Browse Topic: Big data

Items (176)
Metallurgical cranes have a high risk of structural fatigue damage and failure under complex working conditions such as high temperature, heavy load, and strong electromagnetic interference. This article proposes a data-driven structural fatigue damage health monitoring system. This system integrates fiber Bragg grating sensing technology, rigid flexible coupling multi-body dynamics simulation, and big data analysis methods to construct a sensor optimization layout strategy based on rigid flexible coupling virtual prototype simulation, achieving real-time perception of stress states in key parts such as the mid span and end beam corners of the main beam. Develop a visualization system that integrates health monitoring, damage diagnosis, and life prediction. This system can dynamically evaluate the structural health status of metallurgical cranes and predict the remaining life of the structure based on a nonlinear cumulative damage model. On site engineering applications have shown that the monitoring and prediction visualization system can effectively improve the intelligent and safe operation and maintenance level of metallurgical cranes, providing a data foundation and possibility for their predictive maintenance.
Chen, LiZhang, XuDing, Keqin
Under China’s intelligent manufacturing strategy, manufacturing enterprises are expected to achieve digital and networked operations by 2025, with full digital transformation by 2030. Intelligent factories, the core of this transformation, rely on interconnected, integrated, and data-fused systems. This paper focuses on the micro-assembly intelligent workshop at the Nanjing Research Institute of Electronics Technology, which produces micro-circuit modules for large-scale complex electronic systems. The workshop combines discrete and process manufacturing modes, presenting unique challenges for digital management. A digital management platform based on a five-layer architecture (device, network, data, application, and decision layers) is proposed to address multi-dimensional business needs, including production scheduling, logistics, execution, and decision optimization. A hierarchical workflow structure of the workshop, consisting of a main workflow and several sub-processes, is in-depth studied and designed. The platform is constructed based on requirements analysis and workflow design of the workshop and integrates systems such as MES, APS, WMS, and SCADA, supported by AI-driven big data analytics. This study offers a practical framework for advancing digital transformation in the electronics industry.
Zhang, JianWang, JiafengGuo, Yongzhao
Coal is an important component of China's energy structure, mainly transported by three modes: railway, waterway, and highway. In regional coal transportation, highway transport undertakes numerous collection-distribution tasks and medium-short distance transport, playing a vital and indispensable role. Considering the characteristics of the coal highway transportation market and the demand for price indices, a three-tiered coal highway freight price index system has been established, including individual indices, classified indices, and an overall index. Using order data from the logistics platform of the Coal Big Data Center, the coal highway freight price index is compiled by adopting the internationally Laspeyres chain method. The methodological selection has passed the ADF stationarity test. Economically, the coal highway freight price index is closely correlated with coal prices, with the correlation coefficient reaching over 0.7, which can reflect about the coal highway freight market and fill the gap in market highway freight price monitoring.
Zhao, NanxiWang, XinziRong, Haoyu
Roadway departures remain a major cause of crashes, injuries, and fatalities on U.S. roads. Technologies such as lane keeping assist (LKA) and lane centering assist (LCA) can help mitigate these crashes, but their development involves extensive characterization of the parameter space in which they operate. Lane and road departures (LDs/RDs) and lane changes (LCs) must be systematically described and quantified to distinguish kinematic features, identify contributing factors, and benchmark system influence on lateral control. This study developed a unified pipeline to mine over 36 million miles of naturalistic driving study (NDS) data collected from more than 3800 participants. The pipeline integrates various types of signals to detect roadway boundary crossings, classify LKA-relevant scenarios, and extract roadway, driver, environmental, and assistance-related parameters. Lane keeping epochs with and without LKA were also extracted to quantify system influence on lateral control. In the NDS analysis, crashes include both object contact events and RDs, defined as non-premeditated departures from the intended travel surface involving at least one tire. Analysis of pre-identified crashes in the NDS showed that unintentional RDs accounted for 5.67%, unintentional LDs for 1.76%, and intentional LCs for 1.55%, corresponding to lower-bound rates of 2.7, 0.8, and 0.7 crashes per million vehicle miles traveled. RD crashes were predominantly right-sided, LD crashes left-sided, and both were overrepresented on curves and under adverse conditions. Loss of control preceded 22% of RD crashes and 69% of LD crashes. Beyond crashes and near-crashes (CNCs), the algorithm identified approximately 3 million LCs and 0.3 million LDs/RDs. LCs typically involved larger crossing angles that decreased with speed, while departures clustered within 0°–2°. Compared with CNCs, these occurred at higher speeds and smaller angles. LKA consistently reduced lateral variability without biasing the mean offset.
Ali, GibranTerranova, PaoloWilliams, VickiHolley, DustinSaffy, JoshuaAntona-Makoshi, JacoboKefauver, KevinShull, EmilyLi, EricVenegas, Michael
The global automotive industry has reached a new era. If 2025 was defined by the cautious exploration of “experimental pilots” and the collection of vast data lakes from connected vehicle fleets, 2026 marks the year that data finally gains a mind of its own within the assembly plant. We are witnessing a transition from passive automation to integrated, agentic autonomy. This is a shift that moves beyond simple programmed robotic arms and toward systems capable of independent reasoning and real-time optimization. This evolution is not just a technical upgrade; it is a fundamental restructuring of how vehicles are built, de-risked, and scaled in an increasingly volatile global economy.
Panigrahi, Dijam
The scope of this document is to provide considerations, guidelines, and best practices for extracting knowledge from long-term archival data. The document is intended to cover the data generated across all life cycle stages of an aircraft starting from concept to disposal. This document does not standardize the process, nor does it allow regulatory authorities to recognize the document as an acceptable means of compliance. It is only a guideline document to discover, capture, store, retrieve, process, and consume knowledge.
G-31 Digital Transactions for Aerospace
The automotive industry is continuously evolving at high pace to meet rising customer expectations, reliability, reduced maintenance, and most relevant, compliance with stringent emission norms. Traditionally, the analysis of vehicle emissions relies heavily on periodic inspections and manual checks. These conventional methods are often time-consuming, prone to human error, and lack the ability to provide real-time insights. Also, identifying failures due to non-manufacturing issues require meticulous physical inspections and historical data reviews, which are not always accurate or timely. Telematics or Connected cars technology being one of the major technological innovations in recent times revolutionizes these processes by enabling real-time data exchange between vehicles and external systems. The current study presents an innovative approach to utilizing telematics data for real-time monitoring of vehicle emissions and pinpointing Catalytic converter failures by analyzing vehicle probe data retrieved from telematics system aimed to identify fuel adulteration events or CNG kit retrofitments that can compromise vehicle performance and longevity. The methodology involves continuous data transmission from telematics devices to the cloud, where the system monitors vehicle emissions in real-time and alerts customers of potential failures. Further to identify the cause of failure, the telematics raw data is processed and aggregated for analysis using statistical models to detect potential fuel tank cleaning due to incorrect or adulterated fuel filling done in the past. This process is validated through a two-level model, ensuring accuracy in detecting fuel adulteration instances. The key advantage of this approach lies in its server-based high-speed processing, which eliminates the resource burden involved during physical inspection and testing of failed parts and enhances detection capabilities compared to existing solutions. This innovative method not only improves vehicle maintenance and customer satisfaction but also ensures compliance with emission norms, thereby contributing to a cleaner and more sustainable environment.
