A. Gandhimathinathan, B Abishek, Balaji Raghavan, T Dhilip · 5 authors
This research paper proposes a novel network that uses Distributed Ledger Technology (DLT) for automated textile fabric inspection to detect fabric defects accurately and efficiently. The traditional method of visual inspection by human operators is subjective, time-consuming, and prone to errors, resulting in low productivity and increased cost. Automated inspection methods, such as machine vision and machine learning, can help improve the quality of textile products by providing a more objective and accurate inspection process. DLT can provide an additional layer of security and transparency to the fabric inspection process. Overall, the proposed system has the potential to improve the efficiency, accuracy, and reliability of the textile fabric inspection process.
Real-time and vision-based quality control for industrial processes has drawn great interest from both scientists and practitioners, particularly following the transition to Zero Defect Manufacturing (ZDM) and Industry 4.0. Despite considerable progress, most ZDM approaches focus on the accuracy of the inspection process, often neglecting critical factors for application in the shop floor. On one hand, near real-time methods are needed for early defect detection and containment. On the other hand, data scarcity is an issue causing AI methods to overfit. Another concern is the accountability of AI results, since even if an AI pipeline is successfully deployed, its predictions are not verifiable in the long term. In this work, we explore a real-time solution based on lightweight Deep Residual Networks and Blockchain technology to address these issues. Concretely, we propose a two-phase training strategy to boost the performance of baseline classifiers while maintaining low inference times. The performance of the proposed methodology is presented in two different industrial use cases with strict timing requirements, one concerning battery assembly line and the other antenna manufacturing. We validate the proposed method for defect detection and compare the results with common training strategies demonstrating an improvement of 3% and 10% in F1-score and accuracy on the two cases respectively, while lowering inference time by 2.2× compared to existing light architectures. Contributing to the accountability of AI results, we present an IoT framework using Blockchain deployed in Private Ethereum.
The traditional smart contract defect detection method needs to generate corresponding knowledge according to its content when discovering new defects, and then the professionals build assertions and load them into the platform for defect detection. Due to existing methods are prone to loss or misdetection of defects due to human subjective consciousness, a smart contract vulnerability identification method based on Bi-LSTM neural network is proposed. The method firstly vectorizes the smart contract code, then inputs the vectorized data into the LSTM network to generate a model, and finally uses this model to detect defects. Experiments show that this method has a high defect detection rate for different types of new defects, and it is enough to be applied to practical scenarios.
<p>In order to use the fabric blockchain to store ERC 20 tokens, each one must be issued by the owner of the asset. In the case of Fabric, the owner issues the tokens in an "unspent" state on the ledger. The owner can then open a channel on the blockchain to allow others to receive the tokens. Once the token is transferred to a recipient, it is considered "spent" and removed from the channel. When the user redeems the token, it is stored on the Hyperledger Fabric blockchain ledger. This ensures that the token will not be transferred to another owner after the end of the term.</p>\n\n<p>The infrastructure component of Hyperledger Fabric is built on the Ethereum blockchain. It has the same benefits as Ethereum, but it is not as widely adopted as Ethereum. The ERC20 standard has been around for a while, so it can be used to store ERC 20 tokens. The fabric is modular and supports customization to suit different requirements. The use of Hyperledger Fabric 2.0 as a token platform means that the asset can be stored on the blockchain and used by the users.</p>\n\n<p>Unlike Ethereum, Hyperledger is a permissioned network where participants are known and have known identities. Companies with strict data protection laws and regulations will likely be comfortable with this kind of system. However, the network doesn't offer its own token system, so if a company wants to transfer funds between accounts, it should use Ethereum. This is also the case for companies that want to implement a decentralized payment system, such as an ICO.</p>\n\n<p>Aside from its scalability and security, Hyperledger Fabric has many advantages. In the case of ERC 20 tokens, Fabric allows the developers to code "chaincodes", which are smart contracts in Hyperledger parlance. These "chaincodes" are a type of cryptographic data hash. The Fabric consensus algorithm mirrors the workflow of organizations. It also offers secure access to data and can be used to create and execute smart contracts.</p>\n\n<p>As a result of this robust security, EVM chaincode can interact with Ethereum smart contracts and Fabric. By downloading the GitHub repository, developers can customize the chaincode to map to a folder of their choice. The instructions are very detailed. Once you have the appropriate software installed, you can build the multi-container apps and test them. Additionally, you can use "Go" language for Ethereum smart contracts on Fabric.