Showing papers presented at "International Conference on Service Operations and Logistics, and Informatics in 2019"

A Blockchain-based Framework for Collaborative Production in Distributed and Social Manufacturing

Abstract: In recent years, novel collaborative production paradigms, such as (re-)distributed manufacturing and social manufacturing, have attracted intensive attention due to their potential of further changing the existing manufacturing modes and industrial structures. Blockchain, as a new distributed computing architecture, is expected to be successfully used in manufacturing, and help solve issues related to interoperability and collaboration, security and supervision, marketization and protocol, democratic organization and global value chain governance, and so on. In this paper, a blockchain-based framework for collaborative production is proposed, aiming at providing novel ideas for the design, implementation and optimization of collaborative production, and also inspiring the upgrading and transformation of manufacturing industries.

State Evaluation of Power Transformer Based on Digital Twin

Abstract: Transformer is the key substation equipment in power system. The data accumulated during system operation can provide information for further state evaluation and fault diagnosis to ensure safe and stable operation of power system. However, these data are uncertain and the evaluation is based on experience, which makes the state evaluation of transformer still a challenge. At present, digital twin, which could represent the features of any objects or subjects, has received widely attentions from various regions, while its application in power system is still limited. Therefore, this paper proposes a transformer state evaluation approach based on the digital twin. This approach establishes a digital twin model of power transformer, and acquires the sample data and labels from the simulation process and state evaluation under various condition. These sample data and labels will guide the decision making of state evaluation in the real situation. This paper compares the sample data and labels from the digital twin model with those from the physical system, so that the performance of digital twin model and real system would be promoted during this mutual amendment progress.

A Blockchain-based Framework for Central Bank Digital Currency

Abstract: Cryptocurrency and blockchain technologies have developed in parallel in recent years, with technological breakthroughs in currency issuance, payment methods, and currency storage. However, the existing cryptocurrencies cannot replace fiat money. There is a huge gap between decentralized cryptocurrency and central bank digital currency, namely CBDC, in terms of monetary governance and circulation. In this paper, we propose the function and security requirements of CBDC, through a comprehensive analysis of the existing typical cryptocurrency and the prototype of the CBDC scheme. On this basis, we present a blockchain-based framework for CBDC with three layers, including supervisory layer, network layer and user layer, and describe the key business processes of the CBDC's entire lifecycle of issuance-circulation-withdrawal in detail. Finally, we take cross-border payment as an example to explain the transaction process of CBDC. We aim to provide theoretical guidance for CBDC design.

Review on Development of Smart Education

Abstract: Smart education is leading the development direction of Chinese education informatization and becoming a main theme of education development in the era of which technology changes education. There are seven main branches of smart education, namely Intelligent Tutoring System (ITS), smart campus, Big Data in Education (BDE), knowledge graph, educational robots, virtual teachers, and personalized education. Based on literature survey and market research, this paper gives a comprehensive overview of the research and application of smart education from above seven aspects and proposes its development based on the status of smart education development.

A New Approach for Face Anti-Spoofing Using Handcrafted and Deep Network Features

Abstract: In biometrics, face recognition methods are achieving momentum with recent progress in the computer vision(CV). Face recognition is widely used in the identification of an individual's identity. Unfortunately, in recent research work has revealed this face biometrics system is unprotected to spoofing attacks using by very low price instrument such as printed 2D photos attack, 3D masking attack and taking videos using smart devices (reply attack). Therefore, a Liveness Attack Detection (LAD) approach is needed to improve the high-quality security of face recognition system. Most of the earlier worked LAD methods for face anti-spoofing methods have highlight on using the handcrafted features, which are developed by expert knowledge of researcher. As example Gabor filter, Histogram of Oriented Gradients, local ternary pattern, and the Local Binary Pattern. Because of that, the extracted features consider limited factors of the problem, yielding a capture accuracy that is very low and changes with the point of presentation in attack face images. The deep learning method has developed in the computer vision research community, which is proven to be suitable for automatically training. In this article, we approach to mix or combine the handcrafted features and deep neural network features to design the discriminant face spoofing detection. The handcrafted features were based on LBP analysis. We examine the features information from the brightness and the chrominance channels using LBP descriptor. In deep features, we present an approach based on pre-trained convolutional neural network VGG-16 model using static features to recognize video and printed(2D) photo attacks. By attaching this two types of image features on our dataset and public databases, we get good results to identify real and attack images feature, called hybrid features, which has better discrimination ability to understand spoofing image feature.