Dev, TriyambakPrasad, Kakaraparti AgamKalkur, VarunModak, SaikatAGARWAL, ShashankChandra, AnimeshPaul, VarshaGarg, AmitSundararaman, VenkataramanBose, Sushant
Addressing climate issues is a key aspect of good global governance today. A key aspect of managing the threats caused to the environment around is to ensure a sustainable transportation system so that humans exist in peace with nature. According to sources, in 2020 alone, cars accounted for approximately 23% of global CO2 emissions. In addition, they also emit dangerous pollutants thus damaging the ecosystem. To keep pollutants in check there are emission level testing strategies in place in each country. However, we can do better for a sustainable future. On one hand, the huge volume of vehicles around the world makes it an excellent choice and source for a vast emission level dataset comprising of input features as well as the target variable representing the emission band of the vehicle. In addition to the big data available as mentioned above, major advancements in the machine learning algorithms are done today. The advent of algorithms such as Artificial Neural Networks (ANN) has made it possible to develop models with very high accuracy. In this paper, the authors therefore propose the application of Extreme Learning Machines (a type of feedforward neural networks) to solve the pressing challenge of classifying the emission band of a vehicle which can be used by agencies to ensure that healthy vehicles operate on road at large. Extreme Learning Machines (ELM), by definition, is an excellent choice for emission band prediction task as it offers a significant reduction in training time and thereby is scalable such as to the task confronted with in this paper. Results from the metric, namely classification accuracy, are discussed at length. The highly accurate trained model, thus developed, can be used by agencies to then predict the emission band for any given vehicle scenario thus complementing the strategies already in place today for a greener earth.
Sridhar, SriramAswani, Shelendra
The article is devoted to a comprehensive analysis of the digital transformation of education using the example of a project to train engineering personnel for the innovative transport industry in Russia. Special attention is paid to the introduction of hybrid formats, digital platforms, inclusivity, issues of digital inequality, as well as the experience of the National Research Center of the Russian Federation FSUE NAMI and interaction with leading universities in the country. A comparative analysis with foreign initiatives, including modern AI solutions for inclusive education, is presented, as well as the impact of the project to create educational and methodological centers on the professional motivation of teachers.
Shishanov, SergeiKurmaev, RinatRevenok, Svetlana
The traditional Battery Management System (BMS) faces certain limitations in fully utilizing battery capacity and performance during the long cycle life operation of Electric Vehicles (EVs). These constraints include limited real-time data collection, low processing speed, lack of predictive maintenance, and minimal accuracy in predicting health and degradation chemistry. A Battery Digital Twin (BDT) can effectively address these limitations of the BMS. Battery Digital Twins (BDT) can be viewed as a cyber-physical system comprising four key elements: virtual representation, bidirectional connection, Simulation, and connection across the life cycle phases of an EV battery. The performance of a Li-ion battery largely depends on the cathode chemistry, component design, and operating conditions. The battery should be manufactured in a manner (such as cylindrical or prismatic cell) that prevents explosion, leakage, and gas generation inside the battery. To enhance the performance and safety of the battery, sensor data, including current, voltage, and temperature, can be continuously monitored through external measurement devices to generate the battery's State of Charge (SoC). Individual battery parameters such as state of charge, power, energy, health, and safety can continuously send data to a real-time monitoring system. An advanced BDT can contribute to enhanced computational capacity, real-time data collection and analysis, visualization, predictive maintenance, life cycle management, failure prediction based on degradation chemistry, and ML algorithms for various OEMs. Recent developments in key technologies have facilitated the development of innovative features within the Digital Twins (DT) system, such as Big Data for real-time fast and accurate analysis, AI/ML for model training and decision-making, the Internet of Things for live communication, cloud for storage and fast computation, and Blockchain for Life Cycle Management (LCM)/Battery Passport. In this review, we have systematically integrated the early adoption of BDT models and their systematic advancement with continuously evolving physical and AI/ML-based models.
Chaturvedi, VikashM, VenkatesanLanke, SiddhiSubramaniam, AnandKarle, ManishPandit, RugvedGupta, DrishtiKarle, Ujjwala Shailesh
In recent years, the market size of cold chain transportation in China has been expanding, but the industry has problems such as low cold chain circulation rate, low efficiency, high damage rate, and high cost. Under the background of reducing costs and improving quality and efficiency in transportation and logistics, an index set for operational analysis covering average freight rates, daily average number of over-temperature alarm incidents, daily average driving distance, and daily average driving time was established from the perspectives of economic efficiency, quality, and efficiency. Based on data from a third-party platform, including vehicle trajectories, temperatures, speeds, and freight rates, the running situation of road cold chain transportation industry was analyzed. The analysis results show that in 2023, the average freight rate of China’s highway cold chain will rebound, the fluctuation range will significantly narrow, the standardization level of temperature control equipment will continue to improve, the over temperature alarm during transportation will be significantly improved, and the vehicle utilization efficiency and traffic efficiency will be improved.
Li, SicongYe, JingCao, Mengfei
We present a novel processing approach to extract a ship traffic flow framework in order to cope with problems such as large volume, high noise levels and complexity spatio-temporal nature of AIS data. We preprocess AIS data using covariance matrix-based abnormal data filtering, develop improved Douglas-Peucker (DP) algorithm for multi-granularity trajectory compression, identify navigation hotspots and intersections using density-based spatial clustering and visualize chart overlays using Mercator projection. In experiments with AIS data from the Laotieshan waters in the Bohai Bay, we achieve compression rate up to 97% while maintaining a key trajectory feature retention error less than 0.15 nautical miles. We identify critical areas such as waterway intersections and generate traffic flow heatmap for maritime management, route planning, etc.
