Ponazoritev slovenskega železniškega sistema z rastjo sluzavke Physarum polycephalum v dvodimenzionalnem sistemu

Ribič, Peter (author), Zalar, Polona (mentor)

Uporaba standardnih matematičnih algoritmov pogosto ne zadostuje pri reševanju kompleksnih kombinatoričnih problemov, kot je npr. tvorba transportnih omrežij. Kot rešitev za tovrstne probleme se vedno bolj uveljavljajo algoritmi, ki temeljijo na principih iz narave in biologije. Eden izmed bolj preiskovanih organizmov na tem področju je sluzavka Physarum polycephalum. Sposobna je kompleksnega odločanja, asociativnega učenja, ima prostorski in časovni spomin, zmožna je reševanja labirintov po najkrajši poti, z njo lahko tvorimo logična vrata, elektronske komponente ter transportna omrežja. V tej diplomski nalogi smo podrobneje opisali biološko podlago za omenjene lastnosti in raziskali uporabo sluzavke za izboljšanje obstoječega transportnega omrežja na primeru slovenskih železnic. Predpostavili smo, da je sluzavka sposobna tvoriti učinkovito omrežje, ki bo primerljivo ali celo boljše od obstoječega omrežja slovenskih železnic. Po opravljenem poskusu smo to hipotezo tudi potrdili, saj je sluzavka tvorila omrežje, ki je bilo po hitrosti potovanja enako omrežju slovenskih železnic, na račun 29 % daljšega omrežja pa je sluzavka tvorila več kot dvakrat bolj odporno omrežje na okvare od obstoječega sistema. Kljub temu je direktna primerjava težavna zaradi neupoštevanja določenih omejitev, ki vplivajo na tvorbo železniškega omrežja in jih nismo uspeli zajeti v poskusu. Vedenje sluzavke je zato smotrno prevesti v in silico modele, ki imajo precejšen potencial za konstrukcijo učinkovitih algoritmov, ki lahko rešijo širok nabor težav v transportu, informatiki, računalništvu in strojništvu.

• Language: Slovenian
• Keywords: Physarum polycephalum, slovenske železnice, protoplazemski tok, inteligenca roja, tvorba omrežij
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: BF - Biotechnical Faculty
• Year: 2023
• PID: 20.500.12556/RUL-147617
• COBISS.SI-ID: 158361347
• Publication date in RUL: 10.07.2023

RIBIČ, Peter, 2023, Ponazoritev slovenskega železniškega sistema z rastjo sluzavke Physarum polycephalum v dvodimenzionalnem sistemu [online]. Bachelor’s thesis. [Accessed 12 April 2025]. Retrieved from: https://repozitorij.uni-lj.si/IzpisGradiva.php?lang=eng&id=147617

Secondary language

• Language: English
• Title: Depiction of the Slovenian rail network with slime mould Physarum polycephalum in a two-dimensional system

Abstract

When solving complex combinatorial problems such as the formation of traffic networks, standard mathematical algorithms often fall short. As a solution, algorithms inspired by nature and biology are becoming increasingly important. One of the best-studied organisms in this field is the slime mold Physarum polycephalum. It exhibits complex decision-making, associative learning, spatial and temporal memory, can solve mazes by finding the shortest path, and can be used to make logic gates, electronic components, and transportation networks. In this thesis, we have provided a detailed description of the biological basis for these properties and investigated the use of slime mold to improve the existing transportation network using Slovenian railroad as an example. We assumed that the slime mold is capable of forming an efficient network comparable to, or even better than, the Slovenian railroad network. The experiment confirmed our hypothesis, as the slime mold formed a network with the same travel speed as the Slovenian railroad but with more than twice the resilience to failures, albeit being 29% longer. However, a direct comparison is difficult because certain constraints affecting the formation of railroad networks are not taken into account. Therefore, transferring slime mold's behavior into in silico models has potential for developing effective algorithms to solve various problems in the fields of transportation, computer science, information technology, and engineering.

• Keywords: Physarum polycephalum, Slovenian railroad, protoplasmic streaming, swarm intelligence, network formation