prikaz prve stranice dokumenta Analysis of neural network responses in calibration of microsimulation traffic model
  Download
PDF 1.07 MB
Scientific paper - Review paper
Analysis of neural network responses in calibration of microsimulation traffic model
E-GFOS, 6 (2015), 10; 67-76. https://doi.org/10.13167/2015.10.8
Ištoka Otković, Irena; Varevac, Damir; Šraml, Matjaž

Cite this document

Ištoka Otković, I., Varevac, D. & Šraml, M. (2015). Analysis of neural network responses in calibration of microsimulation traffic model. E-GFOS, 6. (10), 67-76. doi: 10.13167/2015.10.8

Ištoka Otković, Irena, et al. "Analysis of neural network responses in calibration of microsimulation traffic model." E-GFOS, vol. 6, no. 10, 2015, pp. 67-76. https://doi.org/10.13167/2015.10.8

Ištoka Otković, Irena, Damir Varevac and Matjaž Šraml. "Analysis of neural network responses in calibration of microsimulation traffic model." E-GFOS 6, no. 10 (2015): 67-76. https://doi.org/10.13167/2015.10.8

Ištoka Otković, I., Varevac, D. and Šraml, M. (2015) 'Analysis of neural network responses in calibration of microsimulation traffic model', E-GFOS, 6(10), pp. 67-76. doi: 10.13167/2015.10.8

Ištoka Otković I, Varevac D, Šraml M. Analysis of neural network responses in calibration of microsimulation traffic model. E-GFOS [Internet]. 2015 July [cited 2024 December 26];6(10):67-76. doi: 10.13167/2015.10.8

I. Ištoka Otković, D. Varevac and M. Šraml, "Analysis of neural network responses in calibration of microsimulation traffic model", E-GFOS, vol. 6, no. 10, pp. 67-76, July 2015. [Online]. Available at: https://urn.nsk.hr/urn:nbn:hr:133:171896. [Accessed: 26 December 2024]

Please login to the repository to save this object to your list.
Metadata
Title (english)Analysis of neural network responses in calibration of microsimulation traffic model
Title (croatian)Analiza odziva neuronskih mreža u postupku kalibracije mikrosimulacijskog prometnog modela
AuthorIrena Ištoka Otković
AuthorDamir Varevac
AuthorMatjaž Šraml
Author's institutionJosip Juraj Strossmayer University of Osijek
Faculty of Civil Engineering and Architecture Osijek
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Civil Engineering
Abstract (english)
Microsimulation models are frequently used in traffic analysis. Various optimization methods are used in calibration, and the one method that has shown success is neural networks. This paper shows the responses of neural networks during calibration of a microsimulation traffic model. We analyzed two calibration methods by applying neural networks and comparing their neural network learning (according to their achieved correlation and the mean error of prediction) and their generalization ability (comparison of generalization results was analyzed in two steps). The best correlation between the microsimulation results and neural network prediction was 88.3%, achieved for the traveling time prediction, on which the first calibration method is based.
Abstract (croatian)
Učestala je primjena mikrosimulacijskih modela u prometnim analizama, a realnost dobivenih rezultata simulacije u funkciji je uspješnosti postupka kalibracije. Različite metode optimiranja primjenjuju se u postupku kalibracije, a jedna od metoda koja se pokazala uspješnom u postupku kalibracije je metoda koja primjenjuje neuralne mreže. U ovom radu prikazana je analiza odziva neuralnih mreža u kalibraciji mikrosimulacijskog prometnog modela. U radu su razmatrane dvije metode kalibracije primjenom neuralnih mreža koje su uspoređene prema rezultatima učenja neuralnih mreža (prema kriterijima postignute korelacije i srednje pogreške predikcije) i prema sposobnosti generalizacije (usporedba rezultata generalizacije analizirana je u dva koraka). Najbolja korelacija između rezultata mikrosimulacije i predikcije neuralne mreže (88,3%) dobivena je za parametar vrijeme putovanja, a na navedenom parametru bazirana je prva metoda kalibracije.
Keywords (english)
Keywords (croatian)
Languageenglish
Publication typeScientific paper - Review paper
Publication statusPublished
Peer reviewPeer review
Publication versionPublished version
Journal titleE-GFOS
Numberingvol. 6, no. 10, pp. 67-76
e-ISSN1847-8948
DOIhttps://doi.org/10.13167/2015.10.8
URN:NBNurn:nbn:hr:133:171896
Publication2015-07
Document URLhttp://hrcak.srce.hr/140696
Type of resourceText
Access conditionsOpen access
Terms of useLicence Attribution-NonCommercial-NoDerivs 3.0 Unported (CC BY-NC-ND 3.0) icon
Created on2018-01-11 13:25:17