Zero downtime manufacturing: Difference between revisions

Browse history interactively

← Previous edit

Content deleted Content added

VisualWikitext

Inline

Latest revision as of 12:34, 18 April 2020

Redirect to:

Intelligent maintenance system

@@ Line 1: / Line 1: @@
-{{mergeto|Intelligent maintenance system|discuss=Talk:Intelligent maintenance system#Merger proposal|date=April 2020}}
+#Redirect [[Intelligent maintenance system]]
-{{Multiple issues|{{Copy edit|date=April 2020}}
-{{Underlinked|date=April 2020}}
-{{Orphan|date=April 2020}}}}
-In the early 2000s, the [[manufacturing]] industry was experiencing turbulent and noticeable changes. As a result, manufacturers were pressured to improve their speed and efficiency in terms of product development, operations, lead times, resource utilization, and quality control, which were coupled with complex supply chain networks, [[manufacturing execution system]]<nowiki/>s (MES) and [[enterprise resource planning]] (ERP) systems. A new concept was evolved to address these issues and enable organizations to dynamically, collaboratively, and cost-effectively integrate, optimize, and configure their manufacturing units and supply networks. This concept was called E-manufacturing<ref>{{cite journal|last1=Lee|first1=Jay|last2=Wang|first2=Haixia|date=2008|title=New Technologies for Maintenance|journal=Complex System Maintenance Handbook|series=Springer Series in Reliability Engineering|language=en|publisher=Springer|pages=49–78|doi=10.1007/978-1-84800-011-7_3|isbn=978-1-84800-010-0}}</ref><ref>{{cite journal|last1=Zhang|first1=David Zhengwen|last2=Anosike|first2=Anthony Ikechukwu|last3=Lim|first3=Ming Kim|last4=Akanle|first4=Oluwaremilekun Mowanuola|date=2006|title=An agent-based approach for e-manufacturing and supply chain integration|journal=Computers & Industrial Engineering|volume=51|issue=2|pages=343–360|doi=10.1016/j.cie.2006.02.012}}</ref>, which was first envisioned by AMR Research Inc. in 2000<ref>{{cite journal|title=AMR Advanced Market Research - Global Data Collection CATI & Online|url=https://amr-research.com/en/|journal=AMR}}</ref> and conceptualized by Lee et al.<ref>{{cite journal|last1=Lee|first1=Jay|date=2003|title=E-manufacturing—fundamental, tools, and transformation|journal=Robotics and Computer-Integrated Manufacturing|volume=19|issue=6|pages=501–507|doi=10.1016/S0736-5845(03)00060-7}}</ref> in 2002. The concept was defined as: “E-manufacturing is a systematic methodology that enables the manufacturing operations to successfully integrate with the functional objectives of an enterprise through the use of the Internet, tether-free (i.e., wireless, web, etc.) and predictive technologies.” E-Manufacturing paradigm was enabled by the ubiquitous internet and telecommunication facilities (i.e. wireless, web, etc.), which helped manufacturing operators to achieve near-zero downtime (nZDM) performance and synchronize with the business systems.
-Maintenance operations are essential to minimize unplanned downtime, to assure product quality, to reduce customer dissatisfaction, and to maintain a competitive edge in the market. However, manufacturers struggle to find balance maintenance strategies without compromising the system's reliability or productivity. Intelligent Maintenance Systems are designed to provide support tools for decision making to optimize maintenance operations. Intelligent prognostic and health management tools are necessary to identify practical, reliable, and cost-saving maintenance strategies to ensure nZDM in production, consistently. Dr. Jay Lee {{fact|date=April 2020}} first introduced the nZDM concept in 2000 while working for NSF I/UCRC<ref>{{Cite web|url=https://www.nsf.gov/eng/iip/iucrc/home.jsp|title=NSF IUCRC {{!}} Home|website=www.nsf.gov|access-date=2020}}</ref> in the area of Intelligent Maintenance Systems (IMS). The key focus was to measure the degradation of machines and processes to predict and prevent potential issues. Analytical techniques based on Big Data, along with a systematic toolbox approach, were first proposed by the IMS Center in 2001.<ref>{{cite journal|last1=Lee|first1=Jay|date=2003|title=E-manufacturing—fundamental, tools, and transformation|journal=Robotics and Computer-Integrated Manufacturing|volume=19|issue=6|pages=501–507|doi=10.1016/S0736-5845(03)00060-7}}</ref><ref>{{cite web|url=http://imscenter.net/|title=Intelligent maintenance system|language=en-us}}</ref>
