Modeling for managing Food-Energy-Water-Social-Environmental- NEXUS security: Novel systems’ analysis approaches

Zagorodny, A.G., Bogdanov, V.L., Ermolieva, T., & Komendantova, N. ORCID: https://orcid.org/0000-0003-2568-6179 (2023). Modeling for managing Food-Energy-Water-Social-Environmental- NEXUS security: Novel systems’ analysis approaches. In: Ювілейна Міжнародна науково-практична конференція “Глушковські читання”. До 100-річчя з дня народження В.М. Глушкова [GLUSHKOV READINGS"- 2023” To the 100th anniversary of the birth of V.M. Glushkov.].

Slideshow Presentation_Zagorodny_Bogdanov_et_al_Glushkov's_leacture261023.pptx - Published Version Available under License Creative Commons Attribution Non-commercial. Download (16MB)

Abstract

The aim of the talk is to introduce and discuss novel systems’ analysis models and methodologies, in particular, robust machine learning, being developed within a joint National Academy of Science, Ukraine, and International Institute for Applied Systems Analysis project -- Integrated modeling for robust management of food-energy-water-social-environmental (FEWSE) nexus security and sustainable development [1-2]. These approaches enable science-based support of policies providing coherent strategic coordination and regulations among food, energy, water, social sectors, accounting for complex linkages and differences in spatial and temporal scales between agriculture, energy and water security, potential systemic risks, and new feasible policies at various levels. Thus often, detailed sectoral and regional models have been used to independently plan desirable developments of respective sectors and regions. However, solutions that are optimal for a sub-system may turn out to be infeasible for the entire system. The talk presents new approaches based on the linkage of detailed distributed models of subsystems (e.g., sectoral and regional models) under joint resource constraints thus allowing for truly integrative decision support through optimal and robust solutions across sectors and regions [2-3]. The linkage solution procedure is based on the parallel solving of equivalent nonsmooth optimization model following a simple iterative subgradient algorithm [4-6]. The procedure can be considered as a new robust machine learning algorithm, namely, as a general endogenous reinforced learning problem of how software agents (models) take decisions in order to maximize the cumulative reward (total welfare). Based on novel ideas of systems’ linkage under asymmetric information and other uncertainties, we discuss nested strategic-operational and local-global welfare models which are used in combination with, in general, non-Bayesian probabilistic downscaling procedures for analyzing and managing systemic interdependencies and risks. Quantile-based indicators [7] are used to cope with new type of endogenous risks and extreme events generated by decisions of various stakeholders. For long-term sustainable functioning of FEWE systems, robust policies comprise both interdependent strategic long-term (anticipative, ex-ante) decisions and short-term (adaptive, ex-post) decisions (adjustments). The methodologies and models will be illustrated with relevant case studies.

Actions (login required)

View Item