Architectural Challenges on the Integration of e-Commerce and ERP

Systems: A Case Study

abio Santos

and Ricardo Martinho

School of Technology and Management, Polytechnic of Leiria, Leiria, Portugal

Keywords:

e-Commerce, Enterprise Resource Planning (ERP), Software Architecture, Enterprise Application Integration,

Data Integration, Business Process Integration.

Abstract:

Many retail companies had to go online before their Enterprise Resource Planning (ERP)-type systems were

ready to fulﬁll all business requirements. Their overall daily operation still heavily depends on these highly

customized systems often mandatory because of legal obligations, which frequently come without e-commerce

“off the shelf” integration. This paper identiﬁes main challenges derived out of the architectural and integration

requirements from a case study at an e-tailer company that operates via two sales channels: online store and

third-party marketplaces. These challenges led to the deﬁnition of a system architecture and implementation

considerations for this common integration scenario, which was validated through its implementation. Our

proposed approach allows ERP-dependent organizations to start selling online with open-source technologies,

avoiding extra ERP licensing and hidden maintenance costs.

1 INTRODUCTION

The growth of Internet access explains the number

of consumers who order products or services on-

line through e-commerce platforms. In the European

Union (28 member states), the share of households

with Internet access had risen to 90% in 2019 vs. ≈

64% in 2009, while 60% of individuals aged 16 to 74

used some sort of e-commerce (Eurostat, 2020). This

scenario led retail companies to adopt online channels

in order to reach more potential consumers and con-

solidate their position in a global market.

Online marketplaces (e.g. eBay) and their clear

dominance in the e-commerce ecosystem represents

an opportunity for e-tailers to boost Business-to-

Customer (B2C) sales. Well-know marketplaces al-

ready have an established high brand name amongst

consumers, and companies who take advantage of this

channel can beneﬁt from that credibility and customer

trust (Ecommerce Europe, 2019).

Since the late-1990s, ERP systems have become

a requirement for most of the worldwide companies

mainly due to the wide range of features offered by

those systems. ERPs incorporate various enterprise

core activities such as manufacturing, ﬁnance, human

https://orcid.org/0000-0001-5916-7022

https://orcid.org/0000-0003-1157-7510

resources, among others, into a single inter-connected

system shared with multiple departments across all

the organization (Holland and Light, 1999). Besides

the complexity and potential incompatibilities with

business needs and existing business processes, ERPs

are generic solutions that require a certain level of

customization that could be translated into an incre-

ment of the overall enterprise’s productivity and long-

term performance (Davenport, 1998).

Despite the beneﬁts of ERP systems, commer-

cial ERP software products are often mandatory for

merchants because of legal requirements related with

certiﬁed software (take, for instance, the Portuguese

case regarding invoicing software (Portuguese Gov-

ernment, 2019)) and the lack of certiﬁed open-source

solutions. Nevertheless, some of these systems do not

offer a complete turnkey solution for a wide range of

businesses such as the absence of e-commerce mod-

ules for retailers who want to sell goods or services

through the Internet. For this reason, companies are

forced to use additional e-commerce software prod-

ucts to meet all business requirements, which leads

to an increase in manual activities (and the hiring of

more human resources) to address the absence of au-

tomated integration processes between ERP and e-

commerce systems.

The coexistence of multiple information systems

and applications to support daily enterprise opera-

Santos, F. and Martinho, R.

Architectural Challenges on the Integration of e-Commerce and ERP Systems: A Case Study.

DOI: 10.5220/0010494903130319

In Proceedings of the 23rd International Conference on Enterprise Information Systems (ICEIS 2021) - Volume 1, pages 313-319

ISBN: 978-989-758-509-8

313

tions and ﬁt all business requirements presents mul-

tiple challenges. Companies should embrace Enter-

prise Application Integration (EAI) as an opportunity

to obtain competitive advantage by optimizing their

core business processes.

This article attempts to determine the software

architectural and integration challenges related with

the integration needed between ERP and e-commerce

systems in e-tailer companies. The speciﬁc research

questions are: 1) what are the main challenges on the

integration of e-commerce and ERP systems; and 2)

how these challenges can be addressed.

To support the research, a case study has been con-

ducted in a Portuguese retail company that operates

an online B2C business via two channels: own online

store and third-party marketplaces.

