Fig 1 - uploaded by Ray Y. Zhong
Content may be subject to copyright.
Source publication
Radio Frequency Identification (RFID) technology has been widely used in manufacturing sites for supporting the shopfloor management. Huge amount of RFID-enabled production data has been generated. In order to discover invaluable information and knowledge from the RFID big data, it is necessary to cleanse such dataset since there is large number of...
Citations
... Second, in-memory level, to preprocess the collected data in order to prepare it for a specific use and facilitate their exploration. At this level, several techniques are used, such as missing values imputation (Ryu et al., 2020), noise treatment (Zhong et al., 2014), discretization, normalization, feature selection (Y. Zhang and Cheung, 2014) and aggregation (Obinikpo and Kantarci, 2019). ...
Smart systems are systems that rely on technological advancements to continuously adapt and improve in order to provide added-value to their users. Designing these systems in a coherent and methodical way is important to set the stage for highly interoperable and collaborative systems with the potential to propel the software community into an era of Systems of Systems. However, the different and numerous concepts, technologies and techniques that have been linked and used recently to develop these systems made this task a challenging endeavor. Indeed, the existing literature on the subject is focused on the technical aspects of developing smart systems with very little effort and thought to how these systems should be designed.To tackle this gap, this thesis proposes and develops a method, called AS3, to analyse and design smart systems. The method starts from a broad definition of the smart system and builds on it to define a smart system loop that provides an integrated view of the main entities that are present in a smart system and their interactions. This smart system loop builds on the adaptability loop as well as the main concepts from context-awareness and service orientation to cover the life cycle of the smart system. Supported by a product metamodel and a process model, the method then provides the intentions and strategies that can be followed in order to design context-aware service-based smart systems. To insure the continuous improvement of the system, the method supports recommendation to allow easy automation of the improvement while keeping the method user in the loop. To showcase the relevance and the efficacy of the AS3 method, this thesis includes a complete rundown of the method to design a system that deals with road security called SMARTROAD.
... Big Data refers to the data sets with so huge volume or complexity that typical data processing technologies or approaches cannot deal with in an efficient and effective way (Ciobanu et al., 2014;Tan et al., 2015;Zhong, Huang, Lan, Dai, Xu and Zhang, 2015;Zhong, Huang, Lan and Wang, 2015;Zhong, Xu, Chen and Huang, 2017). Considering the large data sets from E-commerce logistics, several challenges may be address by using the Big Data Analytics because the data from this domain have some characteristics such as heterogeneous format, multi-dimensional attributes, real-time generation, and different source providers (Zhong et al., 2014;Pang et al., 2015). Big Data Analytics is able to overcome such challenges through decent models, algorithms and mechanisms in different stages like storage, processing, pattern recognition, visualization, standardization, and interpretation (Zhong, Lan, Xu, Dai and Huang, 2016). ...
Purpose
The purpose of this paper is to present the state-of-the-art E-commerce logistics in supply chain management by investigating worldwide implementations and corresponding models together with supporting techniques via furniture industry.
Design/methodology/approach
Typical E-commerce logistics companies from North America, Europe, and Asia Pacific are comprehensively investigated so as to get the lessons and insights from these practices.
Findings
Future technologies like Internet of Things, Big Data Analytics, and Cloud Computing would be possibly adopted to enhance the E-commerce logistics in terms of system level, operational level, and decision-making level that may be real time and intelligent in the next decade.
Research limitations/implications
This paper takes the furniture industry for example to illustrate the E-commerce logistics and supply chain management (LSCM). Other industries like electronic appliance industry are not considered.
Practical implications
Opportunities and future perspectives are summarized from practical implementations so that interested parties like E-commerce and logistics companies are able to get some guidance when they are contemplating the business.
Social implications
E-commerce is booming with the development of new business models and will be continuously boosted in the near future. With large number of enterprises carrying out E-commerce, logistics has been largely influenced.
Originality/value
Insights and lessons from this paper are significant for academia and practitioners for considering E-commerce LSCM.
... Big Data is not only with the huge volume or size, but also with the high complexity. Big Data always includes the transaction and interactive data which have large scale of complexity [7]. Thus, the traditional methods are not able to capture, manage, and process these big datasets. ...
Big Data is becoming more and more significant these years since our daily life is facing huge number of data as the millions of electronic devices. Big Data is not only with the huge volume or size, but also with the high complexity. This paper presents a multi-dimensional matrix model for analyzing the large text datasets based on the attributes, which come from the key words from the texts. These key words form an N dimensional space. Thus, the individual information could be presented by an M×N matrix. The multi-dimensional matrix approach has been compared with GA and PSO algorithm so as to test the efficiency and effectiveness of different approaches on analyzing the text datasets. From the experiments, it is observed that the proposed approach outperforms GA and PSO in sufficiency and computational cost. Some key findings are: For high dimensional Big Text Data, at the beginning, PSO has the best sufficiency from 0 to 10. After that, from 10 to 1000, the prosed multi-dimensional matrix approach significantly outperforms GA and PSO. For Connect-4 data samples, the time cost of proposed approach is only 352153.6 unit of time, while GA takes 613601.4 which is more of about half the time cost and PSO takes 469464.1.