A Novel Blockchain Oracle Implementation Scheme Based on Application Specific Knowledge Engines

Abstract: In blockchain ecosystems, an Oracle is a service tool which provides real-world data for smart contracts and other blockchain applications. At present, there are several Oracle implementation schemes, e.g. centralized Oracles, decentralized Oracles, and hardware Oracles. However, these schemes typically suffer from single source of data and low scalability. Application Specific Knowledge Engine (ASKE) is an integrated topic/application-centered knowledge portal that supports effective information retrieval and analysis. Inspired by ASKE, in this paper, we propose a novel Oracle implementation scheme. The proposed scheme can realize multi-source data extraction and analysis, then working prototypes are demonstrated to show the validity of the scheme.

The Pod Assignment Model and Algorithm in Robotic Mobile Fulfillment Systems

Abstract: This paper studies the pod assignment in the Robotic Mobile Fulfillment Systems which is a parts-to-picker storage system where robots carry movable shelves, called pods, to stationary pickers. On the basis of deciding which product to put in which pod depending on the product correlation, we establish a mathematical model with the objective of minimizing the total picking distance to decide which pod to put in which position. A two-stage hybrid algorithm is proposed to solve the model. In the first stage a Greedy Algorithm is designed to generate initial solution which consider the pod correlation. In the second stage, a simulated annealing algorithm is used to optimize the initial solution and get the final pod layout results. The experimental results show that the pod assignment model and algorithm optimization scheme proposed in this paper can improve the order picking efficiency.

A Framework of Decentralized Electricity Market Based on The Collaborative Mechanism of Blockchain and Edge Computing

Abstract: Distributed generations are expected to develop rapidly in the electric power system. As an essential infrastructure, the structure of electricity market moves to distributed or decentralized network. Without a central coordinator, transaction security and optimal dispatch will be a challenge. In this paper, blockchain and edge computing are used to be a potential solution of decentralized electricity market. A decentralized electricity pricing algorithm is installed in the onsite edge computing terminal, and then the transactions are implemented in blockchain. Case studies are presented to demonstrate the availability of pricing mechanism.

Career Trajectory Prediction based on CNN

Abstract: With the rapid development of online recruitment, it is very important for recruitment enterprises to analyze job seekers' experience and recommend suitable and satisfied job to them. This paper describes the dataset of job seekers' resumes from the internet. According to the basic personal information of job seekers such as gender, age, specialty and education, as well as multiple work experiences including working duration, company industry, company scale, monthly salary and position name, we predict job seekers' future job information. Considering that there are both textual data and numerical data in each resume. In order to extract semantic features accurately, the text data is transformed into vectors with the help of Word2Vec provided by Google. Ultimately, we construct a convolutional neural network (CNN) model to address the career trajectory prediction problem. The validity of the model is verified on a real-world dataset with 70,000 resumes.

Design of Trust Blockchain Consensus Protocol Based on Node Role Classification

Abstract: The blockchain technology has broad application prospects in the financial field, public welfare industry, and Internet with its characteristics of decentralization, openness, autonomy, information unchangeability and anonymity. The most prominent problem it faces is the difficulty in reaching a credible consensus quickly. Based on the consensus mechanism of DPoS and PBFT, a new trust blockchain consensus protocol is proposed. This protocol adds trust value attributes to nodes, dynamically classifies node roles, tracks node behaviors in transactions, and scores trust values for the nodes. The node promotion and demotion mechanism is introduced and the roles of nodes are reclassified according to their latest trust values. Results show that the proposed consensus protocol has high efficiency, low consumption, high fault tolerance and dynamic adaptability, which can meet the needs of most alliance chains.