The remainder of this paper is structured as fol-

lows. Next section mentions related work on the

particular architectural and integration challenges re-

garding ERP and e-commerce systems. Section 3

presents the case study, functional and non-functional

requirements and main challenges. Section 4 justiﬁes

the e-commerce platform selected. Section 5 deﬁnes

the system architecture and implementation details.

Section 6 concludes the paper.

2 RELATED WORK

While an online store can operate without a real-time

connection to accounting systems, ERPs or others, the

cost of manually managing sales and product catalogs

in these multiple systems is often higher than initially

expected. Here, EAI enables stocks, orders, deliver-

ies, returns and invoicing (among other related sales

data) to be automatically exchanged between inde-

pendent systems (Farzaneh, 2014). Existing literature

on the integration of e-commerce and ERP systems

point to potential issues that could be encountered.

For instance, in (Kaya and Aydın, 2019) it is con-

cluded that the automated and direct data transfer

from an e-commerce platform to an ERP database

could cause accounting problems because ERP sys-

tems are often in-use by the whole company every-

day. An intermediate database layer could be created

to mitigate this issue, so data is temporarily stored

prior to importation into the ERP.

In (Nikitovi

c and Mahmutovi

c, 2019), extensive

ERP customization is also referenced as a generator

of hidden costs, since some requirements elicitation

is only concluded during the implementation process.

Despite integration problems, e-commerce solu-

tions face also relevant challenges that should be

taken into account. Noted in (Thaw et al., 2009) is in-

formation security, which plays an important role be-

cause of consumer’s trust in online transactions (cru-

cial for the continuous growth and development of

any e-commerce activity). The usage of technology

enablers such as ﬁrewalls and security protocols (e.g.

3D Secure) could improve the overall security of the

system (Nili et al., 2019).

Maintaining an efﬁcient product inventory is seen

as another challenge. It is important to prevent and ac-

count for stock outs to avoid losing sales and improve

customer experience (Patil and Divekar, 2014).

e-Commerce platforms and ERP systems are het-

erogeneous and complex since they are often de-

veloped independently in potential different tech-

nological ecosystems. Consequently, the industry

adopted technologies and standards which offer cross-

platform support such as eXtensible Markup Lan-

guage (XML) and Simple Object Access Protocol

(SOAP) (Wang and Shi, 2017).

Moreover, the integration of ERP and e-commerce

can be achieved by using speciﬁc agents to ﬁlter, orga-

nize and transfer information efﬁciently through web

services (Kong et al., 2007).

Regardless of the integration dimension/strategy

(presentation, functional or data, as referenced in

(Ruh et al., 2002)), the development of inter-

application middleware is a common architectural ap-

proach to enable connections between legacy sys-

tems, databases, servers, web applications, among

others (Lee et al., 2003). Figure 1 illustrates the com-

parison between the traditional vs. middleware ap-

proaches.

AlternativeTraditional approach

Legacy system Legacy system

Web applicationWeb application

Legacy systems Databases

ServersWeb applications

Inter-application middleware

Figure 1: Comparison of two software architectural ap-

proaches for EAI.

The traditional approach promotes the implementa-

tion of custom changes in each existing node of the

software application ecosystem to allow the commu-

nication between them. It will eventually concen-

trate the same business logic on distinct applications,

which leads to complex maintenance tasks and repet-

itive validation procedures for integration purposes.

Adopting inter-application middleware as a soft-

ware architectural solution makes possible to miti-

gate these problems, since it is focused on mapping

processes, workﬂows and data between diverse ap-

ICEIS 2021 - 23rd International Conference on Enterprise Information Systems

314

plications, systems and databases. This was, indeed,

the overall architectural solution adopted for the case

study described in the next section.

3 CASE STUDY

The case study of this research is focused on a Por-

tuguese retail company that has established to sell

biking gear and cycling clothing through the Internet.

The ﬁrm intends to execute B2C operations via two

sales channels:

• Online Store: company’s e-commerce platform

consisting of a web application with product cata-

log, shopping cart and automatic payment system;

• Marketplaces: third-party (marketplace) plat-

forms where the company’s products can be listed,

advertised and sold online.

ERP PRIMAVERA Professional was the system al-

ready in place used by the company to support stock

management, invoicing and human resources, among

others. It runs on an on-premises environment com-

posed by Microsoft Windows Server and Microsoft

SQL Server (PRIMAVERA, 2020).

