AI Factory /RAILWAY The Evolution (AIF/RE3I)

The project aims to enhance, expand, elaborate, and innovate, denoted as E3I, findings and results from Phase I of the project.

Goal

‘AI Factory /RAILWAY – The Evolution (AIF/RE3I)', i.e. AIF/R Phase II, aims to enhance, expand, elaborate, and innovate, denoted as E3I, findings and results from Phase I of the project. In the first phase of the project, the focus was to initiate and establish some fundamental components in the data lab. In the first phase, we developed and established a software architecture and corresponding digital infrastructure to enable data sharing for the Swedish railway sector.

For Phase II, i.e. AIF/RE3I, we will focus on enablement of data and model sharing required for the establishment of AI-based solutions. Further, this phase of the project will revise and enhance the previously established architecture and infrastructure in the lab to be prepared for the emerging era of Quantum Computing.

AIF/RE3I platform is aimed to be used for sharing of data and model related to: a) railway infrastructure; b) railway rolling stock; c) railway organisations; and d) climate, as depicted in Figure.

Cybersecurity

AIF/RE3I will further provide a cybersecurity platform including authentication mechanisms at various levels. Cybersecurity platform of AIF/RE3I will manage access to data based on data openness characteristic, such as 1) fully open data (requires no authentication); semi-open data (requires weak authentication), and no-open data (strong authentication).

API

The data services in AIF/RE3I provide various Application Programming interfaces (API) to facilitate dynamic and context-based integration between back-ends, middleware, and front-ends services. some examples of these are Logic Apps, Services Bus, Event Grid, REST etc.

Existing commodities

AIF/RE3I will extend the existing digital infrastructure at LTU, i.e. the eMaintenanceLAB, for provisioning of digital services. The maintenance lab is a dynamic, open cloud/fog/edge-based technology platform which is already being utilised by the project's primary and secondary stakeholders and also other railway stakeholders in Sweden and internationally. The lab already manages railway datasets from 2012 and forward. The existing historical data has enormous scientific and practical value. Currently, the datasets are for example used for the model learning and for the assessment of model precision and model accuracy. The existing dataset will be part of the commodities in AIF/RE3I.

Agility

AIF/RE3I will be executed based on the agile methodology. AIF/RE3I will implement Scrum for its process management.

DevOps

AIF/RE3I will implement an integrated development-and-operation methodology for the AI-driven solutions and services.