EMERALD_BPMJ_BPMJ XXXXXXXXXX
A comparative study and
evaluation of ERP reference
models in the context of
ERP IT-driven implementation
SAP ERP as a case study
Hamdan Mohammed Al-Sa
i, Majed Al-Mashari and
Azeddine Chikh
Information Systems Department, College of Computer and Information Sciences,
King Saud University, Riyadh, Saudi Arabia
Abstract
Purpose – The purpose of this paper is to consider the question of what is an appropriate enterprise resource
planning (ERP) reference model for specifying areas of change in the context of IT-driven ERP
implementation and through the model matching. There are other implicit goals to increasing the awareness
of the reference models, as this highlights the principles embedded in ERP systems and explains the
classification of reference models, which is useful in terms of the reuse of knowledge.
Design/methodology/approach – In this paper, a comparison between ERP reference models is conducted
using a suitable decision-making technique and the final results are discussed. The comparison depends on
nine criteria related to conceptual ERP reference models: scope, abstraction, granularity, views, purpose,
simplicity, availability, ease of use for model matching, and target audience.
Findings – This study concludes that the business process reference model is best for specifying areas of
change in the context of IT-driven ERP implementations. The final ranking of the alternatives based on all
criteria places the system organizational model second, followed by the function and data/object reference
models, in that order.
Originality/value – This paper is one of very few studies on the selection of appropriate ERP reference
models according to the ERP implementation approach and model matching factors. This research also
provides an in-depth analysis of various ERP reference model types.
Keywords Decision-making, ERP, Business engineering by model matching, ERP reference models,
IT-driven ERP implementation approach
Paper type Research pape
1. Introduction
Cu
ently, many enterprises aim to support their processes using process-aware
information systems such as enterprise resource planning (ERP) (Pajk, XXXXXXXXXXERP
systems are off-the-shelf comprehensive software packages that support most of the key
functions of an enterprise. Moreover, ERP systems are increasingly being used to automate
and enhance business processes (Pajk et al., XXXXXXXXXXThus, facilitating business process
orientation is considered one of the most important factors when implementing such
systems (Dreiling et al., XXXXXXXXXXMany enterprises perform several activities such as
procurement, finance, sales, production planning, human resources, and material
management, and these can be embedded in an ERP package (Markandeya and Roy,
2014). However, ERP packages have a very complex design that makes them difficult to
understand, implement, and modify (Dodds and Spencer, XXXXXXXXXXERP offers many direct
enefits to an enterprise in terms of inventory reduction, order management improvements,
IT cost reduction, and also present intangible benefits such as information visibility, cost
eductions, integration, standardization, flexibility, and globalization (Gargeya and Brady,
2005). Admittedly, to reap the full benefits of ERP systems, extreme structural changes
Business Process Management
Journal
Vol. 24 No. 4, 2018
pp XXXXXXXXXX
© Emerald Publishing Limited
XXXXXXXXXX
DOI XXXXXXXXXX/BPMJ XXXXXXXXXX
Received 11 July 2016
Revised 18 December 2016
22 Fe
uary 2017
Accepted 24 April 2017
The cu
ent issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/ XXXXXXXXXXhtm
943
SAP ERP as
a case study
within the enterprise may be required, including business process reevaluation and
eengineering (Fryling, XXXXXXXXXXIn a
oad sense, ERP packages are distinguished by
comprehensive functions that apply in certain domains, and present different languages,
accounting standards, cu
encies, and so on (Manoj, XXXXXXXXXXAlthough ERP packages provide
different reference models and configuration tables, ERP projects are risky and have a high
failure rate (Chand et al., 2005; Ehie and Madsen, 2005; Ngai et al., 2008; Larsen, 2009).
Consequently, ERP implementation requires overall knowledge at all levels of the business
process, functions, structure, and objects through the available resource (i.e. reference
models) (Ku
el, 2013).
There are three approaches to implementing ERP systems: IT-driven, process-driven,
and hy
id. The first approach represents the adaptation of the enterprise to ERP
functionalities (Chiplunkar et al., 2003; Markus, 2004; McAdam and Galloway, 2005).
In contrast, the adaptation of an ERP system to the enterprise’s business process constitutes
the process-driven approach (Van Der Aalst and Weijters, 2004; Arao et al., 2005).
The hy
id approach is a mixture of the two (Arif et al., 2005; Soffer et al., 2005; Juntao et al.,
2007). However, research has shown that the IT-driven approach is better for allowing the
enterprise to take all possible advantages from an ERP system ( Juntao et al., 2007; Nazemi
et al., 2012; Panayiotou et al., XXXXXXXXXXResearchers and ERP vendors argue that the adoption of
these best practices makes the implementation of the enterprise systems less costly and
isky, and improves the enterprise’s processes (Stijn and Wensley, XXXXXXXXXXHowever, the
application of an ERP system without any adjustments to the enterprise’s business
processes will result in a risky and costly implementation (Panayiotou et al., 2015).