Therefore, any e-commerce solution would have

to be fully integrated with this system, due to its

supporting role on the company’s main business pro-

cesses, and since it did not offer any e-commerce

module to create an online store and to easily inte-

grate with existing online marketplaces.

3.1 Architectural and Integration

Software Requirements

Before the design of the overall architecture and im-

plementation phase, the requirements were elicited

through multiple in-person meetings with the board of

the retail company. We resume below the main func-

tional and non-functional requirements derived out of

these meetings.

3.1.1 Functional Requirements (FR)

Users such as administrative and warehouse workers

should be able to perform the following actions on the

architectural (integrated) solution:

FR1. View online orders by status (billing, ship-

ping, return, etc.);

FR2. View orders history;

FR3. View documents issued by the ERP;

FR4. Create accounting documents (e.g. invoices,

credit notes);

FR5. Control visibility of articles in the online

store and marketplaces;

FR6. Manage attributes and characteristics of ar-

ticles (e.g. bar codes, title, descriptions, im-

ages);

FR7. View blocked orders and causes;

FR8. View sync failures and causes;

FR9. Force sync of product makes/brands, at-

tributes and catalogs.

In turn, the solution should also account for the fol-

lowing automated features:

FR10. Sync the makes/brands lists;

FR11. Sync attributes (e.g. clothing size charts);

FR12. Sync products with/without variations;

FR13. Sync images of products;

FR14. Sync stock as soon as an order is placed;

FR15. Adopt prevention error mechanisms (e.g.

rollback transactions);

FR16. Reserve stock in ERP after an order is

placed;

FR17. Create accounting documents via ERP.

3.1.2 Non-Functional Requirements (NFR)

The ﬁnal integrated solution should also comply to

the following non-functional requirements:

NFR1. PRIMAVERA ERP integration;

NFR2. Marketplaces integration with Amazon

and Cdiscount;

NFR3. Multi-language support;

NFR4. Search engine optimization;

NFR5. Use only open-source components, li-

braries and frameworks for potentially

needed software development.

3.2 Challenges

The identiﬁcation of the requirements above and re-

lated work determined the following challenges:

C1. Choose a viable e-commerce platform;

C2. Integrate proprietary code with open-source

components;

C3. Automatically synchronize data of heteroge-

neous systems;

C4. Manage stock efﬁciently;

Architectural Challenges on the Integration of e-Commerce and ERP Systems: A Case Study

315

C5. Minimize manual steps required to integrate

information (e.g. customers, orders, product

catalog, etc.);

C6. Avoid direct overuse of the ERP for core daily

operations (e.g. invoicing);

C7. Focus on information security practices;

C8. Optimize development workload and mini-

mize future maintenance effort;

C9. Handle an increase in transactions.

4 e-COMMERCE PLATFORM

DEFINITION

Considering the development effort of building an

e-commerce platform with all the usual base fea-

tures (e.g., catalogs, payments, checkout, customers,

among others), a comparative study was conducted

in order to evaluate existing open-source solutions

(NFR5) and choose a viable platform (C1). These so-

lutions were retrieved from their market share of the

top 1 million sites (BuiltWith, 2021). The selected

platform has direct impact on the process to deﬁne

the overall architecture as it depends on features of-

fered by the e-commerce solution to support systems

integration.

The table 1, presents a weighted scoring analy-

sis of various open-source platforms, based on crite-

ria to meet the case study requirements (e.g. multi-

language support (NFR3) and search engine opti-

mization (NFR4)). Weights were distributed accord-

ing to the relevance of each topic to the case study.

The setup and conﬁguration are not only related

with the initial e-commerce platform setup (e.g. con-

necting to a database) but also how simple and easy an

operator can change any conﬁguration via the back-

ofﬁce (e.g. setup payment providers).

Any existing integration interface (Application

Programming Interfaces (APIs)) (C2) together with a

shallow learning curve of the platform and an exten-

sive marketplace of extensions and themes, are rele-

vant to decrease the development time-frame (C8).

Future maintainability efforts could be minimized

through an active community and support, well-

documented project but also by a clear development

roadmap to anticipate features or bug-ﬁxes (C8).

None of the solutions has native integration with

the PRIMAVERA ERP (NFR1) or with online mar-

ketplaces (NFR2). However, the integration with mar-

ketplaces can be easily achieved via third-party exten-

sions if no customization is required.

