This paper summarizes the execution and results of the research project "Reality Lab Digital Railway". The goal of the project is to open up the Swedish transport administration's (Trafikverket's) core business within railway for development and demonstration of digital information logistic solutions that are tested by primary actors and end users. The results of the project can be summarized in the four areas of regulations, organization, roles and technologies that contribute to Trafikverket's ability to achieve the project goals. Examples of results are methodologies and tools for establishing and actively managing a more dynamic regulatory framework. This framework clarifies the need for innovation as well as the specification of the application of digital technologies to strengthen the railway sector's capability. To support collaboration between railway actors during test and demonstration, physical and digital assets as well as ICT-solutions have also been established. The project has also produced proposals for agreements regarding information management between Trafikverket and external parties.

One major prerequisite for an effective implementation and innovation process is the enablement and provision of a collaborative environment. A common area for multi-organisational collaboration together with a technology platform, enabling data sharing and Big Data Analytics, has been developed called ‘Testbed Railway’ with a corresponding framework ‘Railway 4.0’. Testbed Railway can be used to strengthen the railway industry's adaptability and competitiveness by developing and providing a testbed for research and innovation in the rail industry, nationally and internationally.

Globally, railway is experiencing a major technology transformation (or paradigm shift), triggered by the enhanced utilisation of digital technology. This technological transformation affects not only the technical systems, i.e. railway infrastructure and rolling stock, but also regulations, organisations, processes,and individuals. Hence, hardware, software, but also liveware (i.e. humans) are affected. Today, the digitalisation of railway is characterised by digital services. There are also a range of challenges, e.g. data acquisition,transformation, modelling, processing, visualisation, safety, security, quality, and information assurance. To deal with these challenges, the railway industry needs to define strategies, which enable a smooth transformation of the existing configuration to a digitalised system. Digital railway requires a holistic change management approach based on system-of-systems thinking and a set of appropriate technologies and methodologies. The railway digitalisation strategy should be based on systematic risk management that address aspects of, e.g., information security, traffic safety and project risk. In addition, managing changes for a digitalised railway effectively and efficiently also requires a framework for aspects such as needs finding, requirement identification, and impact of changes for individual, teams and organisation. In this work a major case studywithin the ePilot, has been performed in context of the operation and maintenance processes of the Swedish railway. Therefore, this paper aims to propose a framework for implementing innovations and driving change in a digitalised railway.

The accuracy of traditional anomaly detection techniques implemented on full-dimensional spaces degrades significantly as dimensionality increases, thereby hampering many real-world applications. This work proposes an approach to selecting meaningful feature subspace and conducting anomaly detection in the corresponding subspace projection. The aim is to maintain the detection accuracy in high-dimensional circumstances. The suggested approach assesses the angle between all pairs of two lines for one specific anomaly candidate: the first line is connected by the relevant data point and the center of its adjacent points; the other line is one of the axis-parallel lines. Those dimensions which have a relatively small angle with the first line are then chosen to constitute the axis-parallel subspace for the candidate. Next, a normalized Mahalanobis distance is introduced to measure the local outlier-ness of an object in the subspace projection. To comprehensively compare the proposed algorithm with several existing anomaly detection techniques, we constructed artificial datasets with various high-dimensional settings and found the algorithm displayed superior accuracy. A further experiment on an industrial dataset demonstrated the applicability of the proposed algorithm in fault detection tasks and highlighted another of its merits, namely, to provide preliminary interpretation of abnormality through feature ordering in relevant subspaces.

The Swedish ore line, began in 1898 and was completed and operational by 1902. The line was later electrified and remains so to this day. A study on wheel-rail interaction, to optimize wheel and rail profiles on the Malmbanan and Ofoten lines states that there is a need to develop and apply limit values for the allowable length of single and multiple cracks on wheels. Herein, all stakeholders in the railway organization have to be involved. Provisioning of an integrated decision support system for track and vehicle contributes to increased efficiency and improved effectiveness of the maintenance process, which in turn enables achievement of business excellence. However, an integrated decision support process for maintenance is highly dependent on appropriate information logistics which enables information provisioning to various maintenance stakeholders. One emerging approach for development of eMaintenance solutions is utilization of cloud-based technologies. Cloud-based eMaintenance solutions promises smooth information logistics for maintenance decision support. Hence, the purpose of this paper is to propose an approach for condition-based maintenance decision-making relating to railway vehicle wheels, based on a cloud-based eMaintenance solution. The paper also demonstrates how the proposed approach can be implemented for estimation of Remaining Useful Life (RUL) of railway vehicle wheels.

Purpose – The purpose of this paper is to explore the main ontologies related to eMaintenance solutions and to study their application area. The advantages of using these ontologies to improve and control data quality will be investigated.

