The railway industry is a sector where asset maintenance is paramount in ensuring passenger safety and service continuity. In this context, the application of the blockchain paradigm can represent an innovative and efective solution for improving railway maintenance eficiency. Specifically, blockchain can be used to record every event a railway asset encounters during its useful life cycle, including both the anomalies it has experienced and the maintenance cycles it has undergone. Through the use of technologies such as NFT tokens, IPFS, and IoT, it is possible to create a transparent, verifiable, and immutable railway maintenance management system. The assured on-chain recording of every event, combined with incentive mechanisms based on the distribution of native blockchain tokens, creates a virtuous cycle where maintainers are incentivized to complete all anticipated maintenance cycles to keep assets in optimal conditions. Furthermore, a secondary trade royalty system can be established by using NFT protocols, thereby creating additional incentive for the creation and maintenance of quality assets. This paper will examine the technologies used and methodologies adopted to implement this blockchain-based railway maintenance management system, with the aim of highlighting the main challenges and opportunities related to its implementation. Some case studies will also be analyzed, and potential future prospects for the use of blockchain in this sector will be discussed.
In recent years, we have seen an exponential increase in
new technologies that have greatly impacted our daily lives.
These services are constantly evolving and aim to provide
more and better performance and functionality in fields such
as telecommunications[
Specifically, two case studies were analysed regarding the use of blockchain in managing the activities of corrective and preventive maintenance of railway assets, and in the geographical traceability of Rolling Stock.
Our work, in reference to the following papers under
review [
In summary, the added value of our paper lies in the speciifcity of the blockchain application in railway maintenance management, examining specific case studies and the challenges associated with technology implementation. Meanwhile, other publications focus on broader sectors such as logistics, supply chain, drug management, protection of digital assets, and intelligent transport without concentrating on a specific case.
The work described lays the groundwork for a "readyto-use" solution, providing an efective methodology aimed at improving the safety, eficiency, and quality of railway maintenance
The paper is structured as follows: In Chapter 2, the technologies used in the study are described, introducing a multidisciplinary integration approach aimed at ensuring the reliability of the information supply chain. There is also a mention of the solution that will be demonstrated in the research. In Chapter 3, two case studies are presented for the digitisation of railway assets, one static and one dynamic. Additionally, the decentralised RailChain application for visualising digital asset information is described, including the "Live" status from various NFT HeliumAdapter sources, and the infrastructure of the experimental model. Chapter 4 is dedicated to the use of dynamic NFTs, discussing their benefits and limitations in the proposed solution, and highlighting the integration of IoT and blockchain technologies for the indelible recording of the usage status of assets and the challenges of implementation. A gradual approach to introducing blockchain technology by sector component manufacturers is also proposed. Finally, in Chapter 5, the conclusions are drawn and potential future developments are explored.
The methodology used to apply the blockchain paradigm to railway maintenance management requires a multidisciplinary approach, which involves the integration of various technologies to ensure the reliability of the supply chain that manages the association of information automatically collected from the field and the information recorded by the maintenance team to railway assets. The proposed approach envisages the integrated use of three technologies: • IoT: the Internet of Things (IoT) is a technology that allows for the connection and management of common use objects via a network, utilising sensors, actuators, and embedded devices. This is done with the aim of collecting and analysing data to support decision-making and control activities. The Helium network was used as the communication infrastructure, which is a low-power wireless communication network based on LoRaWAN technology. This allows for the creation of a network of IoT devices. • NFT: an NFT, which stands for "Non-Fungible Token", is a unique and non-interchangeable type of cryptographic token that is used to represent the ownership of digital and physical assets. Each NFT is unique and cannot be replaced with another. This 1–9 means that NFTs can be used to certify the authenticity and exclusivity of an asset • IPFS: the InterPlanetary File System is a peer-topeer distributed file system that allows for the permanent and secure storage and sharing of data, ensuring data traceability, verifiability, and immutability.
We begin by broadly outlining the solution, the feasibility
of which we intend to demonstrate.
