Title: Model based systems engineering—A text mining based structured comprehensive overview
An observed increase in systems scale and complexity has led to a significant momentum in exploring, identifying, and adopting model based systems engineering (MBSE) tools and techniques amongst research communities and industry practitioners. Several attempts to transform systems design and engineering practices through the use of MBSE in academia and industry has led to a considerable increase in the number of articles published containing the keyword “MBSE.” This growth serves as the motivation in this paper to explore the MBSE landscape with the help of text mining techniques to identify the most often used key terms, tools, and languages, in the context of research in MBSE and the thematic aspects defining the use of MBSE by researchers and practitioners. The objective of this paper is to provide a structured comprehensive overview of research contributions across the MBSE landscape by employing text mining techniques for: (a) identifying the concepts and methodologies inferred upon in relation to MBSE, and (b) classifying the literature published to identify commonalities across academic researchers and practitioners using MBSE tools and methods. For this purpose, the abstracts of 2380 relevant articles published in the period of the last two decades from five different databases are mined. It is found that the terms “SysML,” “Cyber Physical Systems,” and “Production” are the most used terms among researchers across the MBSE landscape with SysML being the most widely used modeling language. Further, six major thematic topics are identified that classify articles from over the last two decades with an increasing interest observed in the use of MBSE to support manufacturing and production engineering activities, especially in the cyber physical systems domain. The contributions of this paper provide a leeway on using text mining techniques to understand the research directions that are currently of interest in the field of MBSE and thereby identify potential future research directions. more »« less
Akundi, Aditya; Lopez, Viviana; Tseng, Tzu-Liang Bill(
, 2021 IEEE International Systems Conference (SysCon))
null
(Ed.)
Manufacturing and production systems have become increasingly complex in the past decade to meet the competitive demand in a growing industry. As these systems grow in complexity and flexibility, there is a need for efficient management and analysis of these systems. Model-based systems engineering (MBSE) addresses the complexity inherent with systems development with a model-centric approach that supported tailored modeling languages, methods and tools. This paper identifies the thematic evolution and trends and relationships found in the use and application of MBSE specifically in the manufacturing and production engineering domain. A collection of 471 published article from Institute of Electrical and Electronics Engineers (IEEE) and Science Direct over the past decade were used for the analysis using text mining techniques. Due to the limitation on the access to full text information of all the articles identified, only abstracts were considered for analysis. This effort helps the researchers across the domain to explore the reason behind and understand the change of the thematic perspectives of MBSE application over the last decade. In addition, the finding of the growing interest in addressing the aspects of complexity and systems requirements, and on the aspects of the use of MBSE for identifying and addressing the challenges related to Cyber Physical Systems help in paving a path for future research.
Researchers, evaluators and designers from an array of academic disciplines and industry sectors are turning to participatory approaches as they seek to understand and address complex social problems. We refer to participatory approaches that collaboratively engage/ partner with stakeholders in knowledge creation/problem solving for action/social change outcomes as collaborative change research, evaluation and design (CCRED). We further frame CCRED practitioners by their desire to move beyond knowledge creation for its own sake to implementation of new knowledge as a tool for social change. In March and May of 2018, we conducted a literature search of multiple discipline-specific databases seeking collaborative, change-oriented scholarly publications. The search was limited to include peerreviewed journal articles, with English language abstracts available, published in the last five years. The search resulted in 526 citations, 236 of which met inclusion criteria. Though the search was limited to English abstracts, all major geographic regions (North America, Europe, Latin America/Caribbean, APAC, Africa and the Middle East) were represented within the results, although many articles did not state a specific region. Of those identified, most studies were located in North America, with the Middle East having only one identified study. We followed a qualitative thematic synthesis process to examine the abstracts of peer-reviewed articles to identify practices that transcend individual disciplines, sectors and contexts to achieve collaborative change. We surveyed the terminology used to describe CCRED, setting, content/topic of study, type of collaboration, and related benefits/outcomes in order to discern the words used to designate collaboration, the frameworks, tools and methods employed, and the presence of action, evaluation or outcomes. Forty-three percent of the reviewed articles fell broadly within the social sciences, followed by 26 percent in education and 25 percent in health/medicine. In terms of participants and/ or collaborators in the articles reviewed, the vast majority of the 236 articles (86%) described participants, that is, those who the research was about or from whom data