A Kind of Accuracy Improving Method Based on Error Analysis and Feedback for DLP 3D Printing

Abstract: As an important 3D printing technology, Surface Exposure Rapid Prototyping Technology (SERPT) has developed rapidly in recent years. In order to optimize the printing quality and improve the accuracy of Digital Light Processing (DLP) 3D printing, the main errors in the process of DLP 3D printing is enumerated first in this paper. Further, based on the DLP 3D printing system, a closed-loop transfer function is constructed to measure and analyze all the existing errors. And then, a closed-loop negative feedback mechanism is proposed to compare the formed size and the target size of the model, for the purpose of correcting or compensating the printing errors. Finally, the new printing model size is fitted linearly, which effectively reduces the errors of DLP 3D printing. Then a better 3D printing model could be obtained.

Basis-Prediction Method for Short-Term Traffic Flow Prediction and Its Application

Abstract: Traffic flow forecasting is one of the key issues in smart traffic systems. The process of traffic flow changing involves a high degree of nonlinearity and randomness, environmental interference and measurement noise, which brings difficulties to accurate traffic flow prediction. Aiming at improving the accuracy of short-term traffic flow prediction, this paper proposes a method called Basis Prediction method. A raw traffic flow series can be deemed as a summation of a basis series that indicates the changeable trend of the traffic flow and a deviation series that represents the random interference information involved in the flow. The basis series comprises mainly low-frequency signals and the deviation series is composed of some high-frequency signals. Based on an appropriate extraction of the basis series, prediction merely of the basis series brings more precise prediction of the raw traffic flow. This paper suggests adopting wavelet decomposition to obtain the basis series and the deviation series from the raw traffic flow. To predict the basis series effectively, two algorithms, local weighted partial least squares (LW-PLS) and Kalman filtering, are adopted separately and a comparison between them is also provided. In this paper, real data of traffic flow of Xinbei city in Taiwan was collected and used for validation of the proposed basis prediction method. The results show that the use of that proposed method improves the accuracy of short-term traffic flow prediction by about 2% on average.

Multi-source Heterogeneous Information Fusion of Power Assets Based on Knowledge Graph

Abstract: The effective integration of power assets information plays an important role in the life cycle management of power assets. However, The multi-source heterogeneity of power data makes it inefficient and limited applied under traditional data fusion technology. This paper introduces knowledge graph technology to realize the integration of depth, accuracy, efficiency and visualization of power assets information, and lays the foundation for data value mining. Aiming at “flattening operation and efficient utilization of power assets information”, this study proposes a general process of building professional knowledge graph in power field, and proposes a multi-source heterogeneous information fusion method for power assets based on knowledge fusion. As a case, the transformer information in power transmission and transformation assets in some areas of Hubei Province is selected to quantify the alignment quality and redundancy of entities after knowledge fusion, and the relationship between indicators and thresholds is analyzed to obtain a relatively optimal threshold setting interval.

Trust Degree Calculation Method Based on Trust Blockchain Node

Abstract: Due to the diversity and mobility of blockchain network nodes and the decentralized nature of blockchain networks, traditional trust value evaluation indicators cannot be directly used. In order to obtain trusted nodes, a trustworthiness calculation method based on trust blockchain nodes is proposed. Different from the traditional P2P network trust value calculation, the trust blockchain not only acquires the working state of the node, but also collects the special behavior information of the node, and calculates the joining time by synthesizing the trust value generated by the node transaction and the trust value generated by the node behavior. After the attenuation factor is comprehensively evaluated, the trusted nodes are selected to effectively ensure the security of the blockchain network environment, while reducing the average transaction delay and increasing the block rate.