</p>\n\n<p>Tokens on Ethereum are stored on the Ethereum network, which requires a transaction fee of $400. This fee varies depending on the size of the smart contract. However, this fee can be waived when generating tokens for customers. It's also possible to purchase a service that generates tokens for customers. You'll want to take the time to learn all you can about <a href="https://cryptominnie.com/">Crypto Minnie</a> before deciding. For $20, you can get the service of CoinManufactory and create your own custom ERC 20 tokens.</p>\n\n<p>Tokenization is a significant part of blockchain networks, and it is the process of converting physical assets into digital ones. Once digital, tokens are easily exchanged, purchased, or redeemed for other digital assets. Moreover, they can be categorized as either fungible or non-fungible assets. The latter has a broader utility than the former. You can make a list of different ERC 20 tokens in a Hyperledger Fabric instance and check out the corresponding information for Hyperledger Fabric.</p>\n\n<p> Overledger is a protocol that facilitates secure access to all DLTs. It configures each DLT to accept payments and send cryptocurrency to recipients. Then it provides a secure interface for the recipients to send and receive cryptocurrency. This is a highly effective way to manage distributed assets and tokens. Ultimately, it allows an organization to secure a blockchain while enabling a wide range of applications.</p>
Industrial Internet of Things (IoT) depends on a vast array of sensors, processing elements, and actuators to automate data collection, processing, and response. One of the major challenges in the stage of data collection is the accuracy of electromechanical sensors. The concept of Proof-of-Stake (PoS), which is popular in Ethereum blockchain platform, can be used to improve the robustness of data collection. PoS works by selecting a subset of nodes based on their holdings called “stakes”, to perform a specific task. This algorithm is incorporated into IIoT by assigning a stake value to each node internally. The stake is assigned a predefined value during installation and changes periodically based on accuracy. A stake threshold is set so that only those nodes with stake above that value are considered contributors and others act as validators. During the initial setup, everyone validates every other node. Beyond that, the selection of contributors is repeated periodically. At every time interval, validators take the global average of contributors and find the deviation from the observed value. This is compared with a deviation limit. If none are within the limit, it indicates outliers, and the system invalidates itself. Otherwise, the node with the least deviation will emerge victoriously and send its value across to the processing unit. Those nodes with high deviation will have a part of their stake transferred to those with less deviation. Hence, the total stake remains the same, whereas the distribution keeps changing. The advantage is that this process is entirely distributed once all nodes multicast their readings to the validator group. This also ensures that sensors with persistent errors are singled out because they cannot recover their stake.
This paper presents a complete solution consisting of sustainable IoT-based Reliable Industrial Data Services (RIDS) able to manage the huge amount of industrial data coming from cost-effective, smart, and small size interconnected factory devices for supporting manufacturing online monitoring and control. The i4Q Framework guarantees data reliability with functions grouped into five basic capabilities around the data cycle: sensing, communication, computing infrastructure, storage, and analysis and optimisation. With the i4Q RIDS, factories will be able to handle large amounts of data, achieving adequate levels of data accuracy, precision and traceability, using it for analysis and prediction as well as to optimise the process quality and product quality in manufacturing, leading to an integrated approach to zero-defect manufacturing. The i4Q Solutions efficiently collect the raw industrial data using cost-effective instruments and state-of-the-art communication protocols, guaranteeing data accuracy and precision, reliable traceability and time stamped data integrity through distributed ledger technology and provide simulation and optimisation tools for manufacturing line continuous process qualification, quality diagnosis, reconfiguration and certification for ensuring high manufacturing efficiency and optimal manufacturing quality.
Gianluca Zanella, Charles Zhechao Liu, Kim‐Kwang Raymond Choo
Patent analysis is crucial for technology monitoring, forecasting, and assessment, and facilitates entrepreneurs and different stakeholder groups to develop forward-looking technologies and business strategies. However, the speed and scale in the development of disruptive technologies, such as blockchain, present a challenge for analysts and experts. In this article, we propose an unsupervised systematic patent analysis framework that applies a mixture of cosine-based and density-based outlier analysis to the patent space. A sample of 13 393 blockchain-related patents published between January 2014 and June 2020 is used to test the proposed framework. Specifically, this framework merges cosine and density-based outlier detection methodologies to improve the identification of outliers within clusters of patents. The identified outliers are visualized through an age-outlier technology-opportunity analysis map that represents the different levels of novelty existing in each cluster of the patent sample. The map facilitates companies to better target their R&D efforts and maximize the return of technology investments. Benchmark results show that the proposed outlier detection method improves recall, precision, and f1 score. In addition, the results show that the cluster with a higher percentage of outliers represents the Internet of Things applications of blockchain technology.
