A novel cloud manufacturing framework with auto-scaling capability for the machining industry

Chao Chun Chen, Yu Chuan Lin, Min Hsiung Hung, Chih Yin Lin, Yen Ju Tsai, Fan Tien Cheng

研究成果: Article

9 引文 (Scopus)

摘要

The globalised machine-tool manufacturing enterprises are eager to develop intelligent machine tools and novel business models to increase their competitiveness. In recent years, cloud manufacturing, encapsulating distributed manufacturing resources into cloud services for supporting all tasks in a product life cycle, has emerged as a promising concept and approach for the machining industry to achieve such a goal and gain profits. However, there has been no systematic approach to the development of cloud manufacturing systems (CMSs) for the machining industry so far. In this paper, we propose a novel cloud manufacturing framework (CMF) with auto-scaling capability (called CMFAS) aimed at providing a systematic and rapid development approach for building CMSs. The proposed CMFAS contains a cloud-based architecture which can transform single-user manufacturing functions (MFs) into cloud services that can be accessed by many users simultaneously. Also, a user-acceptable time-based scaling algorithm is designed so that the CMFAS can automatically perform scale-out or scale-in on the number of virtual machines (VMs) according to the user arrival rate, while confining the average service time for a user to be less than a specified user-acceptable time using a minimum number of VMs. Finally, we develop an Ontology Inference Cloud Service (OICS) for machine tools based on the CMFAS and deploy it on a public cloud platform for conducting integrated tests. Testing results show that the OICS can successfully recommend proper machine tools and cutting tools for machining tasks, and the proposed scaling algorithm outperforms traditional CPU-load-based scaling algorithms in terms of a smaller average service time for a user (i.e. quicker processing time) and a smaller number of created VMs (i.e. less cost of leasing cloud resources). The proposed CMFAS, together with its detailed designs, can serve as a useful reference approach for systematically and rapidly building CMSs for the machining industry.

原文English
頁(從 - 到)786-804
頁數19
期刊International Journal of Computer Integrated Manufacturing
29
發行號7
DOIs
出版狀態Published - 2016 七月 2

指紋

Machining
Machine tools
Industry
Ontology
Cutting tools
Program processors
Life cycle
Profitability
Testing
Processing
Virtual machine
Costs

All Science Journal Classification (ASJC) codes

  • Aerospace Engineering
  • Mechanical Engineering
  • Computer Science Applications
  • Electrical and Electronic Engineering

引用此文

@article{f1ec59044e214d46b15cb535b135d49f,
title = "A novel cloud manufacturing framework with auto-scaling capability for the machining industry",
abstract = "The globalised machine-tool manufacturing enterprises are eager to develop intelligent machine tools and novel business models to increase their competitiveness. In recent years, cloud manufacturing, encapsulating distributed manufacturing resources into cloud services for supporting all tasks in a product life cycle, has emerged as a promising concept and approach for the machining industry to achieve such a goal and gain profits. However, there has been no systematic approach to the development of cloud manufacturing systems (CMSs) for the machining industry so far. In this paper, we propose a novel cloud manufacturing framework (CMF) with auto-scaling capability (called CMFAS) aimed at providing a systematic and rapid development approach for building CMSs. The proposed CMFAS contains a cloud-based architecture which can transform single-user manufacturing functions (MFs) into cloud services that can be accessed by many users simultaneously. Also, a user-acceptable time-based scaling algorithm is designed so that the CMFAS can automatically perform scale-out or scale-in on the number of virtual machines (VMs) according to the user arrival rate, while confining the average service time for a user to be less than a specified user-acceptable time using a minimum number of VMs. Finally, we develop an Ontology Inference Cloud Service (OICS) for machine tools based on the CMFAS and deploy it on a public cloud platform for conducting integrated tests. Testing results show that the OICS can successfully recommend proper machine tools and cutting tools for machining tasks, and the proposed scaling algorithm outperforms traditional CPU-load-based scaling algorithms in terms of a smaller average service time for a user (i.e. quicker processing time) and a smaller number of created VMs (i.e. less cost of leasing cloud resources). The proposed CMFAS, together with its detailed designs, can serve as a useful reference approach for systematically and rapidly building CMSs for the machining industry.",
author = "Chen, {Chao Chun} and Lin, {Yu Chuan} and Hung, {Min Hsiung} and Lin, {Chih Yin} and Tsai, {Yen Ju} and Cheng, {Fan Tien}",
year = "2016",
month = "7",
day = "2",
doi = "10.1080/0951192X.2015.1125766",
language = "English",
volume = "29",
pages = "786--804",
journal = "International Journal of Computer Integrated Manufacturing",
issn = "0951-192X",
publisher = "Taylor and Francis Ltd.",
number = "7",