A Driving Warning Method based on YOLOV3 and Neural Network

Abstract: Unmanned driving warning is one of the important core issues of unmanned driving decision. This paper studies the method of prediction on the traffic warning information for unmanned driving warning, which is based on YOLOV3 and BP neural network. Firstly, weight training is carried out through self-made data set. Secondly, the trained YOLOV3 model is used to classify the image and video data: car, bicycle, bus, motorbike, person. Finally, the YOLOV3 target detection information will be classified into 5 categories by the BP neural network: safe, stop, slow, left, right. The result of the simulated experiment shows that the model works well. Based on different data set, the accuracy of the second classification is up to 0.99.

Emojis-Based Recurrent Neural Network For Chinese Microblogs Sentiment Analysis

Abstract: As a novel graphical expression derived from emoticons, emojis have a wide application in social networks. Emojis can assist people in expressing stronger sentiment or show subtler sentiment indirectly which are helpful to sentiment analysis. In this paper, we propose an emojis-based recurrent neural network for sentiment analysis in Chinese microblogs. Firstly, we differentiate ambiguous emojis and explicit emojis by using pre-trained word embedding and a new sentiment lexicon. Then we verify that users' information can eliminate the ambiguity of ambiguous emojis to some extent and confirm the sentiment polarity of ambiguous emojis. On the basis, we obtain emoji representations by utilizing the position vector, semantic vector and sentiment vector of emojis, then put the emoji representations into Bi-directional gated recurrent unit(BiGRU) neural network model to conduct sentiment analysis. The experimental results on a Chinese microblog dataset demonstrate that compared with other baselines, the proposed model can improve the accuracy significantly in sentiment analysis.

A decentralized social networking architecture enhanced by blockchain

Abstract: The privacy issues become a major problem that should be resolved for the existing centralized online social networks, which have prompted researchers to consider the decentralization framework for online social networks. In this paper, we propose a novel decentralized social networking architecture enhanced by blockchain technology. We use a sharding framework to increase the system scalability, a blockchain system to ensure the data integrity and consistency, a reputation-based authority control method to improve the system security.

Intersection Dynamic Signal Control Method based on Road-segment Characteristics and Time-period Characteristics

Abstract: The distribution of traffic flow at intersection is an important basis for traffic signal control, based on the characteristics of time and the characteristics of road segments that affect the dynamic changes of traffic flow, this paper proposes an adaptive control strategy for road traffic lights. Firstly, neural network technology is applied to traffic flow state analysis, scientifically define traffic flow patterns based on road-segment information and completes fast and accurate short-term traffic flow distribution state identification online. Secondly, according to the traffic distribution status of intersections, the signal design process is proposed to improve the traffic signal control efficiency. Finally, the verification of the actual traffic flow of Jingming South Road and Juxian Street in Kunming shows that the adaptive traffic signal control method proposed in this paper can effectively reduce intersection delay, number of stops and queue length.

Analysis of Charging Continuous Energy System and Stable Current Collection for Pantograph and Catenary of Pure Electric LHD

Abstract: Aiming at the problem of limited power battery capacity of pure electric Load-Haul-Dump (LHD), a method of charging and supplying sufficient power through pantograph-catenary current collection system is proposed, which avoids the problem of poor flexibility and mobility of towed cable electric LHD. In this paper, we introduce the research and application status of pantograph and catenary, describe the latest methods and techniques for studying the dynamics of pantograph-catenary system, elaborate and analyze various methods and technologies, and outline the important indicators for analyzing and evaluating the stability of current collection between pantograph-catenary system. Simultaneously, various control strategies for pantograph-catenary system are introduced. Finally, the application of the pantograph-catenary system in high-speed railway and urban electric bus is discussed to illustrate the advantages of pantograph-catenary system charging and energy supply, and it is applied to pure electric LHD charging and energy supply to ensure power adequacy.

A Novel Data-driven Approach for Tropical Cyclone Tracks Prediction Based on Granger Causality and GRU

Abstract: The strong tropical cyclones will make a drastic effect on human life and natural environment. In these days, as meteorological data and monitoring data accumulated to a large amount, traditional methods of predicting tropical cyclones tracks face many challenges about the prediction efficiency and accuracy. Recently, deep learning method has proven to be an efficient and accurate way to forecast time series data. Therefore, this paper proposes a novel data-driven deep learning model based on Granger causality and Gated Recurrent unit(GRU) to predict the tropical cyclones tracks by selecting the meteorological factors that affect the tropical cyclone locations. The model comprises several aspects including data preprocessing, the feature selection layer, and the GRU with a customized batch process. The model was trained using a real-world tropical cyclones dataset from the year 1945 to 2017, and the results proved that our proposed model can improve the prediction accuracy in the experimental scenario.