2.1. IoT
To address one of the challenges of "live" data collection
related to the status of assets (mainly diagnostic alarms
and information related to the usage metrics of the asset
in question) located in remote locations or pertaining to
rolling stock distributed over a wide geographical area, it
is necessary to identify a communication infrastructure to
complement and integrate with more traditional
communication networks, such as Terrestrial Networks (TNs) and
Non-Terrestrial Network (NTNs) [
The Helium Protocol [
Another important feature of the Helium protocol is its
use of a token-based incentive system, which rewards users
who participate in the network by deploying and
maintaining it. This helps ensure that the network remains
decentralised and robust, even as the number of devices connected
to it increases. The Helium protocol, through its Proof of
Coverage [
From a practical perspective, we can say that the essential elements that constitute an NFT are as follows: • Token Identifier • Contract Address • Owner of the Token • Metadata associated with the Token Metadata defines the actual content of NFTs with features, attributes, and addressing related to external content. Metadata is essentially a JSON file that contains at least the following information: a description, a link to the digital media representing the NFT and characteristics.
The proposed solution is implemented on the Ethereum
blockchain [
The type of NFT to be used for modelling must take into consideration that an asset is not a static entity that remains unchanged over time, but on the contrary, it has a history that must be traced. It should be possible to store every maintenance intervention it will be subjected to and every potential move from when it is installed to when it is decommissioned. Therefore, its digital equivalent cannot be a simple static digital ownership certificate identified by a traditional ERC-721 token, but for its modelling, a dynamic NFT must be used. The smart contract that will implement its minting should therefore be written in such a way as to make public the methods that allow the modification of metadata linked to the generated token. The metadata will thus be structured to contain all relevant information related to the modelled assets and their evolution over time, such as: identification code, installation site, installation date, user and maintenance manuals, maintenance reports, links between assets and articles up to, where applicable, the software package installed on the asset or the information related to its configuration. This type of metadata will be manually modified over time by the maintenance managers, reflecting the changes in the asset’s destination and the maintenance activities to which it will be subjected. However, there is another category of information that we want to include in the digital model, which is the aforementioned "live" information. The smart contract will be written in such a way as to implement the reading of this information through LoRaWAN devices interfaced with the field through digital, serial or TCP signals. Since this information is external to the blockchain, this operation will be carried out through an Oracle. To avoid overlapping and synchronising the management of this automatically acquired information with the information that is manually updated by the operators within the same metadata set, an ad hoc smart contract will be implemented that will manage a family of NFTs to model this functionality by associating a specific asset with a specific Helium device; these NFTs will be identified hereafter as HeliumAdapter. Each digital asset that wishes to associate "live" information will therefore have within its metadata the reference to one or more HeliumAdapters. In other words, the referenced adapters will be the "children" of the asset in question.
The design of metadata is a crucial aspect for the representation of relationships that link Assets and Items. These relationships can be divided into two categories: identity relationships and functional hierarchical relationships. The identity relationship expresses the concept that an asset is nothing more than an instance of a specific item located in a particular location. This means that multiple assets can be linked by an identity relationship to the same item, for example, ten cameras of the same type A installed in diferent locations: "CAM-001", ..., "CAM-010" will be ten diferent assets linked to the same item "Type A Camera". Moving on to functional hierarchical relationships, these are of three diferent types (Fig.1): 1. Each asset can be functionally or physically linked to other assets, defined as Child. 2. Each asset can be linked to a list of items that represent spare parts of the same. 3. To each asset, one or more HeliumAdapters (HA), capable of providing "Live" information coming from the field, can be associated. The link can be related to the asset itself, its child assets, or some of its constituent parts.
To complete the picture, we can add a further modelling
rule: all the above-mentioned relationships will be iterable
at the Asset level.
2.3. IPFS
Given the substantial size of the information to be saved
within the metadata, which is conditioned by the
continuous growth throughout the useful life of the asset, storing
the same on-chain was excluded a priori, and it was
chosen to store them of-chain , registering on the blockchain
only the reference URI to the data. For this purpose, the
IPFS system (Inter Planetary File System) [
As for the metadata of the HeliumAdapter, a diametrically opposite design solution has been applied. For representational clarity, the image associated with the HeliumAdapter NFT will reflect the status of alarms and detected variables. To ensure the necessary flexibility, an alternative type of NFT is used, in which data is completely stored on-chain directly in the smart contract. Instead of a link, the tokenURI will contain a JSON-encoded data that includes extremely compact-sized graphical data in SVG format. The SVG vector graphic image is programmatically generated, encoded, and returned by the contract. Inclusion of the states detected by the LoRaWAN devices via the Oracles in the SVG makes the HeliumAdapter responsive to changes in both blockchain data and LoRaWAN devices.
For our paper, we will focus on two case studies. The first case concerns the digitisation of a high-criticality static railway asset, while the second case involves a dynamic asset, namely a Rolling stock that could be located over a wide geographical area.