Kong, XiangyuShao, Guoyu
Amid escalating global warming challenges, the aviation industry must adopt low-carbon and green practices. China, aiming to meet its dual carbon goals, urgently requires enhanced research and development in sustainable aviation fuels (SAF), including their sustainability certification. However, China’s regulatory framework and limited research foundation in biofuels exacerbate this endeavor. This article summarizes the development status of SAF sustainability certification internationally and within China, encompassing the indicator framework, full life cycle greenhouse gas (GHG) calculation methodologies, and emission reduction thresholds. It also highlights issues encountered in the application of current international sustainability certification systems in China, such as high certification costs and inadequate data security. Advancement in domestic sustainability certification in China faces obstacles related to the incomplete foundational database, despite possessing life cycle assessment (LCA) calculation capabilities. To address these challenges, it is imperative to expedite the development of SAF certification systems, research in big data tracking systems, and establish targeted international mutual recognition data tracking platforms. Furthermore, enhancing GHG reduction thresholds in SAF sustainability certification is crucial. These steps will expedite SAF adoption in China, significantly contributing to global decarbonization efforts.
Zhang, ShupingHe, YinJia, QuanxingJia, QinTao, ZanMiao, JiaheShi, YaoZhang, XiangpingWang, Siyu
Warranty claims function as primary source of characterizing field failures across industries, wherein appropriate classification of these claims is critical for further analysis. The classification of warranty claims is a highly laborious effort, involving significant man-hours of warranty analysts. This can be highly optimized and made efficient using direct interpretation of the claim data on 3D model using unity game engine. Additionally, the color perception technique using immersive technology (AR/VR) can help to identify the vital few & drive prioritization of the field failures leading to faster problem resolution. The capabilities of UI/UX & advanced visualization are integrated to develop novel methods to classify the warranty claims & interpret it on a 3D model using immersive technology which is novel and one of its kind in industry. Unique characteristics of this tool is it focuses on the warranty claim classification by claim cost & count of claims and presents the heat map (Red-High, Yellow- Moderate, Green- Low, Gray – No claims) which helps in faster visualization of the claims on actual product. We have also integrated the respective corrective action associated for a failure mode which can help us to understand the status of actions plan to be deployed / closed and track for its future effectiveness. The NLP algorithm has been trained through a multi-representative training dataset which covers all the failure categories based on experiential knowledge. Cross-validation technique is used for optimizing the machine learning algorithm parameters. The test datasets consisting of different mix of warranty claims were then fed into the model wherein the prediction accuracy of more than 90% achieved over multiple runs consistently.
Nankery, Viveksavadatti, SandeepShete, AtulApkare, SanketGanapathi, Poongundran
In-Use emission compliance regulations globally mandate that machines meet emission standards in the field, beyond dyno certification. For engine manufacturers, understanding emission compliance risks early is crucial for technology selection, calibration strategies, and validation routines. This study focuses on developing analytical and statistical methods for emission compliance risk assessment using Fleet Intelligence Data, which includes high-frequency telematics data from over 500K machines, reporting more than 1000 measures at 1Hz frequency. Traditional analytical methods are inadequate for handling such big data, necessitating advanced methods. We developed data pipelines to query measures from the Enterprise Data Lake (A Structured Data storage system), address big data challenges, and ensure data quality. Regulatory requirements were translated into software logic and applied to pre-processed data for emission compliance assessment. The resulting reports provide actionable insights on NOx sensor activity, engine warmup operations, high-risk drive cycles, and load profiles across different operation regimes. This approach significantly reduces the reliance on costly and labor-intensive physical testing with Portable Emissions Measurement Systems (PEMS) by integrating advanced analytical methods into the workflow. By leveraging high-frequency telematics data, this method enables engineers to identify failed machines in the field more efficiently. It also provides valuable insights and reasoning behind these failures, facilitating quicker and more informed decision-making. This not only enhances emission compliance monitoring but also optimizes resource allocation and reduces overall regulatory risks. In summary, the developed methods enable effective emission compliance monitoring, reduce regulatory risks, and help optimize calibration strategies by understanding customer usage patterns. These methods are scalable for various emission regulations.
Arya, Satya PrakashShekarappa, Kiran
Use Decision Making Trial and Evaluation Laborator (DEMATEL) and Analytic Hierarchy Process (AHP) to jointly analysis and determine the key factors of Guangzhou intelligent logistics. Through the questionnaire survey of 92 logistics enterprises in Guangzhou, it is concluded that Information infrastructure, big data, Internet of Things, artificial intelligence, Logistics dynamic updates, and Smart warehousing have a great impact on intelligent logistics. Combining practical engineering with theory to make the implementation of Guangzhou’s smart logistics project more scientific, It is characterized by a higher degree of scientificity. Moreover, it is of great warning value, which can alert relevant parties to potential issues. Meanwhile, it provides essential guidance for the implementation of the smart city project in Guangzhou, facilitating a more efficient and well - directed execution process. This study is limited to logistics business respondents in Guangzhou and may limit the generalizability of the survey results.
Zhang, ShuangshuangChen, NingKhaw, Khai WahLiu, ChenxiJin, Lili
Smart airport is a key driver for the future development of civil aviation and a cornerstone of China’s ongoing “Four Airport” construction initiative. It is important to improve technology in many areas. This includes airport building, daily work, management, and making decisions. As air travel changes, using new tools like artificial intelligence, big data, and the Internet of Things (IoT) is very important. These tools help make airports more efficient, safe, and better for the environment. Because of this, building smart airports is not just a big goal but also a new way to deal with the challenges in today’s air travel systems. A key part in building smart airports is making a full evaluation system to check how well the projects are working. When a strong index system is made for smart airports, people involved can see clearly what is working well and what is not. So, chose using a three-scale hierarchical analysis method gives a clear and step-by-step way to look at different parts of smart airport building. This method helps check many things, like how well the technology fits in, how smoothly the airport works, how it affects the environment, and how users feel [1]. To check if this evaluation framework works in real life, a case study of Beijing Daxing International Airport is used. Daxing Airport is one of the top examples of smart airport building. It uses new technology and modern ways to manage the airport. So it is a good choice for this study. The case study shows that the evaluation system can work. It also gives useful advice for airport managers to make construction projects better.
Li, Shi-lingFu, Lu
In view of the complexity of railway engineering structure, the systematicness of professional collaboration and the high reliability of operation safety, this paper studied the spatial-temporal information data organization model with all elements in whole domain for Shuozhou-Huanghua Railway from the aspect of Shuozhou-Huanghua Railway spatial-temporal information security. Taking the unique spatial-temporal benchmark as the main line, the paper associated different spatial-temporal information to form an efficient organization model of Shuozhou-Huanghua Railway spatial-temporal information with all elements in the whole domain, so as to implement the effective organization of massive spatial-temporal information in various specialties and fields of Shuozhou-Huanghua Railway; By using GIS (Geographic Information System) visualization technology, spatial analysis technology and big data real-time dynamic rendering technology, it was realized the real-time dynamic visualization display of different spatial-temporal information and organization relationship of Shuozhou-Huanghua Railway, which provided auxiliary decision-making and spatial information technology support for independent prediction and early warning of Shuozhou-Huanghua Railway operation safety situation.