Process Route Optimization for Generalized Energy Efficiency and Production Time in Machining System

Abstract: Total consumption of direct electric energy and average embodied energy of cutters and cutting fluid as well as other auxiliary resource consumed in machining a part are defined as generalized energy consumption. In this paper, a process route optimization model oriented for improvement of generalized energy efficiency and production time is proposed. The selection of machining sequence, machine tools, cutting tools, tool access direction and processing parameters in process route is optimized through a SA-QPSO hybrid algorithm. Results indicate that the balance between generalized energy efficiency and machining time is achieved and detailed analysis of process route influence on these two objectives is given afterwards.

E-commerce Supply Chain Cost Optimization with Incentive Compatibility Constraints

Abstract: Even though the e-commerce supply chain operations have paid continuous attention in improving their operational capabilities and optimizing costs, strategic interactions and management issues still exist, thereby impeding the optimal performances. Game theory-based contract models have vastly been adopted in supply chain studies to resolve issues of strategic interactions and decision-making among different supply chain members. In that context, contracts include incentive compatibility constraints to ensure that the players have sufficient incentive to remain within the contract. However, less attention has been given to contract models, which includes multi-levels of e-commerce-based supply chain operations. Therefore, this study develops a cost-sharing contract, including incentive compatibility constraints for the three-level e-commerce supply chain to address the issues of cost information asymmetry. We consider cost information asymmetry issues of the upstream and downstream of the supply chain where e-tailer shares a fraction of operational costs of the product supplier and the 3PL operator. The results of this study emphasize that the cost-sharing contract significantly reduces the overall supply chain cost while proving Pareto-improved outcomes in terms of cost minimization.

Short Term Traffic Flow Forecast Based on CM-GRU Networks

Abstract: Intelligent transportation systems (ITS) have developed for a long time. The rise of deep learning has brought new vitality of the ITS. However, traffic flow data is usually time-correlated and highly randomized. The data distribution will also change dynamically. To actualize the forecasting of traffic flow accurately, we use the historical traffic information to predict the messages of the traffic flow at any time interval. This paper proposes an efficient traffic flow forecast architecture based on deep learning. The method combines the gated recurrent unit (GRU, a type of recurrent neural network) layers and one-dimension convolution layers. Since the performance of these models has a strong dependence on hyper-parameters, this paper conducts a large-scale search of the hyper-parameter space. At the same time, experiments on flow data show that the method proposed in this paper can achieve a better prediction accuracy. Experiments also show lower test errors compared with the existing approaches.

Game-Theoretic Energy Consumption Planning at Distribution Level for Future Smart Grid

Abstract: The prevalent game-theoretic demand-side management methods based on the control of residential smart appliances, and Nash-equilibrium (NE) can be achieved for optimal results. However, commercial customers consume much more energy than residential customers, and seldom work is conducted for academic and commercial buildings with central heating, ventilation and air conditioning (HVAC) systems. HVAC system takes up the largest portion of electricity usage in the commercial sector. The control of HVAC system cannot solely aim to minimizing the energy usage, but also need to maintain the comfort in accordance with building schedule. Therefore, the methods for optimizing HVAC systems are divergent from the optimization methodologies for home appliances. In this manuscript, a novel Game-Theoretic methodology is addressed to control the operations of heating, ventilation, and air conditioning system (HVAC). The approach can help to reduce peak-load, and save monetary expenditure all year around. We also introduce social cost, which is defined as a combination of the energy expense and the cost of human working productivity reduction, to illustrate advantages of the proposed methodology.