Regarding the static asset, we will consider a structured asset consisting of a simplified version of a component subject to a high level of criticality that must meet very restrictive safety criteria: a cabinet that is part of the ventilation control system in a railway tunnel. This system is essential to ensure the safety of passengers and staf in case of ifre. The ventilation system is used to dissipate smoke and hot air from the tunnel, in addition to providing fresh air during evacuation operations. The registration of regular preventative maintenance operations, of any anomalies that have occurred, and of subsequent corrective maintenance 1–9 interventions is crucial to ensure that the system functions correctly in emergency situations. The asset is also associated with two NFT HeliumAdapters that model, respectively, a Helium device that implements Modbus RTU communication with the control Programmable Logic Controller (PLC) for the detection of anomalous conditions (for example, a Jet Fan inverter anomaly) and a Helium device that will detect anomalies related to the operation of the electromechanical components of the system (for example, intervention of auxiliary power circuit breaker) through digital inputs.
As for the dynamic asset, a Rolling stock simulation is modelled in a Black box mode by associating a digitalised asset with an NFT HeliumAdapter, which in turn models an Helium device designed to detect the asset’s position via GPS. This will allow us to store real-time, permanent, and immutable information on the position and performance of the Rolling stock, such as speed, distance travelled, temperature, and energy consumption. This information could then be used to improve the safety and eficiency of the Rolling stock. For example, in the event of an accident, the performance information recorded on the associated dynamic NFT could be used to better understand what happened and how to prevent future accidents, optimise tram performance, and improve energy eficiency, or simply to schedule its inspection or preventive maintenance at the nearest depot.
To conduct the analysis of results arising from the digitisation of assets through dynamic NFTs, it was necessary to develop a decentralised application (Dapp), named RailChain, with the functionality of an Explorer for digitised railway assets. This application acts as the Front-end for displaying all the information related to the digital asset, including the "Live" information status provided by various NFT HeliumAdapters associated with the assets. RailChain also allows us to browse the tree structure of the digitised Product Breakdown Structure (PBS) from the root element to the last LRU, being able to retrieve all the documentary information in support of maintenance activities. RailChain, through functions made available by commercial Web3 APIs, allows us to access and interact with the Ethereum blockchain. These APIs are used to read information on the blockchain, such as the list of all transactions, the content of a particular smart contract, or the content of metadata associated with a specific NFT token. The data received from the blockchain are then processed and presented in a user-friendly format using technologies such as HTML, CSS, and JavaScript. RailChain, acts as an Application Server to a client call in the following format:
https://www.cladiecar.com/railchain/?address= [NFTSmartContractAddress]&id=[ID_NFT]
The application returns the HTML page reporting all the information encoded in the metadata of the NFT described by the call parameters. The application is also capable of independently identifying whether the referenced NFT is an asset or a digital item, proposing the correct HTML page. In the returned HTML page, alongside each asset or item, the QR code encoding of the HTTPS call that returns its clear information is displayed. This solution aims to simplify the maintenance operations related to the asset, making it possible to access all the necessary information simply by scanning the QR code. In particular, the parameterised URL will also be used on the identification label to be applied to the physical asset installed in the field, ensuring a correct correlation between the digital asset represented by the NFT and the physical asset.
An additional feature of the interface is its ability to display the decoding of alarms and states received through the NFT HeliumAdapter associated with the asset. To simplify the interface and to draw the maintenance operator’s attention only to active alarms, the following information will be displayed (Fig.3): • Alarm message associated with the specific bit of the status word. • Link button to the specific procedure to be followed for anomaly resolution. • References to the asset subject to the anomaly, it could be the asset itself to which the HeliumAdapter is associated, one of its child assets (or in any case present in its descending hierarchy) or one of the items associated with it.
All the NFTs minted during the experiment can be viewed from any NFT Marketplace, with the limitation that only the main attributes, listed in the metadata attribute section, will be visible. In this particular case, since these are test NFTs minted on the Goerli Testnet, they can be viewed on the version of the "Marketplace" Rarible reserved for the Testnet: https://testnet.rarible.com/.
Applying all the criteria described so far, the assets deifned in the case studies have been modelled, monitoring the various implementation phases through the web application RailChain. The model obtained shows a signicfiant number of cases representing the diferent possible combinations of identity and hierarchy relationships. For immediate recognition of the type of digitised equipment, the convention was used to mint assets with a representative image having a blue background, while green was chosen as the background colour for the items.