-==Main Elements==
-Near-zero downtime manufacturing is based on two main technologies, namely (1) Accurate Diagnostics and (2) Real-time Prognostics{{citation needed|date=April 2020}}. Both of these technologies are described as follows:
-===Prognostics===
-Prognostics aims to make accurate predictions of the remaining useful life (RUL) of machinery before a failure occurs.<ref>{{cite journal|last1=Heng|first1=Aiwina|last2=Zhang|first2=Sheng|last3=Tan|first3=Andy C.C.|last4=Mathew|first4=Joseph|date=2009|title=Rotating machinery prognostics: State of the art, challenges and opportunities|journal=Mechanical Systems and Signal Processing|volume=23|issue=3|pages=724–739|bibcode=2009MSSP...23..724H|doi=10.1016/j.ymssp.2008.06.009}}</ref> Current prognostic approaches can be categorized into three classes, namely model-based, data-driven and hybrid prognostics approaches.<ref>{{cite journal|last1=Si|first1=Xiao-Sheng|last2=Wang|first2=Wenbin|last3=Hu|first3=Chang-Hua|last4=Zhou|first4=Dong-Hua|date=2011|title=Remaining useful life estimation – A review on the statistical data driven approaches|journal=European Journal of Operational Research|volume=213|issue=1|pages=1–14|doi=10.1016/j.ejor.2010.11.018}}</ref> Rotating machinery is of critical concern in the manufacturing industry. Hence, the majority of the studies in literature concentrated on prognostics and health management (PHM) applications, including but not limited to, standard rotary machinery components, such as bearings, gears, shafts, and motors.<ref>{{cite journal|last1=Lee|first1=Jay|last2=Wu|first2=Fangji|last3=Zhao|first3=Wenyu|last4=Ghaffari|first4=Masoud|last5=Liao|first5=Linxia|last6=Siegel|first6=David|date=2014|title=Prognostics and health management design for rotary machinery systems—Reviews, methodology and applications|journal=Mechanical Systems and Signal Processing|volume=42|issue=1–2|pages=314–334|bibcode=2014MSSP...42..314L|doi=10.1016/j.ymssp.2013.06.004}}</ref> A significant contribution has been made in are using vibration signals for prognostics. However, very few studies have been carried out on stator current monitoring or acoustic signals. A preliminary study by Tandon et al. <ref>{{cite journal|last1=Tandon|first1=N.|last2=Yadava|first2=G.S.|last3=Ramakrishna|first3=K.M.|date=2007|title=A comparison of some condition monitoring techniques for the detection of defect in induction motor ball bearings|journal=Mechanical Systems and Signal Processing|volume=21|issue=1|pages=244–256|bibcode=2007MSSP...21..244T|doi=10.1016/j.ymssp.2005.08.005}}</ref> suggested that current signatures are inexpensive and easy to measure, but acoustic signals reveal fault characteristics early.
-===Diagnostics===
-The fault diagnosis is performed to inspect whether a piece of machinery is healthy or has a fault. In industrial applications, diagnosis is performed to detect the failure modes within a system or sub-system. In this process, time and frequency domain features are extracted from sensors and the data is analyzed using a model, which is designed to pass the information to the failure modes of the system{{citation needed|date=April 2020}}. Using diverse analytic models and a systematic diagnosis approach provides a productive environment for the diagnosis of components during real-time monitoring{{citation needed|date=April 2020}}. Bearing, shaft, electric motors, and gearbox are the most frequently studied components in manufacturing systems for accurate and reliable diagnostics{{citation needed|date=April 2020}}.
-==Enabling Technologies==
-Recently emerging technologies such as the industrial internet of things (IoT), big data analytics, cloud computing, fog computing, cyber-physical systems, digital twin, and industrial artificial intelligence (AI) technologies will enable operation of industries in a flexible, efficient, and with near-zero downtime. These technologies aim to aid the intelligent manufacturing systems based on process monitoring hardware and software, signal processing techniques, sensor fusions tools, and diagnosis and prognosis technologies.
-=== Internet of things (IoT)===