Liu, KunYu, HongshengZhu, PanfengLiu, WenbinWang, Yaoyao
Management of battery systems for electric vehicles has great importance to ensure safe and efficient operation. State-of-Charge and State-of-Health (SoH) are fundamental parameters to be taken under control even though they cannot be directly measured during vehicle operation. Some control approaches have gained increasing interest thanks to advances in sensor availability, edge computing and the development of big data. In particular, SoH estimation through machine learning (ML) and neural networks (NNs) has been thoroughly investigated due to their great flexibility and potential in mapping non-linear relations within data. The numerous studies available in the literature either employ different extracted features from data to train NNs, or directly use measurement signals as input. Additionally, many studies available in the literature are based on a limited number of publicly available datasets, which mainly encompass cylindrical battery cells with small capacity. Starting from the workflow analysis for developing and implementing ML SoH estimators, this work aims to give an overview of the latest application studies in this field, with a special focus on the analysis of the main datasets available in the literature. In the end, the workflow for the implementation of NN-based SoC estimation is demonstrated with a step-by-step procedure on a publicly available dataset, and a final comparison with non-neural regression algorithms is performed.
Chianese, GiovanniCapasso, ClementeVeneri, Ottorino
Big Data technologies have become quite ubiquitous in the last years, allowing for the storage of substantial amounts of data, typically flight test data as recorded by the flight test installation. On recent helicopter prototypes, we generate in excess of 50 GB of raw data per flight hour, usually in a format not adequate for efficient large-scale processing. With some specific optimizations and the setup of a specialized infrastructure, there are now practicable means to store timeseries in ways that allow for requests spanning hundreds or thousands of flights to complete within minutes, opening the way to some substantial savings and new insights. However, to make the most of these data and make informed decisions it is often quite important to store contextual data that go beyond the pure timeseries data, typically on helicopters where optional installations can have a significant impact on aircraft performance or behavior. This paper explores the various kinds of data and metadata related with flight tests, how to collect them and relate them with one another in order to maximize raw data value and avoid some common pitfalls. We define more accurately the data of interest, where to collect them and some ideas to further improve data collection in the future.
Brisset, Nicolas
The automotive aerodynamic development relies on wind tunnel testing and Computational Fluid Dynamics (CFD), where the former provides reliable values to be used for fuel economy calculations, and the latter enables the investigation of flow features responsible for improvement/degradation of the average large-scale performances in terms of aerodynamic coefficients. The abovementioned procedure overlooks a crucial factor however: natural wind. The speed and the direction of natural wind encountered while driving alters the vehicle’s effective yaw angle. Such condition implies that the minimization of the drag coefficient at zero-yaw, commonly performed through wind tunnel and CFD simulations in an industrial context, may not yield real-world optimal shapes. While it is possible to reproduce natural wind-like conditions in a wind tunnel using flaps, for example, the input signal to the flap system must be available beforehand, and such key element is the focus of the present research effort. In this paper, we introduce a methodology behind the selection of a statistically valid test course for wind data sampling on US public roads. Big Data from the US road traffic network and the wind distributions measured at weather stations across the country are used to build a representative country-wide distribution, which is then compared with the local distribution on the road, according to the weather station data. The visualization of the divergence on a road map allows the selection of a suitable course for data collection, providing a sound rationale for downstream fuel economy calculation tasks.
Nucera, FortunatoOnishi, YasuyukiMetka, Matt
Overloading of trucks will not only damage road infrastructure, lead to exhaust pollution, and even cause serious traffic accidents, resulting in huge losses of life and property. However, most of the methods to evaluate truck overloading are limited by environmental factors, so it is impossible to monitor truck overloading in real time. In order to solve this problem, a truck overload detection method based on real-time vehicle diagnosis big data is proposed in this paper. The method comprehensively considers multiple factors affecting the actual power of trucks through mathematical modeling. It based on the effects of overload on fuel combustion efficiency, harmful gas emission, exhaust temperature, and vehicle power loss, The truck overload evaluation model is constructed to judge whether the truck is overloaded or not in real time. Based on the truck overload assessment and truck accident risk factor extraction , a real-time operation risk assessment model based on fault tree analysis is developed to evaluate the safety of overloaded and non-overloaded trucks. The fault tree model is mapped to a Bayesian network model and transformed into equivalent network model by Netica software. The network model is analyzed qualitatively and quantitatively, and the key factors that have great influence on the accidents, such as bad driving conditions, dangerous driving behavior and poor visibility. This research enhances the traffic management department's ability of monitoring and early warning of truck overloading, strengthens the deterrence and efficiency of overload control measures, helps to reduce the occurrence of overloading. This ultimately improves road transport safety, reduces exhaust emissions and environmental pollution caused by overloading.
Chen, YuguangLin, HonghaoWang, Yanan
The operation management of electric Taxi fleets requires cooperative optimization of Charging and Dispatching. The challenge is to make real-time decisions about which is the optimal charging station or passenger for each vehicle in the fleet. With the rapid advancement of Vehicle Internet of Things (VIOT) technologies, the aforementioned challenge can be readily addressed by leveraging big data analytics and machine learning algorithms, thereby contributing to smarter transportation systems. This study focuses on optimizing real-time decision-making for charging and dispatching in large-scale electric taxi fleets to improve their long-term benefits. To achieve this goal, a spatiotemporal decision framework using Bi-level optimization is proposed. Initially, a deep reinforcement learning-based model is built to estimate the value of charging and order dispatching under uncertainty. The model considers the long-term costs and benefits of different tasks and guides whether electric taxis should prioritize charging or order dispatching for the fleet's long-term benefits. Subsequently, a combinatorial optimization approach is employed to determine the specific targets for charging or order dispatching. Case studies are conducted within real-world operation data from electric taxis in Hangzhou City, China. The results validate the efficacy of the proposed method, as compared to a baseline approach. Across various fleet sizes and charging power conditions, the method significantly reduces non-service time during the charging process by optimizing charging time and location. The proposed method is found to be suitable for large-scale fleets and high-charging power scenarios.
Lyu, YelinWang, NingTian, Hangqi
In recent years, the automotive industry has been making efforts to develop vehicles that satisfy customers’ emotions rather than malfunctions by improving the durability of vehicles. The durability and reliability of vehicles sold in the U.S. can be determined through the VDS (Vehicle Dependability Study) published by JD Power. The VDS is index which is the number of complaints per 100 units released by J.D. POWER in every year. It investigates customers who have used it for 3 years after purchasing a new car and consists of 177 specific problems grouped into 8 categories such as PT, ACEN, FCD, Exterior. The VDS-4 has been strengthened since the introduction of the new evaluation system VDS-5 in 2015. In order to improve the VDS index, it is important to gather various customer complaints such as internet data, warranty data, Enprecis data and clarify the problem and cause. Enprecis data is survey of customer complaints by on-line in terms of VDS. In the case of warranty and Enpreics data, it is easy to analyze because it is already categorized, but internet data is difficult to classify because it is unstructured data collected randomly at various internet sites, and the amount of data is big. In this paper, we developed classification technology for internet data using deep learning method such as TF-IDF and Word2Vec. This technology automatically classifies 8 categories and extract keywords of each category even if you don’t read the articles.