The digitisation of the physical asset was made possible through the writing of smart contracts, enabling the minting of NFT tokens in line with the ERC-721 standard. The dynamism of the digitised assets was achieved thanks to the update functions integrated into the smart contracts. This provided a degree of flexibility in managing attributes and documentation that proves maintenance carried out on the physical asset. All updates occur via transactions that are recorded in a certain and immutable manner on the blockchain. The update functions implemented in the smart contracts, in addition to ensuring the dynamism of the NFT tokens, are crucial in the design and fine-tuning phase of the digital asset, before its final release, making possible the correction or modification of metadata without proceeding to a new minting, and the simulation of digital assets behaviour in all possible operational situations.
This latter feature has been widely used throughout the implementation and testing phase, including the verification of asset integration with the NFTs of the HeliumAdapter family.
To experiment with the solution described, a minimal infrastructure was set up by installing some fundamental elements based on the Helium technology (Fig.2).
Firstly, a Helium Hotspot was installed to create a private LoRaWAN network. This device allows IoT devices compatible with the LoRaWAN network to connect and transmit the collected data to the Helium Console.
Secondly, IoT devices compatible with LoRaWAN and Helium technology (a device for managing digital I/O and one for serial communication) were deployed. These devices, also referred to as "nodes," can be placed near physical assets to collect real-time information on the state of the assets.
In the absence of a real field, alarms that arrived at the device through digital signals, such as those generated by the operation of a magnetothermal protection, were simulated on the first device through a breadboard connected to the digital inputs of the IoT device, which provided the voltage level necessary to activate the simulated input.
On the second device, a Modbus RTU serial communication with a control PLC was simulated using an IoT device capable of operating as a Modbus RTU Master using a laptop connected via an RS485-USB interface converter. A simulator was installed on the computer to create a virtual Modbus RTU device configured as a "slave" to respond to requests from the IoT Master device. Through RailChain, the necessary end-to-end checks were performed to verify that the digital model created reflected the operational conditions created in the field, thus validating the implemented architecture.
The last field finding refers to the second case study, which involves the digitisation of a dynamic asset. This functionality is tested by installing a LoRaWAN GPS-001 device on a car, trying to go beyond the coverage area of the installed Helium Hotspot.
Having already tested the integration of NFT HeliumAdapter with NFT assets, we limited ourselves to verify the correct updating of latitude, longitude, and distance travelled information on the NFT HeliumGPSAdapter associated with the IoT GPS device. This was done by viewing the status of the NFT using the Rarible Marketplace Testnet. Also in this case, the result was positive and, upon detecting five positions on an urban route of about 25 Km, the dynamic NFT is correctly updated even when the car goes outside the operational limits of the Helium Hotspot we installed, correctly performing the hand-over with several other Helium Hotspots already present along the test route. Despite the Helium coverage not being optimal, it was still suficient to ensure traceability throughout the route (Fig.4).
Based on the application of the proposed solution to the identified case studies and the experimentation carried out, it can be asserted that the use of dynamic NFTs in the digitisation of railway assets can contribute to the improvement of the predefined areas:
Communication Infrastructure: Dynamic NFTs can be used to collect "live" data on the status of assets located in remote locations or on Rolling stock distributed over a wide geographical area. This was demonstrated through the implementation of the HeliumAdapter smart contracts and the ability to integrate NFTs minted by these contracts with any type of digitised asset. The adapters in question were developed by digitising the LoRaWAN devices connected to the Helium network, the same principle can also be applied to other technological solutions.
Management and Maintenance Systems: The proposed approach can support the decentralisation of maintenance management systems, reducing costs related to the purchase of licences and fees for maintenance services. The proposed solution was developed without using any traditional database but exclusively the Ethereum blockchain and the IPFS distributed file storage and transfer system. This does not mean that a similar system implemented in a practical application can completely eliminate the need for a local database, but it could significantly limit its size, resulting in a consequent reduction of costs.
Time to Market: The usability of services is improved by enhancing the eficiency and speed of information distribution. The NFTs minted according to the principles described represent a digital copy of the physical assets, making information immediately available to maintainers as soon as each individual asset is modelled. Conversely, traditional systems require a complex technological infrastructure, including servers, network storage, a database configured by specialized personnel, a centralised maintenance management system, and the collection of information across the entire railway network. The tuning of the system and the upload of all information onto the system takes a considerable amount of time and the system can only be used upon the completion of this process.