was collected. In contrast to participants, partners/collaborators (n=32; 14%) were individuals or groups who participated in the design or implementation of the collaborative change effort described. In terms of the goal for collaboration and/or for doing the work, the most frequently used terminology related to some aspect of engagement and empowerment. Common descriptors for the work itself were ‘social change’ (n=74; 31%), ‘action’ (n=33; 14%), ‘collaborative or participatory research/practice’ (n=13; 6%), ‘transformation’ (n=13; 6%) and ‘community engagement’ (n=10; 4%). Of the 236 articles that mentioned a specific framework or approach, the three most common were some variation of Participatory Action Research (n=30; 50%), Action Research (n=40; 16.9%) or Community-Based Participatory Research (n=17; 7.2%). Approximately a third of the 236 articles did not mention a specific method or tool in the abstract. The most commonly cited method/tool (n=30; 12.7%) was some variation of an arts-based method followed by interviews (n=18; 7.6%), case study (n=16; 6.7%), or an ethnographic-related method (n=14; 5.9%). While some articles implied action or change, only 14 of the 236 articles (6%) stated a specific action or outcome. Most often, the changes described were: the creation or modification of a model, method, process, framework or protocol (n=9; 4%), quality improvement, policy change and social change (n=8; 3%), or modifications to education/training methods and materials (n=5; 2%). The infrequent use of collaboration as a descriptor of partner engagement, coupled with few reported findings of measurable change, raises questions about the nature of CCRED. It appears that conducting CCRED is as complex an undertaking as the problems that the work is attempting to address.
Zhang, Hao; Li, Zhuohan; Zheng, Lianmin; Stoica, Ion(
, KDD '21: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining)
In recent years, the pace of innovations in the fields of machine learning (ML) has accelerated, researchers in SysML have created algorithms and systems that parallelize ML training over multiple devices or computational nodes. As ML models become more structurally complex, many systems have struggled to provide all-round performance on a variety of models. Particularly, ML scale-up is usually underestimated in terms of the amount of knowledge and time required to map from an appropriate distribution strategy to the model. Applying parallel training systems to complex models adds nontrivial development overheads in addition to model prototyping, and often results in lower-than-expected performance. This tutorial identifies research and practical pain points in parallel ML training, and discusses latest development of algorithms and systems on addressing these challenges in both usability and performance. In particular, this tutorial presents a new perspective of unifying seemingly different distributed ML training strategies. Based on it, introduces new techniques and system architectures to simplify and automate ML parallelization. This tutorial is built upon the authors' years' of research and industry experience, comprehensive literature survey, and several latest tutorials and papers published by the authors and peer researchers.
The tutorial consists of four parts. The first part will present a landscape of distributed ML training techniques and systems, and highlight the major difficulties faced by real users when writing distributed ML code with big model or big data. The second part dives deep to explain the mainstream training strategies, guided with real use case. By developing a new and unified formulation to represent the seemingly different data- and model- parallel strategies, we describe a set of techniques and algorithms to achieve ML auto-parallelization, and compiler system architectures for auto-generating and exercising parallelization strategies based on models and clusters. The third part of this tutorial exposes a hidden layer of practical pain points in distributed ML training: hyper-parameter tuning and resource allocation, and introduces techniques to improve these aspects. The fourth part is designed as a hands-on coding session, in which we will walk through the audiences on writing distributed training programs in Python, using the various distributed ML tools and interfaces provided by the Ray ecosystem.
Chenxi Yuan, Guoyan Li(
, Journal of intelligent manufacturing)
In recent years, driven by Industry 4.0 wave, academic research has focused on the science, engineering, and enabling technologies for intelligent and cyber manufacturing. Using a network science and data mining-based Keyword Co-occurrence Network (KCN) methodology, this work analyzes the trends in data science topics in the manufacturing literature over the past two decades to inform the researchers, educators, industry leaders of knowledge trends in intelligent manufacturing. It studies the evolution of research topics and methods in data science, Internet of Things (IoT), cloud computing, and cyber manufacturing. The KCN methodology is applied to systematically analyze the keywords collected from 84,041 papers published in top-tier manufacturing journals between 2000 and 2020. It is not practically feasible to review this large body of literature through tradition manual approaches like systematic review and scoping review to discover insights. The results of network modeling and data analysis reveal important knowledge components and structure of the intelligent and cyber manufacturing literature, implicit the research interests switch and provide the insights for industry development. This paper maps the high frequency keywords in the recent literature to nine pillars of Industry 4.0 to help manufacturing community identify research and education directions for emerging technologies in intelligent manufacturing.