You, Hanmin
Vehicle efficiency and range, along with the DC charging speed, are deemed as the most important criteria for an electric vehicle currently. The electric vehicle energy consumption is impacted by the change in temperature along with the driving style and average speed of a customer, all other factors being constant. Hence understanding the patterns and impact of different aspects of an EV range & charging speed is crucial in delivering an electric vehicle with robust efficiency across all weather conditions. In this paper we have analysed vehicle parameters of global Jaguar I-PACE customer data. We present and analyse the collated big data of around 50,000+ unique vehicles with a data aggregate of well over 482 million km. In moderate ambient conditions the analysis indicated a good correlation with 50th to 75th percentile drivers’ energy consumption to the EPA label figure. The EPA hot and cold ambient tests also compare well but the correlation is sensitive to long and short trip distances. The cumulative data of the global fleet, for 75th percentile customers, shows that the total consumption of the vehicle increases by 81% and 47%, from the median energy consumption at 20°C, at -10°C and 0°C respectively. Similarly, the global fleet energy consumption, for 75th percentile customers, increased by 7% and 31% respectively, from the median energy consumption at 20°C, at 30°C and 40°C respectively. The paper then deep dives into data bins analysing the consumption sources for certain key drive and ambient scenarios.
Dutta, NilabzaEvans, Davidsapte, Atharva
It is an important factor in electric vehicles to show customers how much they can drive with the energy of the remaining battery. If the remaining mileage is not accurate, electric vehicle drivers will have no choice but have to feel anxious about the mileage. Additionally, the potential customers have range anxiety when they consider Electric Vehicles. If the remaining mileage to drive is wrong, drivers may not be able to get to the charging station and may not be able to drive because the battery runs out. It is important to show the remaining available driving range exactly for drivers. The previous study proposed an advanced model by predicting the remaining mileage based on actual driving data and based on reflecting the pattern of customers who drive regularly. The Bayesian linear regression model was right model in previous study. In addition, in order to improve performance, the driver's regular driving pattern is recognized in advance before driving and it is reflected in the remaining driving mileage model with Bayesian regression. It could be seen that the performance of the model in previous study was improved 10% better compared to the remaining driving mileage existed in vehicle function. The purpose of this study confirms the robust performance with vast data of the more vehicles in long mileage. The big data are from that almost vehicles took drive for 10k mile range. The vehicles had various experience in location, driving time rage, mileages, driving pattern, charge pattern, and so on. Moreover, the personalized model gets improved in big data to get higher accuracy. This study proposes personalized algorithm enable to embed in vehicles.
Joo, Kihyungkim, Lina
This paper presents deep learning-based prognostics and health management (PHM) for predicting fractures of an electric propulsion (eP) drivetrain system using real-time CAN signals. The deep learning algorithm, based on autoencoders, resamples time-series signals and converts them into 2D images using recurrence plots (RP). Subsequently, through unsupervised learning of DeepSVDD, it detects anomalies in the converted 2D images and predicts the failure of the system in real-time. Also, reliability analysis based on fracture mechanics was performed using the detected signals and big data. In particular, the severity of the eP drivetrain system is proportional to the maximum shear stress (τmax) in terms of linear elastic fracture mechanics (LEFM) and can be calculated by summarizing the relationship between cracks (a) and the stress intensity factor (KIII). During this process, the system status can be checked by comparing the stress intensity factor and fracture toughness (KIIIc), and the time from the detection of an abnormal signal in the system to complete failure can be quantitatively determined. Therefore, it is possible to continuously maintain the status of the system by detecting failure signals using deep learning before vehicle parts fail, and with the detected failure prediction signals, a process can be established to enable users to repair defects in the vehicle system before breakdown occurs. By predicting the remaining life of the system and calculating field reliability through these procedures, we introduce innovative technologies aimed at preventing safety accidents, reducing economic costs, and addressing quality issues. In the future, we expect to achieve high business performance by extending and applying this deep learning-based PHM approach to all vehicle components.
Moon, ByungwooLee, SangWonNam, DongJinKim, JeonghwanBae, JaeWoongShin, JeongMin
The term Industry 4.0 is well known in contemporary automotive landscape. It encompasses a smart integrated framework of IIoT (Internet of Things) and industrial automation with machine learning, artificial intelligence and big data analytics to arrive at optimal solutions to running the processes in a streamlined, efficient and effective manner. Industry 4.0 has assumed critical significance in the contemporary era of people working from remote locations to operate processes in order to build products, thereby ensuring business continuity. Consequently, it follows that if industry 4.0 is applied to automotive homologation activity, it will lead to a standardized evaluation, consistent fidelity of testing, accurate judgement of the product under test with regards to its certification, and most importantly, timed delivery to release in the market. The author hereby elucidates a unified Industry 4.0 Framework for Automotive homologation in India which is the need of the hour. This framework consists of five elements. The first is use of IIoT to ensure optimal use and maintenance of the homologation machinery. The next is use of industrial automation to drive the testing in a streamlined manner and make it consistent. Further, machine learning can be deployed to calibrate the learning of the various variables and parameters. Next, artificial intelligence can be deployed to arrive at the best practices of efficient and effective use of the machines. Finally, big data analytics can facilitate insight into the outcome of the test and establish correlation between the output homologation criteria with respect to the test setup parameters. An example is showcased for expounding this methodology in the context of Indian Homologation scenario. The example showcases the tremendous benefits that can be reaped by this framework in terms of efficient machine utilization, empowering and enriching operator knowledge, making the processes streamlined and productive and lending high confidence to the product certification activity, thus holistically overhauling the entire homologation program in a revolutionary manner, much to the benefit of all stakeholders involved therein.
S Thipse, Yogesh
The Auto industry has relied upon traditional testing methodologies for product development and Quality testing since its inception. As technology changed, it brought a shift in customer demand for better vehicles with the highest quality standards. With the advent of EVs, OEMs are looking to reduce the going-to-market time for their products to win the EV race. Traditional testing methodologies have relied upon data received from various stakeholders and based on the same tests are planned. The data used is highly subjective and lacks variety. OEMs across the world are betting big on telematics solutions by pushing more and more vehicles with telematics devices as standard fitment. The data from such vehicles which gets generated in high levels of volume, variety and velocity can aid in the new age of vehicle testing. This live data cannot be simply simulated in test environments. The device generates hundreds of signals, frequently in a fraction of seconds. Multiple such signals can be combined to create KPIs that correspond to specific traits of vehicle health. Specific telematics KPIs can be used as inputs in test benches. With this data, the vehicle gets tested against real-world conditions. Another important aspect of vehicle testing is road conditions. OEMs can only do so much in testing the vehicles in various different road conditions. However, with telematics devices road condition data can be directly generated corresponding to the international roughness index. This data along with KPIs will bring in a new perspective of vehicle development testing and enable the manufacturer to better understand market problems and to take effective countermeasures. With these KPIs, Possibilities are limitless, and these data can be used to create a digital twin of the vehicle, enabling the OEM to assess the vehicle without even physically attending.
