40ARTICLE IN PRESSG ModelSOC 2608 110Applied Soft Computing xxx (2014) xxxxxx
Contents lists available at ScienceDirect
Applied Soft Computing
j ourna l h o mepage: www.elsev ier .com/ locate /asoc
cloud computing platform for ERP applications
hin-Sheng Chena, Wen-Yau Liangb,, Hui-Yu Hsub
Department of Industrial and Systems Engineering, Florida International University, Miami, USADepartment of Information Management, National Changhua University of Education, Taiwan, ROC
r t i c l e i n f o
rticle history:eceived 13 March 2013eceived in revised form 15 October 2014ccepted 11 November 2014vailable online xxx
eywords:eb services compositionRPloud computing
a b s t r a c t
Cloud computing enables many applications of Web services and rekindles the interest of providing ERPservices via the Internet. It has the potentials to reshape the way IT services are consumed. Recent researchindicates that ERP delivered thru SaaS will outperform the traditional IT offers. However, distributinga service compared to distributing a product is more complicated because of the immateriality, theintegration and the one-shot-principle referring to services. This paper defines a CloudERP platform onwhich enterprise customers can select web services and customize a unique ERP system to meet theirspecific needs. The CloudERP aims to provide enterprise users with the flexibility of renting an entire ERPservice through multiple vendors. This paper also addresses the challenge of composing web servicesand proposes a web-based solution for automating the ERP service customization process. The proposedenetic algorithmsough set theory
service composition method builds on the genetic algorithm concept and incorporates with knowledgeof web services extracted from the web service platform with the rough set theory. A system prototypewas built on the Google App Engine platform to verify the proposed composition process. Based onexperimental results from running the prototype, the composition method works effectively and hasgreat potential for supporting a fully functional CloudERP platform.
2014 Elsevier B.V. All rights reserved.
Traditional business applications such as computer aidedesign (CAD), product data management (PDM), Computer aidedanufacturing (CAM), enterprise resources planning (ERP) andanufacturing execution systems (MES) all rely on a central servernd procedural software. These systems are not autonomous orexible enough to support a dynamic business environment .ith the advance of Internet technology and globalization, these
nterprise applications, especially ERP systems have been web-nabled, providing access to information and communications viahe Internet as a part of global business strategy . Along with themerging demand for mobility and on-demand services, the devel-pment of web-based ERP systems becomes an urgent research andevelopment issue .The subscription to web services for ERP applications has two
ssential advantages: ease of integration and reduction in costshrough the hosted application model . Wu et al.  pre-Please cite this article in press as: C.-S. Chen, et al., A cloud computhttp://dx.doi.org/10.1016/j.asoc.2014.11.009
ented a framework for measuring the scalability of service basedpplications in a Cloud Computing environment and propose anssignment strategy to improve the scalability of composite Web
Corresponding author. Tel.: +886 47232105.E-mail address: email@example.com (W.-Y. Liang).
ttp://dx.doi.org/10.1016/j.asoc.2014.11.009568-4946/ 2014 Elsevier B.V. All rights reserved.
61services in terms of services productivity. Recent research indi-cates that ERP delivered thru SaaS will outperform the traditionalIT offers as a consequence of the current economic crisis and willhelps the economies to recover . Although ERP is lagging behindother applications in terms of SaaS based applications there seemsto be a general consensus that ERP in SaaS is gaining momentum.To grab this momentum, the four big players in the ERP systemsmarket SAP, Oracle, Sage and Microsoft are positioning their ERPoffers in SaaS model . However, distributing a service com-pared to distributing a product is more complicated because of theimmateriality, the integration and the one-shot-principle referringto services . Also, the process of analyzing and selecting servicesin the Web services composition process is more complex than theone of analyzing and selecting parts for a product design . Itis further complicated by the customers request in terms of thescope of application. One specific need is the development of effi-cient composition methods which evaluate and optimally integratethese possibly heterogeneous services on the Web, especially in theERP application domain, in response to an enterprise customersrequest.
Therefore, this paper proposes a CloudERP platform on whiching platform for ERP applications, Appl. Soft Comput. J. (2014),
enterprise customers can select web services and customize aunique ERP system to meet their specific needs. The CloudERP aimsto provide enterprise users with the flexibility of renting an entireERP service through multiple vendors. This paper also addresses the
ING ModelA2 ft Com
184ARTICLESOC 2608 110 C.-S. Chen et al. / Applied So
hallenge of composing web services and proposes a web-basedolution for automating the ERP service customization process.his study proposes a method that makes use of the genetic algo-ithm (GA) concept and the rough set theory to solve the Webervices composition problem. The genetic algorithm incorporatesith rough set theory to solve the web services composition prob-
em has been discussed and applied [5,36,37]. However, these allocus on how to use rough set theory to extract rules and ignorehe feature of the application domain. The novelty of the proposedethod lies in the application domain (Cloud ERP).This remainder of this paper is organized as follows: Section 2
eviews the concepts of Web-based ERP and Web services composi-ion. In Section 3, a novel ERP platform called CloudERP is proposed.n Section 4, the proposed composition method for Web ERP ser-ices is presented. In Section 5, a system prototype is presentedlong with experimental data analysis and then followed by Sec-ion 6 which provides concluding remarks and summary of futureesearch directions.
. Literature review
.1. Web-based ERP
ERP systems are one of the most adopted information tech-ology (IT) solutions in organizations . Because of their scalend substantial resources consumption, it is not surprising thatRP systems have been a center of focus by both researchers andractitioners . The key competitive edge for every enterprisen the 21st century is in its ability to prescribe, standardize, anddapt its business activities and collaborations with customers,uppliers, partners and competitors . Most ERP vendors todayecognize this interoperability issue as significant and have built upnternet-enabled supply chain/logistics modules to facilitate inte-ration with the back-end systems of supply chain partners relyingn a diverse set of legacy databases, IT infrastructure and applica-ions . For example, Gollakota  reported a company whichreated kiosks with Internet and computer access and operated aeb portal serving the needs of the farming industry. The portalrovided information relating to farming techniques,