Methodology for the Characterisation of Linear Rail Transport Infrastructures with the Machine Learning Technique and Their Application in a Hyperloop Network

Abstract

The layout solution for linear rail transport infrastructure will always alternate ‘surface’ sections with ‘tunnel’ and ‘viaduct’ sections. The capital expenditure (CapEx) linked at the planning stage to this type of public asset is strongly connected to the quantity of tunnels and viaducts planned. In this context, for similar lengths, a railway line using 15% tunnels and 7% viaducts to link two cities should not have the same financial viability as one using 8% tunnels and 3% viaducts to link the same cities. The process of planning, design and construction of linear works is heavily scrutinised by public administrations in all countries, and in many cases similar standards of work are shared. Firstly, this research paper highlights the existence of hidden geometric patterns in all linear transport infrastructures worldwide. Secondly, it proposes to exploit the existence of such patterns for the benefit of planners through the computational power available today in machine learning-as-a-service (MLaaS) platforms. This article demonstrates how geometric features extracted from any succession of rectangular trapeziums in linear infrastructures can predict the quantity of kilometres in ‘surface’, ‘tunnel’ and ‘viaduct’ sections in future linear rail transport infrastructures that have not yet been built. The practical application of the proposed working methodology has made it possible to intuit the characteristics of a future Hyperloop transport network in Europe of more than 12,000 km in length.

Keywords

Infrastructures planning / Machine learning / Rail transport infrastructure / Hyperloop / Project finance / Capital expenditure

References

[1] Musk, E (2013) Hyperloop Alpha. SpaceX

[2] AASHTO (2001) A policy on geometric desing of highways and streets. pp. ix (Preface)

[3] LópezArchila MC, EcheverryArciniegas CA. Manual de DiseñoGeométrico para Carreteras. Ciencia e IngenieríaNeogranadina, 1999, 8(145–156): 155

[4] National Association of Australian State Road Authorities (1972) Guide Policy for Geometric Design o Freeways and Expressways

[5] The Indian Road Congress (1973) Recommendations about the Alignment Survey and Geometric Design of Hill Roads

[6] Secretaria de Comunicaciones y Transportes, Dirección General de Servicios Técnicos de los Estados Unidos Mexicanos (1976) Manual de Proyecto Geométrico de Carreteras

[7] Actisa Development Team (2014) Tadil software. Spain

[8] Zhu Y, Chen F, Wang Z. Spatio-temporal analysis of rail station ridership determinants in the built environment. Transportation, 2019, 46: 2269-2289

[9] Sysyn M, Gerber U, Nabochenko O. Prediction of rail contact fatigue on crossings using image processing and machine learning methods. Urban Rail Transit, 2019, 5: 123-132

[10] Ghasedi M, Sarfjoo M, Bargegol I. Prediction and analysis of the severity and number of suburban accidents using logit model, factor analysis and machine learning: a case study in a developing country. SN Appl. Sci., 2021, 3: 13

[11] Yan X, Radwan E, Abdel-Aty M. Characteristics of rear-end accidents at signalized intersections using multiple logistic regression model. Accid Anal Prev, 2005, 37(6): 983-995

[12] Ochmanski M, Modoni G, Bzowka J. Prediction of the diameter of jet grouting columns with artificial neural networks. Soils Found, 2015, 55(2): 425-436

[13] Hadj-Mabrouk H. Contribution of artificial intelligence to risk assessment of railway accidents. Urban Rail Transit, 2019, 5: 104-122

[14] Faghih-Roohi S, Hajizadeh S, Núñez A, Babuska R, De Schutter B (2016) Deep convolutional neural networks for detection of rail surface defects. In: International joint conference on neural networks (IJCNN), Vancouver, BC, 2016, pp 2584–2589

[15] You J, Wang J, Guo J. Real-time crash prediction on freeways using data mining and emerging techniques. J Mod Transport, 2017, 25: 116-123

[16] Shamshirband S, Rabczuk T, Chau K (2019) A survey of deep learning techniques: application in wind and solar energy resources

[17] Santur Y, Karaköse M, Akin E (2017) A new rail inspection method based on deep learning using laser cameras. In: International artificial intelligence and data processing symposium (IDAP), Malatya, pp 1–6 (2017)

[18] TayefehHashemi S, Ebadati OM, Kaur H. Cost estimation and prediction in construction projects: a systematic review on machine learning techniques. SN ApplSci, 2020, 2: 1703

[19] Amin M. Development of cost estimation model for residential building. Int J CivStructEng Res, 2017, 5(1): 1-4.

[20] Jiang Q. Estimation of construction project building cost by back-propagation neural network. J Eng Des Technol, 2019, 18(3): 601-609.

[21] BigML (2011) BigML, Inc. Corvallis, Oregon, USA

[22] Berthold M. KNIME (Konstanz Information Miner), 2006 Universität Konstanz

[23] Actisa SL (2020) High speed railway line from Santiago de Chile to Valparaiso and Viña del Mar. Northern Corridor or Southern Corridor?

[24] Centro Nacional de Información Geográfica, file: RT-ESPANA-PORMODOS.ZIP. 15 November 2018

[25] QGIS Development Team (2002) QGIS Noosa. Windows—GNU/Linux—Mac OS X—BSD—Android. USA: Open Source Geospatial Foundation

[26] Schneider A (2002) GPS visualizer. USA: GPS Visualizer