... High-dimensional wavelet basis is needed. The proposed dual wavelet is based on the advantages of biorthogonality which is used for constructing linear filter that could meet the high-dimensional applications [9][10][11]. ...
... Complex event processing has been implemented in shop floor to monitor for event-driven manufacturing processes [71]. Besides, with the help of RFID techniques, huge amount of RFID-enabled production data have been generated which call for the participation of "Big Data" techniques to monitor and track the product quality in real time [72]. ...
Recently, “Big Data” has attracted not only researchers’ but also manufacturers’ attention along with the development of information technology. In this paper, the concept, characteristics, and applications of “Big Data” are briefly introduced first. Then, the various data involved in the three main phases of product lifecycle management (PLM) (i.e., beginning of life, middle of life, and end of life) are concluded and analyzed. But what is the relationship between these PLM data and the term “Big Data”? Whether the “Big Data” concept and techniques can be employed in manufacturing to enhance the intelligence and efficiency of design, production, and service process, and what are the potential applications? Therefore, in order to answer these questions, the existing applications of “Big Data” in PLM are summarized, and the potential applications of “Big Data” techniques in PLM are investigated and pointed out.
... Firstly, multi-variables dimensional system is common with physical, chemistry, and other principles to combine all the parameters together, forming a complex variable-oriented system. In order to express the process, different parameters must be integrated to build up a high dimensional description, resulting difficulties in understanding the data [4][5][6][7][8][9]. Secondly, the surveillance data is nonlinear. ...
As the development of science and technology, especially the network and IT techniques, lots of enterprises are facing the floods of data, which brings new challenges such as the storage, processing, and application under the new era. As in the industry area, the data will increase in tremendous level and their relations are more and more complex. That results in great issues when enterprises are contemplating to using the rich information or knowledge hidden from the huge data sets. In order to investigate the trends of the machine status from big data, this paper examines the coal milling machine's abrasion, which is usually monitored. The surveillance is changing along with the time, which is able to reflect the abrasion trends so that the machine maintenance and breakdown could be easily carried out. Given the characteristics of big data in industrial field, this method proposes a multi-scale system with the entropy and energy to reflect the machine status in large scales. Two experiments are carried out to examine the methods. It is founded that, as the time increasing, the changes are becoming larger.
Internet of things (IoT) and Radio Frequency Identification (RFID) technologies are gradually adopted in manufacturing recently. With the aid of them, numerous data is generated from daily manufacturing operations. Big data analytics is used in locating deficiencies and thus improving the productivity of a manufacturing shopfloor. Many studies have also examined the effect of “Blue Monday” and “post-lunch slump” on worker’s performance. This paper provides a big data approach on analyzing worker’s performance with the data collected from a manufacturing shopfloor. By evaluating the worker’s performance at different time periods, a better decision can be arranged for improving overall productivity.
Mining Big Data is the capability of finding new useful information in complex massive datasets, that may be continuously changing and may have varied data types. Big data is helpful only when it is transformed into knowledge or useful information. Data Intelligence is about transforming data into information, information into knowledge, and knowledge into value. It refers to the intelligent interaction with data in a rich, semantically meaningful ways, where data is used to learn and to obtain knowledge.
However, extracting valuable information from this data by following the classical Knowledge Discovery process reveals new previously unknown challenges, due to Big Data properties. These challenges have received a lot of attention in recent years, and still need more and more contribution and research. A large number of publications have
yielded a plethora of proposed methods and algorithms.
In this paper, we provide a comprehensive literature review on Big Data current status. We present the Data Intelligence framework in the context of Big Data from data acquisition until insight extraction, we highlight its main issues, and identify its progress in both technological and algorithmic perspectives. We summarize and analyse relevant research papers in the field, collected from different scientific databases. This investigation will help researchers to understand the current status of Data Intelligence, discover new research opportunities, and gain information about this field.
The emergence of Internet of Things (IoT) and new manufacturing paradigms have brought greater complexity of massive datasets. Radio frequency identification (RFID), as one of the key IoT technologies, has been used to collect real-time production data to support the manufacturing decision-making in smart factories. The adoption of these technologies results in a large amount of data collection. To extract useful information from this data, this paper utilizes a big data approach to figure out useful insights from RFID-enabled data regarding possible bottlenecks or inefficiencies on the shop floor so as to improve the quality management. Time and quality are the main metrics measured in this paper, where the longest process times, part accuracy percentage, and failure rate are determined for each of the workers (UserIDs) and process types (ProcCodes). Key findings and observations are significant to make advanced decisions in the smart factory by making full use of the RFID captured data.
