System Structure

1. Data Acquisition, Transformation and Evaluation: Big Data

2. IOT: Transferring Data to Cloud in Real Time during Operation

3. Artificial Intelligence (AI): Analyzing Data through Machine Learning

4. Total Maintenance Management System:

- Maintenance Workshop: big data analysis, predictive maintenance, and managing work order and assets in real time in whole work place in real time. For doing so the system is fully connected with MMIS.

- Operation Control Center: monitoring and recording all the train information while train is moving. All data is monitored in real time while train is in operation. Real time monitored data information is including exact Location within the Map, Speed, head side where the train is heading, Temperature (in & out), Electrical Systems, Doors Status (Open & Close), Mileage of Train as well as CCTV (all internal and external cameras)

Continuous data collection is processed from various systems and subsystems in trains, enabling monitoring of mechanical and electrical conditions, operational efficiency and many other performance indicators on a real time basis.

All the collected data is all recorded through on-board train computer to the server. All collected and recorded data goes through the VISION IT’s program for monitoring, controlling, operation, operation safety and maintenance purposes. Railway predictive maintenance or RPM will be performed by following approach:

• Data-driven: driven by asset digitalization, maintenance engineering has an increasing volume of multisource and heterogeneous data. These data are analyzable in their entirety through a holistic approach, applying artificial intelligence techniques, machine learning and predictive analytics.

A predictive diagnostic system will be able connected to air pressure, currents, velocity, voltages and so forth. For this reason, a system of smart sensors directly and digitally connected with the Train Control and Management System (TCMS).

1. Data Acquisition, Transformation and Evaluation: Big Data

Sensors will create both exogeneous data that measures external factors, such as the weather or line conditions, and endogenous data synthesized from within the train’s subsystems. Once the data is created, the flow required to convert raw data into useful information.

Extract the Right Data

The following list is a sample of potential functions and components that will be monitored and controlled through on-board computer which is connected with Train Main Computer. Whole data from train main computer is flowed into the on-board computer, then it is sent to the cloud through the internet connectivity (4G/5G):

* Axles.

* Bogies.

* Brakes. * Door systems.

* Filters.

* Flat wheel (degradation of the steel wheel).

* Harmful currents or voltages.

* Pantographs.

* Rotating parts.

* Water and air pressure.

* Wheel bearings.

Data received through already installed sensors in the trains, and additionally with following sensors:

^•¨^¨Sound: vibrations generate acoustics. Measuring the acoustics level through an electromagnetic microphone can be an effective means of detecting vibrations.

^•¨^¨Temperature: increased friction leads to an increase of temperature of the monitored asset. Thermistors or other temperature sensors can detect these variations.

^¨¨^•Vibrations: Shock pulse measurement, envelope technique and acoustic emissions are a few different techniques used to measure vibrations. Moreover, several properties of the carriages can be analyzed using accelerometers installed along the train.

Data Acquisition, Transformation and Evaluation

Data acquisition is gathered and measured from heterogeneous sources (such as the different trains’ subsystems) and related targeted variables in an established, systematic trend. The acquisition process will be processed through each train computer system where all the data is collected from each and every sensor. System structure will be as follows:

1. Collect and store the data produced by IP sensors and other external sources.

2. Perform a first analysis of the data in real time.

3. Share through wireless connectivity all the data acquired during the trip to the cloud. This data is then consolidated and processed through data transformation tools.

Through data evaluation, we analyze data and search for patterns that predict potential faults through advanced algorithms, expertise, domain know-how and best practices. For example, patterns might predict the circumstances in which a traction drive, electronic door motor or a wheel set will fail.

The data evaluation phase deals with both short-term and long-term analysis. The short-term analysis is performed on board and provides real-time information to the driver about the running trip. The long-term analysis provides an end-to-end view of the maintenance framework to make it more efficient, identify new patterns, and improve decision- making and future planning.

2. IOT: Transferring Data to Cloud in Real Time during Operation

Multiple data types from many sources (such as engine variables, bogie sensors, GPS position within the line, and atmospheric data) are ingested into the data lake through on board computer connected with train main computer. Then the data flows into the cloud (server which is prepared and installed for this purpose) by 4G and/or 5G internet connectivity.

3. Artificial Intelligence (AI) and Predictive Maintenance It is possible to use several capabilities and technologies to achieve these results by gaining insights from data. The following list will be used as potential techniques:

* Descriptive analytics techniques provide simple summaries and observations about the data.

*Data mining analyzes large quantities of data to extract previously unknown interesting patterns and dependencies.

* Machine learning enables the software to learn from the data and predict accordingly. For example, when a train’s subsystem fails, several factors come into play. The next time those factors are evident, the software will predict the failure.

* Simulation enables what-if scenarios for specific assets and/or processes; for example, how running specific components for a certain period of time impacts the likelihood of failure.

^•¨^¨Text mining is a subset of data mining, where data is composed by natural language texts. It enables the understanding of and alignment between computer and human languages.

^¨¨^•Predictive analytics uses machine learning and data mining techniques to predict future outcomes.

* Prescriptive analytics adds a decision-management framework to the predictive analytics outcomes to align and optimize decisions according to analytics and organizational domain knowledge. The goal to achieve is not just to identity when an asset fails, but also to suggest actions, and to show the implications of each decision.

Value-Added Outcomes

The value-added outcomes from Artificial Intelligence solution to the Predictive Maintenance are as follows:

* Predicting when, subject to specific border conditions, a part will fail, and which maintenance actions are required.

* Planning the maintenance actions in advance, allowing a just-in-time sourcing for replacement of parts, and optimizing procurement and inventory.

* Identifying systems that might be affected by potential design problems based on their history of poor performance.

* Identifying a track's problem when a train goes through a specific point in line, treating the vehicle like a sensor on wheels.

By understanding the reasons behind various failure patterns and categorizing them into various action buckets, it is possible to address both short-term and long-term objectives.

4. Total Maintenance Management System

All data will be managed through single MS SQL data base. After all data processed and analyzed, different teams based on need will be informed automatically. For example, if there is any failure, work order will automatically be issued to the corrective maintenance team while spare part need related information separately send to the logistic team.