Design/methodology/approach – A literature study has been done to explore the eMaintenance ontologies in the different areas. These ontologies are mainly related to content structure and communication interface. Then, ontologies will be linked to each step of the data production process in maintenance.

Findings – The findings suggest that eMaintenance ontologies can help to produce a high quality data in maintenance. The suggested maintenance data production process may help to control data quality. Using these ontologies in every step of the process may help to provide management tools to provide high quality data.

Research limitations/implications – Based on this study, it can be concluded that further research could broaden the investigation to identify more eMaintenance ontologies. Moreover, studying these ontologies in more technical details may help to increase the understandability and the use of these standards.

Practical implications – It has been concluded in this study that applying eMaintenance ontologies by companies needs additional cost and time. Also the lack or the ineffective use of eMaintenance tools in many enterprises is one of the limitations for using these ontologies.

Originality/value – Investigating eMaintenance ontologies and connecting them to maintenance data production is important to control and manage the data quality in maintenance.

The term eMaintenance emerged in the early 2000s and has become a popular topic in maintenance related literature because of ongoing technological improvements. This paper uses a recent approach, i.e. cloud-based technology, to provide an eMaintenance solution for online time data analysis to make effective and efficient railway maintenance decisions. Due to increased traffic, the Swedish railway sector needs to optimise maintenance, using predictive maintenance to a much higher degree so that unplanned breakdowns and downtime are drastically reduced. The paper shows how research within the railway sector is developing eMaintenance solutions using the cloud and web-based applications for improved condition monitoring, better maintenance and increased uptime. In the proposed solution, data are acquired from railway measurement stations and sent to the eMaintenance cloud, where they are filtered, fused, integrated and analysed to assist maintenance decisions. The paper provides a concept for a web-based eMaintenance solution to assist railway maintenance stakeholders make fact-based decisions and develop more efficient and economically sound maintenance policies.

Radio Frequency Identification (RFID) helps automatic identification of objects using radio waves. This is not a new technology instead decades old and has been used during the World War II, when it was used by allied ground forces to track German bombers. It is a technology for wireless communication between a reader and a transponder/tag. This technology permits the transfer of data to the most diverse objects without the need for physical contact and uses intelligent barcodes to track items and have been successfully applied in military, security, healthcare, real time location tracking, vehicle identification and other areas. This paper is based on applications of radio frequency identification technology with eMaintenance cloud for railway system to analyze and visualize data of trains for the cost effective maintenance planning. Further, cloud computing is an emerging research area that can be utilised for acquiring an effective and efficient information logistics. Specifically, the widespread use of RFID will enable wagons to be tracked leading to better resource utilization, lower freight costs, and better maintenance. Therefore, it helps to provide greater control of the train carriages, making it easier to plan resources. However, RFID is a powerful tool that can help to improve industry proficiency, implementing this technology is not easy. Furthermore, operating RFID systems can be a challenging process. Thus, this paper is based on the application of RFID in the context of railway operation.

Increased environmental awareness in the industry combined with the globalized market economy makes increasing demands for sustainable and efficient resource utilization. In this context, maintenance plays a critical role by linking business objectives to the strategic and operational activities aimed at retaining the system’s availability performance, cost-efficiency and sustainability. Performing maintenance effectively and efficiently requires corresponding infrastructure for decision-support provided through eMaintenance solutions. A proper eMaintenance solution needs to provide services for data acquisition, data processing, data aggregation, data analysis, data visualization, context-sensing etc. To en Quality of Service (QoS) in eMaintenance solutions, the performance of both system-of-interest, enabling systems and related processes have to be measured and managed. However, the QoS has to be considered on all aggregation levels and encompass the aspects of Content Quality (CQ), Data Quality (DQ) and Information Quality (IQ). Hence, the purpose of this paper is to study and describe some aspects of DQ in eMaintenance related to process industry in northern Europe

Reliability, availability and maintainability analysis is one of the most important tools for measuring system performance. The performance of a traction power supply system (TPSS) can be measured using the data collected from frequency converters, as these converters constitute the main part of the TPSS. The quality of the collected data should be good enough to provide the correct and complete information necessary for assessment of frequency converter performance. Many methods can be used to assess the performance of converters such as neural networks, fuzzy logic and standards. The IEEE 762 Standard offers a methodology that can provide key performance indicators for power generation units. This standard has been chosen for its widespread acceptance and applicability. To be able to evaluate a converter’s performance, IEEE 762 indexes should be calculated using data such as the downtime, reserve shutdown hours and service hours. Therefore, the purpose of this study is to assess the performance of the Swedish TPSS frequency converters using IEEE 762, and to assess the quality of data by inspecting their compatibility with this standard. In this study, an application has been developed to generate the missing information and to calculate the indexes provided by the standard, in order to evaluate the power converters’ performance. A case with sample data is also discussed in this paper.