Analiza fizioloških indikatora stresa i performanci u simulacijskim vježbama

master's thesis

• Title (english): Analysis of physiological indicators of stress and performance in simulated exercises
• Author: Daria Vanesa Cvitković
• Mentor: Siniša Popović (mentor)
• Committee member: Siniša Popović (predsjednik povjerenstva)
• Committee member: Alan Jović (član povjerenstva)
• Committee member: Marko Đurasević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Stres je riječ koja opisuje ljudske reakcije na emocionalna, kognitivna i fizički zahtjevna iskustva. Previsoke razine stresa mogu izazvati smanjenu efikasnost u obavljanju zadataka i razne mentalne i fizičke posljedice ukoliko je pojedinac njemu dugo izložen, a ne raspolaže adekvatnim metodama suočavanja sa stresom. Iako ne postoji izravan način mjerenja stresa, moguće je mjeriti sa stresom povezane fiziološke promjene, kao što su primjerice promjene u kardiovaskularnoj, elektrodermalnoj i respiracijskoj aktivnosti.
Jedno od visoko stresnih zanimanja je kontrolor zračnog prometa, s obzirom na odgovornost za sigurnost i živote ljudi u avionima. Zbog ovog razloga, ovakvo radno mjesto zahtijeva visoke razine rezilijentnosti na stres, prilagodljivosti i samokontrole.
Simulatori kontrole zračnog prometa su okruženja u kojima se u strogo kontroliranim uvjetima i na siguran način mogu provoditi obuke i istraživanja s kontrolorima zračnog prometa. U radu je opisan jedan takav simulator niske razine vjernosti na temelju kojeg je provedeno istraživanje o povezanosti fizioloških značajki i performanci na simulacijskim vježbama. Tijekom simulacije ispitanicima su između ostalog mjereni srčani ritam, vodljivost kože i respiracija. U okviru ovoga rada iz signala srčanog ritma i vodljivosti kože izvučene su određene fiziološke značajke, kao i značajke performanci koje pokazuju uspješnost u izvedbi simulacijske vježbe. Rezultati su pokazali da nema statistički značajne razlike u predloženim fiziološkim značajkama između ispitanika viših i nižih performanci na simulacijskoj vježbi, ali postoji par statistički značajnih korelacija fizioloških značajki sa značajkama koje su korištene u rangiranju ispitanika prema performancama. Algoritmi strojnog učenja pri predviđanju razine performanci samo na temelju predloženih fizioloških značajki dali su ograničeno dobre preliminarne rezultate, koje je moguće dalje unaprjeđivati i proširivati u okviru budućeg rada.

Abstract (english)

Stress is a word that describes human reactions to emotionally, cognitively and physically demanding experiences. Excessively high levels of stress can cause reduced efficiency in task performance, as well as various mental and physical consequences if the individual is exposed to such stress for a long time and does not have adequate stress coping methods. Although there is no direct way to measure stress, it is possible to measure stress-related physiological changes, such as changes in cardiovascular, electrodermal and respiratory activity.
One of the highly stressful occupations is the air traffic controller, given the responsibility for the safety and lives of people on airplanes. For this reason, this type of job requires high levels of resilience to stress, adaptability, and self-control.
Air traffic control simulators are environments where training and research with air traffic controllers can be conducted under strictly controlled conditions and in a safe manner. This thesis describes one such low-fidelity simulator, on the basis of which research was conducted regarding the relationship between physiological features and performance on simulation exercises. During the simulation, the subjects’ heart rate, skin conductance and respiration were measured, among other things. Within the scope of this work, certain physiological features were extracted from the heart rate and skin conductance signals, as well as features of performance on the simulation exercise. The results showed that there are no statistically significant differences in the proposed physiological features between subjects with higher and lower performance on the simulation exercise, but there are a couple of statistically significant correlations of physiological features with the features that were used in the ranking of subjects according to performance. Machine learning algorithms for predicting the level of performance based only on the proposed physiological features obtained limited good preliminary results, which may be further improved and expanded upon within the scope of future work.

• Keywords: simulator; kontrola zračnog prometa; stres; rezilijentnost; fiziološke značajke; performance; strojno učenje
• Keywords (english): simulator; air traffic control; stress; resilience; physiological features; performance; machine learning
• Language: croatian
• URN:NBN: urn:nbn:hr:168:798269
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 20:13:32