Sahoo, PriyabrataSingh, SaurabhPrasad, Kakaraparti Agam
Using current technologies, a single “entry level” vehicle has millions of electrical signals sent through dozens of modules, sensors and actuators, and those signals can be sent over the air, creating a telemetry data that can be used for several ends. One electrical device is set up to have diagnosis, in order to make maintenance feasible and support repair, plus giving improvement directions for specialists on new developments and specifications, but in several cases the diagnosis can only determine the mechanism of failure, but not the event that triggered that failure. Current evaluation method involves teardown, testing and knowledge from the involved specialized team, but this implies in recovering of failed parts, which in larger automakers with thousands of dealers/repair shops, reduces the sample for analyses when there is a systemic issue with one component. This specificity is usual in Propulsions systems, regarding electro-mechanical devices, and sensors, also in electrochemical devices, such as batteries and others, when a systemic issue appears, the teardown reveals its failure, but now why it failed. Based on that information and needs a methodology using big data mining and tools combined with available telemetry data in order to detect statistically main events or contributors/variables that triggers a failure event. That sort of methodology is helpful and more agile since it doesn’t depend on recovering of parts to give directions of which potential event may trigger a failure event, supporting in systemic/application comprehension of any component failure which uses electrical signals monitored within vehicle, and doesn’t depend on extraction of failed components, it can use and consider every single failed vehicle, for one specific component, as basis for analyses and identification of failure event, which will support in systemic correction/improvement and adjustment/improvement of specification for future and specific developments.
Prazeres, ChristopherHachyia, AfonsoTakahashi, Marcio
Heavy vehicles are major fuel consumers in road transportation, and the traditional way to reduce fuel consumption is to reduce weight, resistance, improve mechanical transmission efficiency, and improve engine thermal efficiency. However, European heavy-duty truck companies took the lead in realizing predictive cruise control (PCC) technology on the basis of cruise through intelligent network technology, based on ADAS maps, and achieved good fuel saving effects. In this paper, by studying the fuel consumption characteristics of trucks, designing the dynamic parameters of the load and whole vehicle, the predictive adaptive cruise control (PACC) technology is realized based on the predictive cruise strategy, and the statistics of fuel saving rate under different cruise ratio conditions are analyzed through the big data platform.
Qian, GuopingLu, ZhenghuaTian, JuntaoLiu, LianfangXi, ChongZhou, Xiaoying
Synthesized driving cycles which can reflect the real world driving scenarios are essential for electrification and hybridization of powertrains of heavy duty logistics vehicles (HDLV). Current synthetic methods always neglected weight variation which is crucial for logistic vehicle driving scenarios. This paper proposed a method based on multi-dimensional Markov chains and big data to generate typical driving cycles with consideration of vehicle weight and slope. The validation of the synthesized driving cycle was based on a statistical analysis and the adequacy of the representative to real world driving data was demonstrated.
Liu, Zemin EitanLi, YongTan, GuikunXu, LubingShuai, Shijin
Numerous researchers are committed to finding solutions to the path planning problem of intelligence-based vehicles. How to select the appropriate algorithm for path planning has always been the topic of scholars. To analyze the advantages of existing path planning algorithms, the intelligence-based vehicle path planning algorithms are classified into conventional path planning methods, intelligent path planning methods, and reinforcement learning (RL) path planning methods. The currently popular RL path planning techniques are classified into two categories: model based and model free, which are more suitable for complex unknown environments. Model-based learning contains a policy iterative method and value iterative method. Model-free learning contains a time-difference algorithm, Q-learning algorithm, state-action-reward-state-action (SARSA) algorithm, and Monte Carlo (MC) algorithm. Then, the path planning method based on deep RL is introduced based on the shortcomings of RL in intelligence-based vehicle path planning. Finally, we discuss the trend of path planning for vehicles.
Hao, BingZhao, JianShuoWang, Qi
In view of the structural accidental events in the ongoing airworthiness stage of civil aircraft, it is necessary to conduct a risk assessment to ensure that the risk level is within an acceptable range. However, the existing models of risk assessment have not effectively dealt with the risk of accidental structural damage due to random failure. This article focuses on probabilistic risk assessment using the Transport Airplane Risk Assessment Methodology (TARAM) of accidental structural damage of civil aircraft. Based on the TARAM and probability reliability integral, a refined failure frequency probability calculation model is established to elaborate on composite structure failure frequency. A case study is analyzed for the outer wing plane of an aircraft having impact damage of composite materials. Finally, results of the risk assessment without correction and risk assessment with correction are presented for detailed visual inspection and general visual inspection.
Jia, BaohuiFang, JiachenLu, XiangXiong, Yijie
The automotive industry is going through one of its greatest restructuring, the migration from internal combustion engines to electric powered / internet connected vehicles. Adapting to a new consumer who is increasingly demanding and selective may be one of the greatest challenges of this generation, Original Equipment Manufacturers (OEM) have been struggling to keep offering a diversified variety of features to their customers while also maintaining its quality standards. The vehicles leave the factory with an embedded SIM Card and a telematics module, which is an electronic unit to enable communication between the car, data center. Connected vehicles generate tens of gigabytes of data per hour that have the potential to be transformed into valuable information for companies, especially regarding the behavior and desires of drivers. One of the techniques used to gather quality feedback from the customers is the NPS it consists of open questions focused on top-of-mind feedback. Here is where AI and ML comes into play, using NLP and several other computational techniques to download, extract, structure, read, process, understand and categorize all this data into specific predetermined categories, allowing engineers to accelerate fixing quality issues and improving user experience. The ML model developed in this article identify costumer complains in an enormous data lake and groups them into categories. After a significative amount of data is collected and grouped into it enables the algorithm to predict future trends and together with real time connected vehicle data the model can alert the responsible engineers to develop an action to solve the problem without more customers even actually experience the failure. The ML algorithm is still on its development phase, but the initial results are promising, we have successfully processed more them 6 millioncustomers feedback finding problems with precision and accuracy close to 90%.
