IN2CLOUD: A novel concept for collaborative management of big railway data

Abstract

In the EU Horizon 2020 Shift2Rail Multi-Annual Action Plan, the challenge of railway maintenance is generating knowledge from data and/or information. Therefore, we promote a novel concept called “IN2CLOUD,” which comprises three sub-concepts, to address this challenge: 1) A hybrid cloud, 2) an intelligent cloud with hybrid cloud learning, and 3) collaborative management using asset-related data acquired from the intelligent hybrid cloud. The concept is developed under the assumption that organizations want/need to learn from each other (including domain knowledge and experience) but do not want to share their raw data or information. IN2CLOUD will help the movement of railway industry systems from “local” to “global” optimization in a collaborative way. The development of cutting-edge intelligent hybrid cloud-based solutions, including information technology (IT) solutions and related methodologies, will enhance business security, economic sustainability, and decision support in the field of intelligent asset management of railway assets.

Keywords

railway / intelligent asset management / collaborative learning / big data / hybrid cloud / Bayesian

References

[1] Asplund M, Lin J (2016). Evaluating the measurement capability of a wheel profile measurement system by using GR&R. Measurement, 92: 19–27

[2] Ben-Daya M, Kumar U, Murthy D N P (2015). Introduction to Maintenance Engineering: Modelling, Optimization and Management. New York: Wiley

[3] Cai B, Liu Y H, Fan Q, Zhang Y W, Yu S L, Liu Z K, Dong X (2013). Performance evaluation of subsea BOP control systems using dynamic Bayesian networks with imperfect repair and preventive maintenance. Engineering Applications of Artificial Intelligence, 26(10): 2661–2672

[4] Creamer L (2017). The Best Asset Management Services of 2017. Retrieved from http://uk.pcmag.com/cloud-services/79270/guide/the-best-asset-management-services-of-2017, 2017-9-10

[5] Galar D, Kumar U, Karim R (2017). Big data in railway operations and maintenance. Globe Railway Review, 4: 11–14

[6] Figueres-Esteban M, Hughes P, Gulijk  C (2016). Visual analytics for text-based railway incident reports. Journal of Safety Science, 89: 72–76

[7] Karim R, Westerberg J, Galar D, Kumar U (2016). Maintenance analytics—The new know in maintenance. IFAC-PapersOnLine, 49(28): 214–219

[8] Lee J, Kao H, Yang S (2014). Service innovation and smart analytics for Industry 4.0 and big data environment. Procedia CIRP, 16: 3–8

[9] Lee J, Yang S, Lapira E, Kao H A, Yen N (2013). Methodology and framework of a cloud-based prognostics and health management system for manufacturing industry. Chemical Engineering Transactions, 33: 205–210

[10] Lin J, Asplund M (2015). Bayesian semi-parametric analysis for locomotive wheel degradation using gamma frailties. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 229(3): 237–247

[11] Lin J, Asplund M, Parida A (2014). Reliability analysis for degradation of locomotive wheels using parametric Bayesian approach. Quality and Reliability Engineering International, 30(5): 657–667

[12] Lin J, Pulido J, Asplund M (2015). Reliability analysis for preventive maintenance based on classical and Bayesian semi-parametric degradation approaches using locomotive wheel-sets as a case study. Reliability Engineering & System Safety, 134: 143–156

[13] Meeker W, Hong Y (2014). Reliability meets Big data: Opportunities and challenges. Quality Engineering, 26: 102–116

[14] Thaduri A, Galar D, Kumar U (2015). Railway assets: A potential domain for Big data analytics. Procedia Computer Science, 53: 457–467

[15] Yang S, Bagheri B, Kao H, Lee J (2015). A unified framework and platform for designing of cloud-based machine health monitoring and manufacturing systems. Journal of Manufacturing Science and Engineering, 137(4): 040914

[16] Zhang L W, Lin J, Karim R (2015). An angle-based subspace anomaly detection approach to high-dimensional data: With an application to industrial fault detection. Reliability Engineering & System Safety, 142: 482–497

[17] Zhang L W, Lin J, Karim R (2017). Sliding window-based fault detection from high-dimensional data streams. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 47(2): 289–303

Acknowledgements

The authors thank Luleå Railway Research Centre (Järnvägstekniskt Centrum, Sweden) and Swedish Transport Administration (Trafikverket) for initiating the research study and providing financial support. This work was also partly supported by NSFC under a key project (Grand No. 71731008). We thank researchers involved in discussions of the concept, including from Luleå Tekniska Universitet (Sweden), Universidad Politecnica De Madrid (Spain), Technische Universiteit Delft (The netherlands), The University of Huddersfield (UK), Schneider Electric Espana SA(Spain), Infranord (Sweden), Integrasys SA (Spain), Digital Rail Limited (UK), and Limmat M&M Sociedad Limitada (Spain).

RIGHTS & PERMISSIONS

2017 The Author(s) 2017. Published by Higher Education Press. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0)