-The basic concept is the pervasive presence of things around us, such as sensors, actuators, Radio-Frequency Identification (RFID) tags, etc., which can interact and cooperate to achieve common goals.<ref>{{cite journal|last1=Jbara|first1=Yosef Hasan|date=2020|title=Data Reduction in MMBD Computing|journal=Multimedia Big Data Computing for IoT Applications: Concepts, Paradigms and Solutions|series=Intelligent Systems Reference Library|language=en|publisher=Springer|volume=163|pages=217–245|doi=10.1007/978-981-13-8759-3_8|isbn=978-981-13-8758-6}}</ref>
-===Big Data Analytics===
-Analyzing the vast amount of data collected from the manufacturing shop floor plays a vital role in managing intelligent maintenance activities and reducing unplanned downtime{{fact|date=April 2020}}. Usually, analytic tools are used to convert raw data into meaningful information to increase transparency in manufacturing systems.
-===Cloud Computing===
-Cloud computing helps in the realization of cloud-based maintenance systems, where maintenance activities such as prognosis and diagnostic are offered as on-demand to different clients in the manufacturing industry{{citation needed|date=April 2020}}.
-===Cyber-Physical Systems===
-Making informed decisions on maintenance activities require advanced connectivity, integration of physical systems, and cyber computational resources. This interconnectivity is realized through the 5-level cyber-physical system known as 5C architecture<ref>{{Cite journal|last=Jiang|first=Jehn-Ruey|date=2017|title=An improved Cyber-Physical Systems architecture for Industry 4.0 smart factories|url=http://dx.doi.org/10.1109/icasi.2017.7988589|journal=2017 International Conference on Applied System Innovation (ICASI)|publisher=IEEE|doi=10.1109/icasi.2017.7988589|isbn=978-1-5090-4897-7}}</ref>.
-===Industrial AI===
-Industrial AI is a systematic discipline to enable engineers to develop and deploy algorithms with consistent success. The traditional definition of industrial AI usually refers to the application of [[artificial intelligence]] applied to industry.<ref name=":0">{{cite web|url=https://www.forbes.com/sites/mariyayao/2017/04/14/unique-challenges-of-industrial-artificial-intelligence-general-electric/#59e377551305|title=4 Unique Challenges Of Industrial Artificial Intelligence|last1=Yao|first1=Mariya|website=Forbes|accessdate=2017}}</ref> Successful implementation of industrial AI may improve decision making and provide more in-depth insight to business users for various purposes, including, but not limited to, reducing asset downtime, improving manufacturing efficiency, automating production, predicting demand, optimizing inventory levels or enhancing risk management. A unified architecture for industrial AI includes four primary components, namely (1) Data Technology, (2) Platform Technology, (3) Analytic Technology, and (4) Operation Technology.<ref>{{cite arxiv|eprint=1908.02150|class=cs.CY|first1=Jay|last1=Lee|first2=Jaskaran|last2=Singh|title=Industrial Artificial Intelligence|date=2019|last3=Azamfar|first3=Moslem}}</ref> These core functionalities guarantee the successful implementation of industrial AI leading to new benefits. Unlike general artificial intelligence, which is a frontier research discipline to build computerized systems to perform tasks requiring human intelligence, industrial AI is more concerned with addressing industrial pain-points for customer value creation, productivity improvement, and insight discovery.<ref>{{cite web|url=http://deloitte.wsj.com/cio/2015/07/29/artificial-intelligence-goes-mainstream/|title=Artificial Intelligence Goes Mainstream|last1=Sallomi|first1=Paul|website=The Wall Street Journal|publisher=The Wall Street Journal - CIO Journal - Deloitte|accessdate=2017}}</ref> Although, in a dystopian vision, AI-based machines may take away manual jobs currently done by humans, causing ethical and social issues. However, the industry holds a more positive view of AI and sees it as a big business opportunity and assumes that this transformation of the economy is unstoppable.<ref name=":1">{{cite web|url=https://obamawhitehouse.archives.gov/sites/default/files/whitehouse_files/microsites/ostp/NSTC/preparing_for_the_future_of_ai.pdf|title=Preparing for the Future of Artificial Intelligence|publisher=National Science and Technology Council|accessdate=2017}}</ref>
-== References ==
-<!-- Inline citations added to your article will automatically display here. See en.wikipedia.org/wiki/WP:REFB for instructions on how to add citations. -->
-{{reflist}}
-[[Category:Manufacturing]]