Reference models (RMs) are generic and conceptual models that implement the best
practice in a specific domain. In general, the term “reference model” denotes a formulation that
describes the structure and functionality of enterprise systems (Pesic and Van Der Aalst,
2005). Such an entity is usually developed from known modeling tools in the domain
according to a basic meta-model. The main purpose of RMs is to facilitate the use of complex
packages, adapt them to the specific problem, and configure an overall solution from these
adapted parts (Ramesh and Jarke, XXXXXXXXXXIn the same vein, ERP-specific RMs are the best way
to present the capabilities, functionalities, and different features of the ERP system (Pajk et al.,
2011). Additionally, ERP RMs exist in the form of models for business processes, functions,
system organization, objects, and data (Rosemann and Van Der Aalst, XXXXXXXXXXAlthough the
construction of RMs has received considerable attention, the process of reusing knowledge in
these RMs has been somewhat neglected (Reinhartz-Berger et al., XXXXXXXXXXAs a result, ERP RMs
equire further investigation and analysis to ensure they are employed co
ectly in any
desired implementation.
The main problem faced by enterprises and government organizations when
implementing solution-based ERP systems is how to define the areas of change within
the enterprise so as to meet the ERP capabilities quickly and cost-effectively (Pe
one, 2004;
Daneva and Wieringa, 2005; Law and Ngai, 2007; Alsulami et al., 2014; Gao et al., 2010).
Specifying the areas of change in an IT-driven implementation refers to the business
engineering that focuses on the enterprise structure, readiness, and business processes that
fit with the enterprise system capabilities. Unfortunately, one of the challenges in applying
RMs when implementing ERP is that the models do not generally include any information
about what can be modified and how the modifications will appear (Kirchmer, XXXXXXXXXXHence,
establishing sound and consistent predefined areas of change through the RMs, and in the
context of model matching, involves selecting and reusing suitable RMs. Furthermore, using
and employing the RMs appropriately requires an in-depth analysis of different degrees of
abstraction, components, and so on.
This research stresses the necessity for studying different types of ERP RMs and
choosing the most suitable using an appropriate decision-making technique that can help to
944
BPMJ
24,4
specify the enterprise areas of change in the context of business engineering and model
matching. To address these concerns, the objectives of this research are: to demonstrate the
key concepts and principles of the ERP package levels and ERP-specific RMs, to investigate
usiness engineering using the RMs, to review the academic literature on RM comparison
criteria, to study and analyze ERP RM types, and to select an appropriate ERP RM using a
decision-making technique in the context of business engineering through the process of
model matching.
The structure of this paper is as follows: Section 2 introduces the theoretical background
elevant to ERP, as well as basic concepts and components of the RMs. In Section 3,
usiness engineering using IT RMs will be discussed. The evaluation criteria for comparing
RMs and related work are presented in Sections 4 and 5, respectively. A case study using
SAP ERP RMs is described in Section 6. The research methodology and apply the AHP
technique to select an appropriate ERP RM are outlined in Sections 7 and 8, respectively.
Our results are discussed in Section 8. Finally, the conclusions to this study and ideas fo
future research are given in Section 9.
2. Preliminaries
2.1 RMs: theoretical background
A model is a simplified representation or abstraction of the reality. The main purpose of
graphical representations is usually to facilitate an understanding of an abstract concept
(Monk andWagner, 2012), and to make complex information systems controllable (Thomas,
2006). Moreover, graphical schemes are typically more intuitive than a written description.
Models increase user acceptance and reduce the cost of ERP implementation (Scheer and
Habermann, XXXXXXXXXXConstructing a model from a set of components with a description of how
to use these components is called a modeling method (Scheer and Habermann, 2000).
Overall, the aim of models is to reduce the complexity of the reality in order to bette
understand a complex system. RMs are generic conceptual models that present or formalize
the best practices for a certain domain (Frank, 1999; Fettke and Loos, 2003; Rosemann and
Van Der Aalst, XXXXXXXXXXIn practice, capturing common processes in a variety of enterprises
ather than a specific process in one enterprise produces an RM (Reinhartz-Berger et al.,
2010). These are sometimes called generic models, universal models, or model patterns
(Fettke et al., 2006; Pajk, XXXXXXXXXXOften, the term of best practice refers to the capture of
eusable state-of-the-art methodologies, with a focus on a specific application domain
(Pesic and Van Der Aalst, XXXXXXXXXXThe main purpose of RMs is to describe the functionality
and structure of a package for off-the-shelf solutions such as ERP (Rosemann and
Van Der Aalst, XXXXXXXXXXMoreover, the lifecycle of RMs is determined by enterprise systems
vendors who represent the designers. Thus, the application of RMs is motivated by the
design-by-reuse paradigm (Rosemann and Van Der Aalst, 2007; Assy et al., 2015).
RMs are developed for several objectives: to streamline best practice processes and
provide a generic solution (Rosemann and Van Der Aalst, 2007), to improve the
understanding of complex enterprise systems by describing their functionality and
structure (Rosemann, 2000), to provide end-users with a comprehensive view of software
functionality, to assist in capturing specific domain knowledge (Reinhartz-Berger et al.,
2010), and to reduce the costs of implementing enterprise systems (Winkelmann, 2012).
Furthermore, RMs are expected to fit the