Seminar for
DECISION MAKING – THEORY, TECHNOLOGY AND PRACTICE

 

PROGRAM

---

Plan rada Seminara Odlučivanje - teorija, tehnologija, praksa za DECEMBAR 2023.

---

Četvrtak, 07.12.2023. u 13:00, Pariske Komune bb, Niš i Online
Ivana Petkovski, Mathematical Institute of the Serbian Academy of Sciences and Arts
MULTI-CRITERIA DECISION-MAKING IN THE EVALUATION OF ENERGY AND ENVIRONMENTAL INDICATORS IN THE DIGITAL SOCIETY
In the era of accelerated digitalization, rational energy consumption and environmental pollution are becoming central topics of discussion and analysis by experts. This research paper examines how digitalized societies manage energy consumption and mitigate carbon dioxide emissions. The focus is on European countries that distinguish themselves for their digital development. Through the application of the hybrid CRITIC-TOPSIS multi-criteria decision-making model, this research provides a practical framework for ranking countries according to their energy and environmental attributes. The research results contribute to the identification of leaders in efficient energy and environmental management within the framework of digital development.

---

Utorak, 12.12.2023. u 14:15, Pariske Komune bb, Niš i Online
Predrag Stanimirović, Prirodno-matematički fakultet, Niš
RECURRENT NEURAL NETWORKS IN NUMERICAL LINEAR ALGEBRA
Recurrent Neural Networks (RNNs) allow the output from some nodes to influence the subsequent values of the inputs of the same nodes. The lecture discusses RNNs dedicated to finding the zeros of some matrix equations or minimizing nonlinear functions. Such RNN models represent dynamic systems based on nonlinear optimization. Networks for optimization can be divided into two classes: Gradient Neural Networks (GNN) and Zhang Neural Networks (ZNN). The ZNN design was proposed in 2001 and is intensively used in solving various time-varying problems. The dynamics of the ZNN model derives from the selection of the appropriate error function, which is known as the Zhang function (Zf). Some specific ZFs that result from the application of non-linear optimization methods are shown. The convergence of ZNN dynamic systems intended for solving time-varying matrix and vector equations and calculating general inverses is investigated. GNN dynamic systems intended for solving some matrix equations and calculating general inverses of constant matrices are presented and analyzed. The connection of GNN and ZNN dynamic systems with nonlinear optimization methods is analyzed.

---

Četvrtak, 14.12.2023. u 13:00, Online
Una Stanković, Mathematical Institute of the Serbian Academy of Sciences and Arts
UNMANNED AERIAL VEHICLES (UAV) PATH PLANNING TECHNIQUES AND CONSTRAINTS IN URBAN AIRSPACE INTEGRATION: LITERATURE REVIEW
Unmanned Aerial Vehicles (UAVs), commonly known as drones, have revolutionized numerous industries by offering unprecedented capabilities and possibilities. However, the widespread adoption of UAVs presents significant challenges related to safety, airspace management, and regulatory compliance. I this review paper, we emphasize the importance of UAVs and how the implementation of an efficient path-planning system will ensure safe and effective UAV operations. We provide an overview of the most recent methodologies found in the literature. Firstly, we present the wide range of applications that UAVs have found in various industries. Secondly, we highlight the most critical challenges that we must overcome to enable the seamless integration of UAVs into the airspace. Finally, we provide an overview of the most recent methodologies and approaches to tackle path planning. This analysis highlights the diversity of algorithms used for UAV path planning and provides insights into their strengths and limitations. Our main goal is to become acquainted with the problem variants, applied methodologies, and the state-of-the-art challenges in this field and to identify the possible points of improvements and potential contributions.
This is a joint work with Emir Ganić, University of Belgrade, Faculty of Transport and Traffic Engineering, Tatjana Davidović and Dragan Urošević, Mathematical Institute of the Serbian Academy of Sciences and Arts, Belgrade.

---

Ponedeljak, 25.12.2023. u 16:00, Knez Mihailova 36, sala 301f i Online
Slobodanka Boljanović, Mathematical Institute of the Serbian Academy of Sciences and Arts
FATIGUE ENDURANCE ASSESSMENTS OF AIRCRAFT LUG WITH CRACK-LIKE FLAW
The vast majority of large moving systems inevitably face safety-critical flaws such as corrosion and cracks. Therefore, the need to develop reliable computational models that allow real-time quantification of fatigue responses for such surface flaws is still one of key aspects in the failure relevant investigations. The present research proposes a computational design model for assessing the failure strength of lug-pin linkage under the threat of crack-like flaw, exposed to dynamic loading. In order to quantify the interaction between environmental effects and applied service loading effects, the fracture mechanics-based concept will be employed. The time-variant fatigue vulnerability of aircraft lug-type joint will be also theoretically examined through novel damage tolerance-based solutions for residual life evaluations taking into account the lug-pin effect. Further, the suitability of the developed model will be analyzed through case studies by employing experimental datasets available in the literature. Different theoretical outcomes, successfully verified, will demonstrate that damage tolerance-based analytical tool can be used as an appropriate guide for precise and specific management in fatigue relevant applications to protect the load bearing capacity and safe integrity of failure critical lug-pin joints.
Zajedničko predavanje sa seminarom iz Biomehanike, bioinžinjeringa i matematičke biologije.

---

Predavanja su namenjena sirokom krugu slusalaca, ukljucujuci studente redovnih i doktorskih studija. Seminar će se održavati svakog drugog četvrtka od 13:00 - 14:00^h, CIITLAB, Elektronski fakultet Niš, Aleksandra Medvedeva 14, Niš

---