Smart manufacturing systems are growing based on the various requests for predicting the reliability and quality of equipment. Many machine learning techniques are being examined to that end. Another issue which considers an important part of industry is data security and management. To overcome the problems mentioned above, we applied the integrated methods of blockchain and machine learning to secure system transactions and handle a dataset to overcome the fake dataset. To manage and analyze the collected dataset, big data techniques were used. The blockchain system was implemented in the private Hyperledger Fabric platform. Similarly, the fault diagnosis prediction aspect was evaluated based on the hybrid prediction technique. The system's quality control was evaluated based on non-linear machine learning techniques, which modeled that complex environment and found the true positive rate of the system's quality control approach.
Tie Qiu, Min Zhang, Xize Liu, Jing Liu · 6 authors
As the application of the industrial Internet of Things (IIoT) becomes more widespread, the IIoT is being combined with social networks. Nodes in the network can be users, machines, and so on. Using the sensing detection technology of the IIoT, industrial machines can realize real-time informatization, which is convenient for users to perform remote management. Nodes can communicate with each other and make ratings. These ratings can be modeled as directed weighted edges between nodes and form directed weighted networks (DWNs). The edge weight represents the “strength” of relationship and the direction of edge points from the edge generator to the edge receiver. Predicting edge weights in DWNs is critical to predicting unknown ratings or recovering lost data. In this article, we propose a directed edge weight prediction model (DEWP) using decision tree ensembles. It extends the local similarity indices to DWNs and extracts a series of similarity indices between nodes as features of each edge. These features are used to construct a blended regression model of random forest, gradient boost decision tree, extreme gradient boosting, and light gradient boosting machine. The proposed algorithm was evaluated experimentally with the Bitcoin OTC and Bitcoin Alpha datasets by removing 10% to 90% of edges in the original network. Compared with other classical algorithms, DEWP has higher prediction accuracy and robustness.
In the Architecture, Engineering, construction and Operations (AEcO) there is a growing interest in the use of the building Information modelling (bIm). Through integration of information and processes in a digital model, bIm can optimise resources along the lifecycle of a physical asset. Despite the potential savings are much higher in the operational phase, bIm is nowadays mostly used in design and construction stages and there are still many barriers hindering its implementation in Facility management (Fm). Its scarce integration with live data, i.e. data that changes at high frequency, can be considered one of its major limitations in Fm. The aim of this research is to overcome this limit and prove that buildings or infrastructures operations can benefit from a digital model updated with live data. The scope of the research concerns the optimisation of Fm operations. The optimisation of operations can be further enhanced by the use of maintenance smart contracts allowing a better integration between users' behaviour and maintenance implementation. In this case study research, the Image recognition (Imr), a type of Artificial Intelligence (AI), has been used to detect users' movements in an office building, providing real time occupancy data. This data has been stored in a bIm model, employed as single reliable source of information for Fm. This integration can enhance maintenance management contracts if the bIm model is coupled with a smart contract. Far from being a comprehensive case study, this research demonstrates how the transition from bIm to the Asset Information model (AIm) and, finally, to the Digital Twin (i.e. a near-real-time digital clone of a physical asset, of its conditions and processes) is desirable because of the outstanding benefits that have already been measured in other industrial sectors by applying the principles of Industry 4.0.
Supply chain traceability is one of the most promising use cases to benefit from characteristics of blockchain, such as decentralization, immutability and transparency, not required to build prior trust relationships among entities. A plethora of supply chain traceability solutions based on blockchain has been proposed recently. However, current systems are limited to tracing simple goods that have not been part of the manufacturing process. We recommend a method that allows for the traceability of manufactured goods, including their components. Products are represented using non-fungible digital tokens that are created on a blockchain for each batch of manufactured products. To create a link between a product and the components that are needed to produce it, we propose “token recipes” that define the amount of tokenized goods required for minting a new token. As input tokens are automatically and transparently consumed when creating a product token, the physical process of producing a new item out of existing components is projected onto the ledger. This ultimately leads to the complete traceability of goods, including the origin of inputs. Evaluating the performance of the system, we show that a prototypical implementation for the Ethereum Virtual Machine (EVM) scales linearly with the amount of the input and goods tracked.
The goal of this paper is to ascertain with what accuracy the direction of Bitcoin price in USD can be predicted. The price data is sourced from the Bitcoin Price Index. The task is achieved with varying degrees of success through the implementation of a Bayesian optimised recurrent neural network (RNN) and a Long Short Term Memory (LSTM) network. The LSTM achieves the highest classification accuracy of 52% and a RMSE of 8%. The popular ARIMA model for time series forecasting is implemented as a comparison to the deep learning models. As expected, the non-linear deep learning methods outperform the ARIMA forecast which performs poorly. Finally, both deep learning models are benchmarked on both a GPU and a CPU with the training time on the GPU outperforming the CPU implementation by 67.7%.