In recent years, driven by Industry 4.0 wave, academic research has focused on the science, engineering, and enabling technologies for intelligent and cyber manufacturing. Using a network science and data mining-based Keyword Co-occurrence Network (KCN) methodology, this work analyzes the trends in data science topics in the manufacturing literature over the past two decades to inform the researchers, educators, industry leaders of knowledge trends in intelligent manufacturing. It studies the evolution of research topics and methods in data science, Internet of Things (IoT), cloud computing, and cyber manufacturing. The KCN methodology is applied to systematically analyze the keywords collected from 84,041 papers published in top-tier manufacturing journals between 2000 and 2020. It is not practically feasible to review this large body of literature through tradition manual approaches like systematic review and scoping review to discover insights. The results of network modeling and data analysis reveal important knowledge components and structure of the intelligent and cyber manufacturing literature, implicit the research interests switch and provide the insights for industry development. This paper maps the high frequency keywords in the recent literature to nine pillars of Industry 4.0 to help manufacturing community identify research and education directions for emerging technologies in intelligent manufacturing.
Akundi, Aditya, and Mondragon, Oscar. Model based systems engineering—A text mining based structured comprehensive overview. Retrieved from https://par.nsf.gov/biblio/10294267. Systems Engineering . Web. doi:10.1002/sys.21601.
Akundi, Aditya, & Mondragon, Oscar. Model based systems engineering—A text mining based structured comprehensive overview. Systems Engineering, (). Retrieved from https://par.nsf.gov/biblio/10294267. https://doi.org/10.1002/sys.21601
@article{osti_10294267,
place = {Country unknown/Code not available},
title = {Model based systems engineering—A text mining based structured comprehensive overview},
url = {https://par.nsf.gov/biblio/10294267},
DOI = {10.1002/sys.21601},
abstractNote = {An observed increase in systems scale and complexity has led to a significant momentum in exploring, identifying, and adopting model based systems engineering (MBSE) tools and techniques amongst research communities and industry practitioners. Several attempts to transform systems design and engineering practices through the use of MBSE in academia and industry has led to a considerable increase in the number of articles published containing the keyword “MBSE.” This growth serves as the motivation in this paper to explore the MBSE landscape with the help of text mining techniques to identify the most often used key terms, tools, and languages, in the context of research in MBSE and the thematic aspects defining the use of MBSE by researchers and practitioners. The objective of this paper is to provide a structured comprehensive overview of research contributions across the MBSE landscape by employing text mining techniques for: (a) identifying the concepts and methodologies inferred upon in relation to MBSE, and (b) classifying the literature published to identify commonalities across academic researchers and practitioners using MBSE tools and methods. For this purpose, the abstracts of 2380 relevant articles published in the period of the last two decades from five different databases are mined. It is found that the terms “SysML,” “Cyber Physical Systems,” and “Production” are the most used terms among researchers across the MBSE landscape with SysML being the most widely used modeling language. Further, six major thematic topics are identified that classify articles from over the last two decades with an increasing interest observed in the use of MBSE to support manufacturing and production engineering activities, especially in the cyber physical systems domain. The contributions of this paper provide a leeway on using text mining techniques to understand the research directions that are currently of interest in the field of MBSE and thereby identify potential future research directions.},
journal = {Systems Engineering},
author = {Akundi, Aditya and Mondragon, Oscar},
editor = {null}
}
Warning: Leaving National Science Foundation Website
You are now leaving the National Science Foundation website to go to a non-government website.
Website:
NSF takes no responsibility for and exercises no control over the views expressed or the accuracy of
the information contained on this site. Also be aware that NSF's privacy policy does not apply to this site.