All evaluated solutions were developed with the

PHP and JavaScript programming languages and rela-

tional database management systems such as MySQL

or PostgreSQL. These technologies allow the usage of

free Operating Systems (e.g. Ubuntu Server) which

reduces overall implementation costs.

WooCommerce is a plugin of WordPress, a widely

used content management system. However, its adop-

tion would cause more complexity and potential se-

curity breaches since it would be necessary to in-

stall additional third-party plugins due to the lack of

some core features (e.g. search engine optimization

(NFR4)).

Regarding Zen Cart, it has a very small number of

maintainers and noticeable low contributions or new

versions. Therefore, we considered it not to be a good

option to take for a long-term scenario, once there is

no clear roadmap deﬁnition and it lacks relevant ex-

tensions in the marketplace and multi-language sup-

port (NFR3).

OpenCart providers do not maintain a change log

in order to anticipate negative impacts of upgrading

versions. Additionally, the native API does not pro-

vide products and attributes’ synchronization in an

easy and efﬁcient way.

The analysis determined that Magento and

PrestaShop were the best options to consider on the

current case study. Both platforms provide web ser-

vices to facilitate the integration with other enterprise

applications and an extensive documentation which

promotes the acquisition of technical knowledge by

users.

While Magento offers both open-source and

commercial solutions, developers who maintain

PrestaShop are totally focused on open-source devel-

opment. Furthermore, Magento’s learning curve is

considerably higher and could demand for paid train-

ing and support. Therefore, PrestaShop was consid-

ered the best option for this case study.

5 ARCHITECTURE &

IMPLEMENTATION

The proposed architecture for the integrated solu-

tion is illustrated in Figure 2, using different nodes

and associations according to the Uniﬁed Modeling

Language (UML) standard, allowing to represent the

physical view of deployment and various component

dependencies (Patig, 2010).

The Figure shows an UML deployment diagram

composed by 3 nodes: Client, Windows Server

and Ubuntu Server. The selection of software and

technologies took into account open-source options

(NFR5).

ICEIS 2021 - 23rd International Conference on Enterprise Information Systems

316

Table 1: Comparative analysis of e-commerce platforms with scoring from 1 to 5.

Weight (%)

OpenCart

PrestaShop

Magento

WooCommerce

Zen Cart

Setup and conﬁguration 5 4 4 2 4 3

Learning curve 10 4 5 2 5 3

Multi-language support 10 3 5 5 5 1

Search engine optimization 10 4 4 4 3 3

Marketplace of extensions and themes 5 4 4 2 4 2

Community and support 15 3 4 3 5 2

API 25 3 5 5 4 1

Documentation 10 5 5 5 5 1

Development roadmap 10 1 5 3 3 1

Total (%) 100 66 93 76 85 34

<<device>>

:Server

<<OS>>

Ubuntu Server

HTTPS

<<device>>

:Client

Browser

MySQL

Databases

Apache

Web Server

HTTPS

<<device>>

:Server

<<OS>>

Windows Server

SQL Server

Database

Ecommerce

Platform

HTTPS API

Sync

Web Application

Primavera

ERP

Staff

Web Application

Terminal

SSH

HTTPS API

IIS

Web Server

WebserviceERP

WCF Web Service

SqlClient

WebserviceSQL

WCF Web Service

COM Interops

Figure 2: Simpliﬁed system architecture.

Both “Windows Server” and “Ubuntu Server”

nodes were deployed on-premises, within the com-

pany’s local area network, and assuring that only the

“Ubuntu Server” node had Internet connection. Net-

work isolation and ﬁrewalls minimized the risk of oc-

currence of security breaches (C7).

ERP licensing and maintenance costs could repre-

sent a major slice of expenses in any enterprise appli-

cation integration. An inter-application middleware

was adopted to prevent hidden costs since the middle-

ware will host most of the integration business logic

(C8), leaving the opportunity to deﬁne which type of

technologies should be used (C2).

The performance of synchronization operations

across different enterprise applications and databases

could be critical in terms of scalability. As the cata-

log and order volume increases, processes slow down.

Thus, synchronization processes were optimized by

Architectural Challenges on the Integration of e-Commerce and ERP Systems: A Case Study

317

placing software applications and databases under the

same data center, so that applications could commu-

nicate using a local network and avoid extra-latency

linked to Internet usage (C9).