Documentation: Dynamic NFTs can provide a decentralised solution for maintenance teams to access technical documentation, configuration information, and software related to the assets to be maintained, thereby improving the eficiency and quality of maintenance activities. This is ensured by associating with the assets a QR code, the scanning of which will invoke an instance of RailChain that will return the associated information. The QR code will be printed on the identification label applied to the physical asset; by scanning the QR code with any mobile device, the information will be immediately available.
Event Logging: In our view, this represents the strength of the proposed solution, namely the integration of IoT and Blockchain technologies. IoT technology integrated with the digitisation of assets via dynamic NFTs allows for the indelible registration on the blockchain of the usage state and anomalies that have occurred to an asset throughout its lifecycle. This constitutes an incentive to carry out the necessary preventive and corrective maintenance cycles and to register them on the blockchain concurrently with the actions taken to demonstrate that the asset has been kept in optimal conditions. As a further reinforcement of the virtuous cycle, it is conceivable to incentivise maintenance workers to fulfil their duties by rewarding them with tokens for every correctly completed maintenance cycle.
It should be kept in mind that the application of the proposed solution in a real-world context entails considerable complexity and the need to overcome certain technological and regulatory challenges related to scalability, security, privacy, regulation, and interoperability.
The system’s security is a critical aspect in the implementation of solutions based on dynamic NFTs and IoT technologies. It is vital to prevent possible external attacks that could compromise the security of digitised information and its integrity. To face these challenges, various countermeasures can be adopted, such as smart contract auditing to identify any vulnerabilities, increasing the number of nodes to ensure system redundancy, the use of authentication and authorisation protocols to limit access to sensitive information, and encrypting information to ensure its security in case of theft or loss.
To ensure maximum security, it is important to develop an integrated approach to security that includes constant risk assessment and the implementation of appropriate security measures to mitigate the identified risks.
Particular attention should be paid to the audit of smart contracts, an important security practice that involves the review and analysis of smart contracts to identify any vulnerabilities. To carry out an audit of smart contracts, an approach based on static and dynamic analysis techniques is required. Static analysis involves the review of the smart contract’s source code to identify any vulnerabilities or errors in its implementation. Dynamic analysis involves simulating the behaviour of the smart contract to identify any vulnerabilities.
These analyses should be carried out using specialised audit services ofered by third-party companies. These services involve the analysis of the smart contracts used for the digitalisation of assets and the generation of detailed reports on any vulnerabilities or problems encountered.
The proposed solution as a proof of concept responded
optimally to the project objectives. Further steps for
improvement are possible, for example, by utilising the NFT ERC-998
1–9
protocol [
Therefore, in a case of real implementation, the recommendation is certainly to implement the modelling using the ERC-998 protocol.
The study in question was based on a simplified model implemented using the ERC-721 protocol, elucidating the relationships between the assets within their metadata and assuming the following constraints: • Any potential changes in the relationship between the assets are recorded manually, ensuring the metadata and related tokenURI are updated On-chain. • All implemented NFTs belong to the same owner and are not partially transferred.
In our view, the primary limitation to the use of blockchainbased solutions remains the negative perception that many people have of this technology, often associated with financial speculation on digital assets and seen as unreliable and a potential source of scams. This negative perception hinders the adoption of this technology by many organisations, preventing the full exploitation of the potentialities of the blockchain, such as transparency, security, and decentralisation.
However, it should be noted that the spread of the application of the blockchain paradigm in asset tracking and railway maintenance does not necessarily have to occur with a Top-down approach, that is, starting from the need of a railway operator to implement an eficient maintenance management system for the entire transport network managed. We can also envisage a more gradual introduction approach of the Bottom-up type, starting from the product, where the component manufacturer used in the sector provides digisation of the marketed asset as an additional plus.
The use of NFTs in the railway maintenance context could be applied by Rolling stock manufacturers for public transport by implementing a digital ownership certificate for each asset, providing detailed information on manufacturing, certification, the PBS, spare parts, technical manuals, and preventive maintenance cycles to be performed. The operational manager of the infrastructure would thus be incentivised to use the same technology for managing the life cycle of the asset.
In conclusion, we can afirm that blockchain technology, enhanced by integration with IoT technology, has reached a degree of maturity such that its applications can be considered the new technological frontier. The blockchain paradigm is gradually overcoming the negative perception and prejudice to emerge as a reliable, secure, and decentralised solution for a wide range of applications in railway public transport, metro, and tram, surpassing traditional challenges related to data and process management. 1–9