Torres Fernandes Veiga, Daniel Thadeude Miranda Junior, Airton WagnerNascimento Silva, LuanaSena Cavalcante, Mairondos Santos, Maria da Conceição
The main purpose of this research is to identify how the established quality methodologies, known worldwide as TQC (Total Quality Control) and TQM (Total Quality Management) are supported by the tools of the Quality 4.0 concept that similarly received influence from the disruptive technologies of Industry 4.0 in the last decade. In order to crosscheck the relationship among TQC and TQM and how Quality 4.0 supports these quality systems a qualitative investigation method was adopted through a survey questionnaire applied to one of the most important worldwide automobile company, based also in Brazil, Toyota of Brazil. Based on a literature review and relationship of concepts and synergy among them it was possible analyse and find out conclusions of this research work. The main results were identified as TQC and TQM are very well established concepts of quality and Quality 4.0 concepts and tools have been implemented on a path according to the markets importance prioritization, so then Toyota of Brazil is implementing in a slow motion, but started up using technologies of Industry 4.0 on Quality, promoting a leverage of excellence on quality function overall organization and settling down this Quality 4.0 level.
da Silva Bento, NelsonCavalcanti Bortoleto, WilliamIbusuki, Ugo
Scenario-based testing is a promising approach to solving the challenge of proving the safe behavior of vehicles equipped with automated driving systems (ADS). Since an infinite number of concrete scenarios can theoretically occur in real-world road traffic, the extraction of scenarios relevant in terms of the safety-related behavior of these systems is a key aspect for their successful verification and validation. Therefore, a method for extracting multimodal urban traffic scenarios from naturalistic road traffic data in an unsupervised manner, minimizing the amount of (potentially biased) prior expert knowledge, is proposed. Rather than an (elaborate) rule-based assignment by extracting concrete scenarios into predefined functional scenarios, the presented method deploys an unsupervised machine learning pipeline. The approach allows for exploring the unknown nature of the data and their interpretation as test scenarios that experts could not have anticipated. The method is evaluated for naturalistic road traffic data at urban intersections from the inD and the Silicon Valley Intersections datasets. For this purpose, it is analyzed with which clustering approach (K-means, hierarchical clustering, and DBSCAN) the scenario extraction method performs best (referring to an elaborate rule-based implementation). Subsequently, using hierarchical clustering the results show both a jump in the overall accuracy of around 20% when moving from 4 to 5 clusters and a saturation effect starting at 41 clusters with an overall accuracy of 84%. These observations can be a valuable contribution in the context of the trade-off between the number of functional scenarios (i.e., clustering accuracy) and testing effort. The possible reasons for the observed accuracy variations of different clusters, each with a fixed total number of given clusters, are discussed. The findings encourage the use of this type of data and unsupervised machine learning approaches as valuable pillars for the systematic construction of a relevant scenario database with sufficient coverage for testing ADS.
Weber, NicoThiem, ChristophKonigorski, Ulrich
The Advancement in Connected vehicles Technology in recent years has propelled the use of concepts like the Internet of Things (IoT) and big data in the automotive industry. The progressive electrification of the powertrain has led to the integration of various sensors in the vehicle. The data generated by these sensors are continuously streamed through a telematics device on the vehicle. Data analytics of this data can lead to a variety of applications. Predictive maintenance is one such area where machine learning algorithms are applied to relevant data to predict failure. Field vehicle malfunction or breakdown is costly for manufacturers’ aftermarket services. In the case of commercial vehicles, downtime is the biggest concern for the customer. The use of predictive maintenance techniques can prevent many critical failures by tending to the root cause in the early stages of failure. Engine overheating is one such problem that transpires in diesel engines. Overheating of an engine may lead to various catastrophic failures like a warped cylinder head or cracked cylinder. It is essential to curb such problems at the early stages to save on warranty costs and establish confidence regarding our product in the customer's mind. Here Gaussian Mixture Model is applied to cleaned data to obtain the Engine Coolant Temperature Distribution. based on which the vehicles showing Overheating trends are classified into a separate class. These vehicles are then monitored and Early Failure Alerts for Overheating are triggered in the system for taking proactive measures to prevent the failure.
Hiwase, Shrikant DeokrishnaJAGTAP, PRAMODKrishna, Dinesh
"Bosch Connected and Virtualized Development: Use of In-Vehicle Data to Optimize and Validate Braking Systems"132819/13/2022
"Virtual development of electronic braking systems Value creation through in-vehicle data and cloud services Connected cars generate in-vehicle data from its electronic control units and sensors about how they are used, where they are, and how they feel. The amount of in-vehicle data will grow further exponentially, with progress in higher levels of driving automation, personalization and electrification. Therefore, exploiting the chances of generated in-vehicle data will become a key theme for the value creation and new business models in the automotive and mobility industry. While this will only happen in small steps, it will come quickly and will have a huge and lasting impact on both industries. Bosch connected braking systems Connected braking systems (CBS) is a new and connectivity-based approach from Bosch. It aims to enable its automotive customers and itself to transform the way of developing braking systems by exploiting in-vehicle data continuously and remotely from start-of-design through calibration to validation. It is a generic approach not limited to braking systems only. It supports the validation of simulation models and therefore virtualized releases required for ambitious projects with less prototype vehicles. Carmakers gain benefits of shorter learning cycles and higher efficiency in the optimization and distribution of new software. Once the approach is successfully applied in the development phase, carmakers face the next big evolution step in exploiting in-vehicle data beyond series production has started. Field validation and exploration continues the connected braking systems approach in the field. Based on an edge computing architecture, an embedded data client access pre-defined internal component and system information that is typically not available on vehicle networks. Thereby, Bosch is following a smart instead of a big data approach and triggers data acquisition and preprocessing only after predefined events to limit busloads and data transmission volumes. Once the data is transmitted to the Bosch IoT (internet-of-things) cloud, a fully managed cloud service collects, processes and stores in-vehicle data. For analysis, data is queried using NoSQL or MongoDB database. Analytic results are visualized through dashboards and automatic reports. Bosch has established a strong partnership with its pilot customers using field validation and exploration for its braking systems in the field. Pre-defined braking system data of more than 10.000 vehicles is now continuously and almost in real-time sent to the Bosch IoT cloud. Based on the valuable field insights such as electrical, thermal and hydraulic loads, function activations or driver requests, Bosch and its customer are able to redefine specifications and to improve performance of braking systems. Field validation and exploration of in-vehicle data serves carmakers in a second step as a data source and key enabler for the development of new data based services such as predictive diagnostics, anomaly detection or even connected digital twins. "
Nesbitt, Richard
Cloud Control Platform, from Information System to Digital TwinSAE-PP-002959/7/2022
The move towards an automated driving system (ADS) is being driven by both potential benefits and challenges of the technology, such as telecommunication, vehicle industry, ITS and cybersecurity. Cooperative Vehicle-Infrastructure System (CVIS) is the core development of ADS while the establishment of Cloud Control Platform (CCP). The CCP integrated automation, 5G and AI technology to provide physical world, digital world and Internet of Vehicles world based on big data, algorithm and variety of traffic scenarios. The utilization of Digital Twin in CCP will cover the whole vehicle-to-cloud (V2C) and CVIS process, which will develop from fragmentation to integration and static to dynamic. It opens the door to real-time monitoring and synchronization of physical activities with the virtual reality. Digital twin, a novel digitalization paradigm of cyber physical systems, has been attracted interest over the past years. The real function of digital twin is to setup real-time link between physical and digital world and realize their connection, communication and operation. The CCP with digital twin will merge billing system, monitoring system and other individual information systems into a unified Twin Road Cloud Platform which is the Vehicle-Road Collaboration Cloud Control Platform fulfilling the needs of intelligent vehicles and smart roads. The intelligent vehicles upload date to the CCP server through on-board devices and 4G/5G cellular network. The intelligent vehicles received the data from CCP which creates a virtual world based on the received data and proposed models. The cloud computing supports the digital twin framework which benefit the smart transportation systems with high quality communication and the acceptable communication delays and packet losses.