5.1 “Client” Node

This node represents the device used to access the sys-

tem through the company’s internal network or over

the Internet. It is used to access the e-commerce plat-

form and the “Staff” web application.

Some user proﬁles can execute routines of the

“Sync” web application component via a computer

terminal (FR93) and Secure Shell (SSH) connection

(C7).

5.2 “Windows Server” Node

This node provides the environment for the ERP

system execution and it is composed by a database

management system – MS SQL Server – and two

Windows Communication Foundation (WCF) web

services. Both web services are executed on the

Microsoft web server Internet Information Services

(IIS).

The development of those web services using Mi-

crosoft technologies allows to take advantage of the

ERP libraries, but also promotes the integration with

any other enterprise application through widespread

standard and protocols such as REST/JSON and Hy-

pertext Transfer Protocol (HTTP) (C2).

“WebserviceSQL” acts as a middleware which al-

lows secure operations on the ERP database by exter-

nal applications (NFR1). Incoming connections are

ﬁlter based on a set of permissions composed by a

list of allowed IP addresses for certain types opera-

tion (e.g. create, read, update and delete) (C7).

On the other hand, “WebserviceERP” facilitates

the access to core features of the ERP (NFR1). ERP

system’s APIs were leveraged in order to execute op-

erations such as create, void and retrieve documents,

from any other application, avoiding administrative

workers to access directly to the ERP and saving li-

censing costs (C6).

5.3 “Ubuntu Server” Node

This node is composed of a database management

system – MySQL – and a web server – Apache –

to support the correct functioning of the e-commerce

platform and of the two web applications: “Sync” and

“Staff”.

The “Sync” web application is an inter-application

middleware which supports data synchronization and

automatic operations related to e-commerce activity

(FR10 to FR16) (C3, C5). It connects to “Webser-

viceSQL” to execute database operations on the ERP

system, “WebserviceERP” to perform core ERP oper-

ations (e.g. create invoices) (FR17) (C2) and also the

e-commerce platform API. Exposes an API to provide

the list orders and the ability to create/void invoices or

change orders statuses (C5).

In contrast, “Staff” web application allows admin-

istrative and warehouse workers to perform daily op-

erations (e.g. view order statuses) without accessing

ERP directly (FR1 to FR6) (C6). It connects to the

API exposed by “Sync” and provides interfaces for

order management, manual invoicing, product man-

agement and errors details (e.g. synchronization fail-

ures, blocked orders, etc.) (FR7 and FR8) (C5, C6).

5.4 Implementation Considerations

The synchronization of e-commerce data occurs

through scheduled routines which can be executed

manually (C3). These routines are susceptible to fail

due to multiple reasons, including temporary network

issues and missing data in the ERP system, among

others. A retry mechanism was implemented to by-

pass temporary issues and every failure is logged by

the system for post-analysis.

Regarding payments, most of the e-commerce

platforms already support third-party providers such

as Paypal via extensions widely used and tested. Se-

curity protocols such as Secure Socker Layer (SSL)

and Transport Layer Security (TLS) improved the se-

curity of communication channels used not only be-

tween applications and users, but also in transactions

without human interaction (e.g. automated synchro-

nization processes) (C7).

The cost of the software components implementa-

tions was optimized by using open-source technolo-

gies, frameworks and libraries. Furthermore, the tech-

nical background of the development team has been

also taken into account since past experiences were

useful to build a robust solution (C8).

Existing extensions to integrate payment services,

online marketplaces, among other features, were im-

portant to reduce the development workload but also

future maintenance costs since these extensions are

maintained by authors or the community (C8).

Maintaining an efﬁcient product inventory is im-

portant to prevent stock outs and improve overall cus-

tomer experience. Although invoicing is carried out

later manually by operators, it was essential to make

a stock reservation of articles immediately after re-

ceiving the payment conﬁrmation as it prevents erro-

neous stock in both e-commerce platform and ERP

ICEIS 2021 - 23rd International Conference on Enterprise Information Systems

318

system. A virtual warehouse was used for this pur-

pose (C4), allowing for temporary stock on both ERP

and e-commerce systems as new orders are placed,

and preventing inconsistent stocks due to synchro-

nization processes.

6 CONCLUSIONS

This paper identiﬁes the software architectural chal-

lenges behind a common integration scenario be-

tween ERP systems and e-commerce platforms.