}

TY - JOUR

T1 - A novel cloud manufacturing framework with auto-scaling capability for the machining industry

AU - Chen, Chao Chun

AU - Lin, Yu Chuan

AU - Hung, Min Hsiung

AU - Lin, Chih Yin

AU - Tsai, Yen Ju

AU - Cheng, Fan Tien

PY - 2016/7/2

Y1 - 2016/7/2

N2 - The globalised machine-tool manufacturing enterprises are eager to develop intelligent machine tools and novel business models to increase their competitiveness. In recent years, cloud manufacturing, encapsulating distributed manufacturing resources into cloud services for supporting all tasks in a product life cycle, has emerged as a promising concept and approach for the machining industry to achieve such a goal and gain profits. However, there has been no systematic approach to the development of cloud manufacturing systems (CMSs) for the machining industry so far. In this paper, we propose a novel cloud manufacturing framework (CMF) with auto-scaling capability (called CMFAS) aimed at providing a systematic and rapid development approach for building CMSs. The proposed CMFAS contains a cloud-based architecture which can transform single-user manufacturing functions (MFs) into cloud services that can be accessed by many users simultaneously. Also, a user-acceptable time-based scaling algorithm is designed so that the CMFAS can automatically perform scale-out or scale-in on the number of virtual machines (VMs) according to the user arrival rate, while confining the average service time for a user to be less than a specified user-acceptable time using a minimum number of VMs. Finally, we develop an Ontology Inference Cloud Service (OICS) for machine tools based on the CMFAS and deploy it on a public cloud platform for conducting integrated tests. Testing results show that the OICS can successfully recommend proper machine tools and cutting tools for machining tasks, and the proposed scaling algorithm outperforms traditional CPU-load-based scaling algorithms in terms of a smaller average service time for a user (i.e. quicker processing time) and a smaller number of created VMs (i.e. less cost of leasing cloud resources). The proposed CMFAS, together with its detailed designs, can serve as a useful reference approach for systematically and rapidly building CMSs for the machining industry.

AB - The globalised machine-tool manufacturing enterprises are eager to develop intelligent machine tools and novel business models to increase their competitiveness. In recent years, cloud manufacturing, encapsulating distributed manufacturing resources into cloud services for supporting all tasks in a product life cycle, has emerged as a promising concept and approach for the machining industry to achieve such a goal and gain profits. However, there has been no systematic approach to the development of cloud manufacturing systems (CMSs) for the machining industry so far. In this paper, we propose a novel cloud manufacturing framework (CMF) with auto-scaling capability (called CMFAS) aimed at providing a systematic and rapid development approach for building CMSs. The proposed CMFAS contains a cloud-based architecture which can transform single-user manufacturing functions (MFs) into cloud services that can be accessed by many users simultaneously. Also, a user-acceptable time-based scaling algorithm is designed so that the CMFAS can automatically perform scale-out or scale-in on the number of virtual machines (VMs) according to the user arrival rate, while confining the average service time for a user to be less than a specified user-acceptable time using a minimum number of VMs. Finally, we develop an Ontology Inference Cloud Service (OICS) for machine tools based on the CMFAS and deploy it on a public cloud platform for conducting integrated tests. Testing results show that the OICS can successfully recommend proper machine tools and cutting tools for machining tasks, and the proposed scaling algorithm outperforms traditional CPU-load-based scaling algorithms in terms of a smaller average service time for a user (i.e. quicker processing time) and a smaller number of created VMs (i.e. less cost of leasing cloud resources). The proposed CMFAS, together with its detailed designs, can serve as a useful reference approach for systematically and rapidly building CMSs for the machining industry.

UR - http://www.scopus.com/inward/record.url?scp=84951268771&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84951268771&partnerID=8YFLogxK

U2 - 10.1080/0951192X.2015.1125766

DO - 10.1080/0951192X.2015.1125766

M3 - Article

AN - SCOPUS:84951268771

VL - 29

SP - 786

EP - 804

JO - International Journal of Computer Integrated Manufacturing

JF - International Journal of Computer Integrated Manufacturing

SN - 0951-192X

IS - 7

ER -