Wang, Jian
The global big data market had a revenue of $162.6 billion in 2021.1 Data is becoming more valuable to companies than gold. However, this data has been used, historically, without contributors’ informed consent and without them seeing a penny from the discoveries the data led to. This article discusses how non-fungible tokens (NFTs) can provide a helpful tool for pharmaceutical companies to track contributed data and compensate contributors accordingly. NFTs are unique, untradable cryptographic assets that can be tracked on a blockchain. NFTs provide a unique traceable token that cannot be replicated, providing a perfect tool to store biodata. The term biodata refers to details regarding a patient’s history and behavioral patterns.
Deep neural network models have been widely used for environment perception of intelligent vehicles. However, due to models’ innate probabilistic property, the lack of transparency, and sensitivity to data, perception results have inevitable uncertainties. To compensate for the weakness of probabilistic models, many pieces of research have been proposed to analyze and quantify such uncertainties. For safety-critical intelligent vehicles, the uncertainty analysis of data and models for environment perception is especially important. Uncertainty estimation can be a way to quantify the risk of environment perception. In this regard, it is essential to deliver a comprehensive survey. This work presents a comprehensive overview of uncertainty estimation in deep neural networks for environment perception of intelligent vehicles. First, we provide a systematic and intuitive understanding of the classification and modeling of uncertainty and then summarize methods for uncertainty estimation in deep neural networks. Considering the research of epistemic uncertainty estimation as a study-worthy branch, the methods on epistemic uncertainty estimation are also illustrated in chronological order. Next, we present the application of uncertainty estimation in environment perception tasks including object detection, segmentation, trajectory prediction, depth estimation, optical flow, and so on. For these typical tasks, we make a detailed analysis in aspects of uncertainty type, baseline, or other features, in order to provide a macroscopical view of models. Finally, we give the outlooks for the uncertainty estimation in environment perception of intelligent vehicles in three aspects: epistemic uncertainty estimation with less computation, aleatoric uncertainty estimation with modified loss function, and uncertainty estimation for 3D perception based on LiDAR point cloud.
Yin, HuilinChen, ZhaoruYan, JunRigoll, Gerhard
Traditional methods of municipal domestic waste analysis and prediction lack precision, while most data’s sample size is not suitable for many neural networks. In this paper, combining the advantage of deep learning methods with the results of association analysis, a waste production prediction method TLSTM is proposed based on long short-term memory(LSTM). It is found that the most influencing factors are population, public cost, household and GDP. Meanwhile, the garbage production in Shanghai will continue to decline in the future, indicating the policy of refuse classification is effective. The R-square index and MSE index of the model were 0.55 and 76571.73 respectively, surpassing other state-of-the-art models. In cooperation with School of Environmental Science and Engineering at Shanghai Jiao Tong University, the dataset comes from the average data of the Shanghai Household Waste Management Regulation from 1980 to 2020. This research method has a certain guiding significance to both the related fields of municipal solid waste management and environmental planning and the application of neural network models in other fields.
Tu, YunXiao, Zi XinShen, Na
As the complexity of systems expands with increasing emphasis for digital transformation, the aerospace industry is generating big data to meet customer requirements. The ability to that data to solve challenging problems is limited by many factors, including the capabilities of current classical computing systems. Impact of Quantum Computing in Aerospace discusses how quantum computing systems offer (possibly quadratic to exponentially) greater computational power over classical computers. The power of quantum computing is tremendous and has many potential impacts on the aerospace industry; however, there are also many unsettled topics surrounding the future of the technology. Click here to access the full SAE EDGETM Research Report portfolio.
Walthall, RhondaDixit, Sunil
Kontron and Intel experts explain how rugged, modular COM Express solutions reduce complexity and allow retrofit of autonomous systems on heavy mobile equipment. Continually transformed with more than a century's advances in capabilities, hydraulics and fuel efficiency, today's heavy mobile equipment must also become more intelligent and better connected. Technologies such as artificial intelligence (AI), deep learning, big data, GPS, 5G and computer vision are proving their mettle - empowering far more efficient ways of carrying out unique and demanding tasks via advanced telematics, advanced driver assistance systems (ADAS) or varying levels of autonomy. Heavy mobile equipment (HME) that can gather and apply data in real time operates and makes decisions in ways that humans cannot. This evolution toward automation promises not only leadership for manufacturers of more advanced systems, but also increased safety, economy, efficiency and ecological compatibility.
London, JackThomas, Andrea
On the last generation of Airbus Helicopters rotorcrafts such as H175 or H160, dynamic systems data are collected in a systematic manner in order to perform advanced analytics. Main gearbox (MGB) oil temperature and oil pressure are key parameters to assess the overhaul status of the lubrication and cooling systems. This paper describes new ways of monitoring lubrication and cooling systems behavior, taking advantage of big data capabilities and advanced analytics such as machine learning and physical modeling based approaches.
Martin, OcéaneMermoz, EmmanuelMechouche, Ammar
Driver Assist Technologies are complex systems for which it can be difficult to objectively estimate customer experience in a repeatable and quantitative manner. We must assess the designed feature operation at a massive scale to better understand the eventual customer impact and cost of a variety of engineering decisions. We will present the Leveraging Aggregated Vehicle Analytics (LAVA) methodology for improved understanding of the impact of these dynamic and subjective problems by utilizing connected vehicle (CV) data. Several examples of the LAVA methodology will be discussed and examined in detail. Using the LAVA methodology, minimal and anonymized data collected from CVs can be used to answer many engineering decision questions with high confidence in a controlled and scientific manner.
Lerner, JeremyTayim, DinaPervez, NahidZwicky, Timothy
Items per page:
1 – 50 of 176