These challenges are derived from functional and

non-functional software requirements elicited for a

particular case study.

The proposed architecture for this case study in-

cludes the development of several components to

comply with the elicited requirements, and addresses

the various challenges. These also include the selec-

tion of an open-source e-commerce platform through

a weighted scoring model, in order to reduce the de-

velopment effort of the integration. Its successful im-

plementation validated the presented architecture as a

solution to address the case study and identiﬁed chal-

lenges.

Further research and case studies should be con-

ducted for the identiﬁcation of architectural and in-

tegration challenges that can be generalized for EAI

between ERP systems and e-commerce platforms.

Nevertheless, software architects and managers can

beneﬁt from the case study presented and solutions

adopted as this kind of scenarios is quite common

within enterprise information systems.

REFERENCES

BuiltWith (2021). ecommerce usage distribution in the top

1 million sites. https://trends.builtwith.com/shop. Last

checked on Jan 23, 2021.

Davenport, T. H. (1998). Putting the enterprise into the en-

terprise system. Harvard business review, 76(4).

Ecommerce Europe (2019). European ecommerce report:

2019 edition. https://www.ecommerce-europe.eu/

wp-content/uploads/2019/07/European Ecommerce

report 2019 freeFinal-version.pdf. Last checked on

Jan 23, 2021.

Eurostat (2020). Digital economy and soci-

ety statistics - households and individuals.

https://ec.europa.eu/eurostat/statistics-explained/

index.php/Digital economy and society statistics -

households and individuals. Last checked on Jan 23,

2021.

Farzaneh, M. K. (2014). Evaluation of use of erp in e-

commerce: Methods and strategies. research Jour-

nal of Applied sciences, engineering and technology,

7(20):4171–4174.

Holland, C. and Light, B. (1999). A critical success fac-

tors model for ERP implementation. IEEE software,

16(3):30–36.

Kaya, A. and Aydın,

O. (2019). E-commerce in turkey

and SAP integrated e-commerce system. Interna-

tional Journal of eBusiness and eGovernment Studies,

11:207–225.

Kong, Z., Wang, D., and Zhang, J. (2007). A strategic

framework for enterprise information integration of

ERP and e-commerce. In Research and Practical Is-

sues of Enterprise Information Systems II, pages 701–

705. Springer.

Lee, J., Siau, K., and Hong, S. (2003). Enterprise integra-

tion with ERP and EAI. Commun. ACM, 46(2):54–60.

Nikitovi

c, M. and Mahmutovi

c, A. (2019). Hidden costs

of ERP implementation. In 2019 42nd International

Convention on Information and Communication Tech-

nology, Electronics and Microelectronics (MIPRO),

pages 1314–1318. IEEE.

Nili, A., Barros, A., Johnstone, D., and Tate, M. (2019).

Technological enablers for preventing service failure

with e-commerce websites. In 27th European Confer-

ence on Information Systems (ECIS 2019), Stockholm

& Uppsala, Sweden.

Patig, S. (2010). Modeling deployment of enterprise ap-

plications. In International Conference on Advanced

Information Systems Engineering, pages 253–266.

Springer.

Patil, H. and Divekar, B. R. (2014). Inventory management

challenges for B2C e-commerce retailers. Procedia

Economics and Finance, 11:561–571.

Portuguese Government (2019). Decreto-Lei n.º 28/2019.

ario da Rep

ublica n.º 33/2019, S

erie I de 2019-02-

15. Regulamentac¸

ao das obrigac¸

oes relativas ao pro-

cessamento de faturas.

PRIMAVERA (2020). PRIMAVERA proﬁs-

sional. https://pt.primaverabss.com/pt/software/

software-de-gestao/professional/professional/. Last

checked on Dec 05, 2020.

Ruh, W. A., Maginnis, F. X., and Brown, W. J. (2002). En-

terprise application integration: a Wiley tech brief.

John Wiley & Sons.

Thaw, Y. Y., Mahmood, A. K., and Dominic, P. (2009).

A study on the factors that inﬂuence the con-

sumers trust on ecommerce adoption. arXiv preprint

arXiv:0909.1145.

Wang, Y. and Shi, Y. (2017). Analysis on the integration

of erp and e-commerce. AIP Conference Proceedings,

1864(1):020137.

Architectural Challenges on the Integration of e-Commerce and ERP Systems: A Case Study

319