultaneous forecasting of several natural disasters - conception. Academia Letters, Article 5562. 6 References [1] ScienceDirect_articles_08Sep2021_19-17-34.6 z “Remote Sensing of Environment”, journal finder “Elsevier” [2] Emmer “Geographies and scientometrics of research on natural hazards”, Geosciences 2018, 8, 382] [3] Roger W. Hoerl, Diego Kuonen, and Thomas C. Redman “To Succeed With Data Science, First Build the ‘Bridge’, To Succeed With Data Science, First Build the ‘Bridge’”, MIT Sloan Management Review, October 22, 2020 [4] Roger Hoer, Diego Kuonen, and Thomas C. Redman “The Data Science Management Process”, MIT Sloan Management Review July 12, 2021 [5] Peter Gray, Rob Cross, and Michael “Use Networks to Drive Culture Change”, November 30, 2021 [6] Esther Tippmann, Pamela Sharkey Scott, and Mark Gantly. Driving Remote Innovation Through Conflict and Collaboration MIT Sloan Management Review April 15, 2021 [7] Allison Ryder “The Key to Success With AI Is Human-Machine Collaboration”, MIT Academia Letters, June 2022 ©2022 by the author — Open Access — Distributed under CC BY 4.0 Corresponding Author: Sania Kalitska, skalitska@gmail.com Citation: Kalitska, S. (2022). Simultaneous forecasting of several natural disasters - conception. Academia Letters, Article 5562. 7