Comparison of Feature Selection Algorithms for Traffic Micro-Simulation Model Calibration

Year 2022, Volume: 14 Issue: 2, 752 - 761, 31.07.2022

Erdem Doğan

https://doi.org/10.29137/umagd.1096157

Abstract

A significant number of advanced microsimulation models have now been developed to perform traffic simulations, but these models contain a large number of parameters that must be calibrated to model all traffic conditions. Attempting to calibrate all of these parameters can be costly and even reduce calibration accuracy. In this study, an analysis of the effects of various Feature Selection Algorithms (FSA) on calibration accuracy is conducted and an approach is proposed to determine the appropriate FSA type. As part of the proposed approach, a model parameter set was created from SUMO's vehicle type, car following and lane change model parameters, and an experimental set was created utilizing the Latin Hyper Cube. The experiments were carried out for a 9.2 km long road section equipped with detectors capable of collecting high time resolution data. As a result, it was observed that using FSA can significantly improve the calibration performance. In addition, the calibration method proposed in this study can be functional for traffic simulation practitioners and researchers.

Keywords

Traffic, calibration, simulation, feature selection

Trafik Mikro-Simülasyon Model Kalibrasyonu için Özellik Seçim Algoritmalarının Karşılaştırılması

Abstract

Günümüzde trafik simülasyonlarını gerçekleştirmek için önemli sayıda gelişmiş mikro simülasyon modeli geliştirilmiştir, ancak bu modeller tüm trafik koşullarını modellemek için kalibre edilmesi gereken çok sayıda parametre içermektedir. Tüm bu parametreleri kalibre etmeye çalışmak maliyetli olabilir ve hatta kalibrasyon doğruluğunu azaltabilir. Bu çalışmada, çeşitli Özellik Seçim Algoritmalarının (ÖSA) kalibrasyon doğruluğu üzerindeki etkilerinin bir analizi yapılmış ve uygun ÖSA tipinin belirlenmesi için bir yaklaşım önerilmiştir. Önerilen yaklaşım kapsamında, SUMO'nun araç tipi, araç takip ve şerit değiştirme model parametrelerinden bir model parametre seti oluşturulmuş ve Latin Hiper Küpü kullanılarak deney seti oluşturulmuştur. Deneyler, yüksek zaman çözünürlüğünde veri toplama yeteneğine sahip detektörlerle donatılmış 9,2 km uzunluğundaki bir karayolu kesimi için gerçekleştirilmiştir. Sonuç olarak, ÖSA kullanımının kalibrasyon performansını önemli ölçüde iyileştirebileceği gözlemlenmiştir. Ayrıca bu çalışmada önerilen kalibrasyon yönteminin trafik simülasyonu uygulayıcıları ve araştırmacılar için fonksiyonel olacaktır.

Keywords

Trafik, kalibrasyon, simülasyon, özellik seçimi

References

• Arkatkar, S., Velmurugan, S., Puvvala, R., Ponnu, B., & Narula, S. (2016). Methodology for simulating heterogeneous traffic on expressways in developing countries: A case study in India. Transportation Letters, 8(2), 61–76. https://doi.org/10.1179/1942787515Y.0000000008
• Azam, M., Puan, O. C., Hassan, S. A., & Mashros, N. (2019). Calibration of microsimulation model for tight urban diamond interchange under heterogeneous traffic. IOP Conference Series: Materials Science and Engineering, 527(1). https://doi.org/10.1088/1757-899X/527/1/012077
• Balakrishna, R., Antoniou, C., Ben-Akiva, M., Koutsopoulos, H. N., & Wen, Y. (2007). Calibration of microscopic traffic simulation models: Methods and application. Transportation Research Record, 1999(1), 198–207. https://doi.org/10.3141/1999-21
• Ciuffo, B., Punzo, V., & Montanino, M. (2014). Global sensitivity analysis techniques to simplify the calibration of traffic simulation models. Methodology and application to the IDM car-following model. IET Intelligent Transport Systems, 8(5), 479–489.
• Essa, M., & Sayed, T. (2015). Simulated traffic conflicts: Do they accurately represent field-measured conflicts? In Transportation Research Record (Vol. 2514, pp. 48–57). https://doi.org/10.3141/2514-06
• Ge, Q., & Menendez, M. (2014). An efficient sensitivity analysis approach for computationally expensive microscopic traffic simulation models. International Journal of Transportation, 2(2), 49–64.
• Kira, K., & Rendell, L. A. (1992). A practical approach to feature selection. In Machine learning proceedings 1992 (pp. 249–256). Elsevier. Li, G.-Z., Meng, H.-H., Yang, M. Q., & Yang, J. Y. (2009). Combining support vector regression with feature selection for multivariate calibration. Neural Computing and Applications, 18(7), 813–820.
• Lopez, P. A., Behrisch, M., Bieker-Walz, L., Erdmann, J., Flötteröd, Y.-P., Hilbrich, R., Lücken, L., Rummel, J., Wagner, P., & Wießner, E. (2018). Microscopic traffic simulation using sumo. 2018 21st International Conference on Intelligent Transportation Systems (ITSC), 2575–2582.
• McKay, M. D., Beckman, R. J., & Conover, W. J. (1979). A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code. Technometrics, 21(2), 239–245. https://doi.org/10.2307/1268522
• Mirjalili, S., Mirjalili, S. M., & Lewis, A. (2014). Grey Wolf Optimizer. Advances in Engineering Software, 69, 46–61. https://doi.org/https://doi.org/10.1016/j.advengsoft.2013.12.007
• Rakha, H., Hellinga, B., Van Aerde, M., Perez, W., Aerde, M. Van, Perez, W., Van Aerde, M., & Perez, W. (1996). Systematic verification, validation and calibration of traffic simulation models. 75th Annual Meeting of the Transportation Research Board, Washington, DC.
• Rasmussen, C. E. (2003). Gaussian processes in machine learning. Summer School on Machine Learning, 63–71.
• Saeys, Y., Inza, I., & Larrañaga, P. (2007). A review of feature selection techniques in bioinformatics. Bioinformatics, 23(19), 2507–2517. https://doi.org/10.1093/bioinformatics/btm344
• Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological), 58(1), 267–288.
• Vigneau, E., & Thomas, F. (2012). Model calibration and feature selection for orange juice authentication by 1H NMR spectroscopy. Chemometrics and Intelligent Laboratory Systems, 117, 22–30.
• Yang, W., Wang, K., & Zuo, W. (2012). Neighborhood component feature selection for high-dimensional data. Journal of Computers, 7(1), 162–168. https://doi.org/10.4304/jcp.7.1.161-168