MFFNN and GRNN Models for Prediction of Energy Equivalent Speed Values of Involvements in Traffic Accidents / Trafik Kazalarında tutulumunun Enerji Eşdeğer Hız Değerleri Tahmininde MFFNN ve GRNN Modelleri

Year 2015, Volume: 4 Issue: 2, 102 - 109, 11.01.2015

Ali Yılmaz , Cigdem Aci Kadir Aydin

https://doi.org/10.18245/ijaet.78159

Cited By: 1

Abstract

Accident reconstruction is a scientific study field that depends on analysis, research and drawing. Scientific reconstruction of related traffic accident on computer eliminates making decisions depending on initiative or experience of the expert and yields impartial decisions and evidences especially on events like matter for the courts or forensic investigations. In this study, data collected from accident scene (police reports, skid marks, deformation situation of involvements, crush depth etc.) were inserted properly into the software called “vCrash” which is able to simulate the accident scene in 2D and 3D. Then, 784 parameters, related to calculating Energy Equivalent Speed (EES) with a prediction error, were prepared according to several accidents. These parameters were also used as teaching data for the Multi-layer Feed Forward Neural Network (MFFNN) and Generalized Regression Neural Network (GRNN) models in order to predict EES values of involvements, which give idea about severity and dissipation of deformation energy corresponding to the observed vehicle residual crush, without requirement of performing simulation for probable accidents in future. Using 10-fold cross validation on the dataset, standard error of estimates (SEE) and multiple correlation coefficients (R)of both models are calculated. The GRNN-based model yields lower SEE whereas the MFFNN-based model yields higher R.

Özet: Kaza yeniden analiz, araştırma ve çizim bağlıdır bilimsel bir çalışma alanıdır. Bilgisayardaki ilgili trafik kazası Bilimsel yeniden inisiyatifi veya bilirkişinin deneyimine bağlı olarak kararlar ortadan kaldırır ve özellikle mahkemeler veya adli soruşturma için madde gibi olaylara tarafsız kararlar ve delilleri verir. Bu çalışmada, veriler kaza sahnesi (polis raporlarında, kızak işaretleri, tutulumunun deformasyon durumuna vs. ezilme derinliği) 2D ve 3D kaza sahnesini taklit edebilen "vCrash" olarak adlandırılan yazılım içine düzgün bir şekilde yerleştirildi toplanan. Daha sonra, tahmin hatası Enerji eşdeğer Speed ​​(EES) hesaplanması ile ilgili 784 parametreleri, çeşitli kazalar göre hazırlandı. Bu parametreler İleri Sinir Ağı (MFFNN) ve Genelleştirilmiş Regresyon Sinir Ağı (GRYSA) yapılan ÇO şiddeti ve karşılık gelen deformasyon enerjisinin dağılımı konusunda fikir vermek bulguların EES değerlerini tahmin etmek için modeller Yem Çok katmanlı öğretim veri olarak kullanıldı Gelecekte muhtemel kazalara karşı simülasyon gerçekleştirme gereksinimi olmadan gözlenen araç artık ezmek. Veri kümesi üzerinde 10 kat çapraz doğrulama kullanarak, tahminler (GDA) ve her iki model çoklu korelasyon katsayılarının (R) standart hatası hesaplanır. MFFNN-tabanlı model yüksek R. verir, oysa GRNN-tabanlı model alt SEE verir.

Keywords

Accident reconstruction, EES, artificial neural network, MFFNN, GRNN / Kaza rekonstrüksiyonu, EES, yapay sinir ağları, MFFNN, GRNN

References

• S. Miaou and H. Lum, "Modeling vehicle, accidents and highway geometric design relationships,"Accident Analysis and Prevention, vol. 25, pp. 689–709, 1993.
• C. Murray and A. Lopez (Eds.), The Global Burden of Disease, Harvard, Cambridge, 1996.
• Traffic Statistical Bulletin, Republic of Turkey Ministry of Internal Affairs, 2006.
• Virtual CRASH (vCrash), Traffic Accident Reconstruction Tool, College of Industrial Engineers and Surveyors, Graduate Industrial Province of Naples, Italy, 2014.
• E. Tomasch, "Accident Reconstruction Guidelines," Graz University of Technology, A.T., 2004.
• Generalized Extreme Value (GEV) Available: http://en.wikipedia.org/wiki/Generalized_extreme_value_distribution, 05/09/2011.
• E. Alpaydın, Introduction to Machine Learning, 2nd ed. MIT press, London, 2010.
• E. I. Zayid and M. F. Akay “Predicting the performance measures of a message-passing multiprocessor architecture using artificial neural networks,” Neural Computing and Applications, vol. 23, pp. 2481-91, 2013.
• M. Khashei, A. Z. Hamadani and B. Bijari, “A novel hybrid classification model of artificial neural networks and multiple linear regression models,” Expert Syst App.,vol. 39, pp. 2606–20, 2010.
• E. J. Jackson, A user’s guide to principal components. Wiley press, 2001.
• H. Demuth and M. Beale, Neural network toolbox user’s guide, The MathWorks,Inc., 1997.
• M. Firat and M. Gungor, “Generalized regression neural networks and feed forward neural networks for prediction of scour depth around bridge piers,” Adv. Eng. Softw., vol. 40, pp. 731–37, 2009.