3D Deep Learning for 3D Printing of Tooth Model

Abstract: Recently, with the development of both 3D sensors and 3D virtual network that bring the needs of interaction with the real world, many 3D applications burst out. However, it is difficult to understanding these three-dimensional scenes with a fixed program. Then, a data-driven method is required to process these 3D data, which brings a strong demand of 3D Deep Learning in 3D data. Towards this goal, with an end-to-end deep learning, the experiment is based on PointNet++, a well proposed method for feature extraction. The experiment optimizes the network structure and parameters to improve the classification results. Finally, the network is applied to tooth model for classification and identification so that the dental model can be found from different perspectives.

Economic Evaluation Method of Power Grid Planning Based on Fuzzy SEC Comprehensive Index System

Abstract: With the rise and development of power market, saving investment is more than the only goal of power grid planning. Improving the system reliability, reducing the risk level and achieving the pursuit of maximize the benefit of the whole society are the needs of power grid planning department. This paper considers the relationship between the safety, efficiency and cost to establish a power grid planning evaluation model based on the analysis of Life Cycle Cost (LCC). At the same time, the economic evaluation model is established, and the evaluation indices are given. Then, according to the fuzzy multiple attribute decision making and economical evaluation model, this paper constructs the comprehensive evaluation index to evaluate the scheme. Finally, this paper verifies the effectiveness of the evaluation model through one engineering examples.

Trusted collection, management and sharing of data based on blockchain and IoT devices

Abstract: This paper takes the Internet of Things(lot) and Blockchain technology as the research direction, and makes a comprehensive analysis of the trusted data collection, management and sharing at the device end of the Internet of Things. It mainly solves the problems of data acquisition, data protocol, data application and data security, and provides solutions. The Internet of Things (IOT) technology is used to solve the problems of mass data information collection, multi-communication protocol integration, directional data acquisition and analysis, sensor direct acquisition and application, and wireless Internet of Things integration. Block chain technology is used to solve multi-center data acquisition and storage, data traceability can not be tampered with, data sharing is safe and reliable.

Design of Smart Park Venue Rental System based on Blockchain Technology

Abstract: Through the installation of IoT smart devices in smart parks, this paper uses blockchain technology to collect and transmit information in all aspects of the venue rental business, and realizes automatic business processing through smart contracts, to solve the problems of high management cost and low work efficiency of the current renting operators.

Sci-Tech Business Incubator - Favorable Network Operation Management Mode for Small and Medium Enterprises Development

Abstract: Sci-tech Business Incubator is a specifically designed service organization to cultivate and support the small and medium enterprises in high-tech. It can provide a series of support for those newly-established technological enterprises. It has a significant social and economic meaning for the development of high_tech industry, improvement of national and regional innovation system. With the development of Information technology, the operation and management mode of sci-tech business incubator also met new changes. This thesis aims to explore this new change in the operation and management mode. More and more sci-tech business incubator came into being. The demand of its service function is getting deepen. The network operation of Chinese sci-tech business incubator is becoming an inexorable trend to meet its interior demand and development. Meanwhile, the network operation of Chinese sci-tech business incubator in our country is still in an initial stage. Compared with the advanced level of foreign countries, we need to make more exploration and research.

Bitcoin Fee Decisions in Transaction Confirmation Queueing Games Under Limited Multi-Priority Rule

Abstract: In the Bitcoin system, transaction fees serve not only as the fundamental economic incentive to stimulate miners, but also as an important tuner for the Bitcoin system to define the priorities in the transaction confirmation process. In this paper, we aim to study the priority rules for queueing transactions based on their associated fees, and in turn users' decision-making in formulating their fees in the transaction confirmation queueing game. Based on the queueing theory, we first analyzed the waiting time of users under non-preemptive limited multi-priority (LMP) rule, which is formulated to adjust users' waiting time over different priorities. We then established a game-theoretical model, and analyze users' equilibrium fee decisions. Towards the end, we conducted computational experiments to validate the theoretical analysis. Our research findings can not only help understand users' fee decisions under the LMP rule, but also offer useful managerial insights in optimizing the queueing rules of Bitcoin transactions.