Computers in Industry 56 (2005) 573–587

www.elsevier.com/locate/compind

A classification for better use of ERP systems

Valérie Botta-Genoulaz a,*, Pierre-Alain Millet b
a
Laboratoire PRISMa, INSA de Lyon, INSA-GI, bât Jules Verne, 19 Avenue Jean Capelle, 69621 Villeurbanne Cedex, France
b
Laboratoire PRISMa, INSA de Lyon, INSA-IF, bât Blaise Pascal, 7 Avenue Jean Capelle, 69621 Villeurbanne Cedex, France
Received 29 March 2004; received in revised form 14 December 2004; accepted 13 February 2005
Available online 14 July 2005

Abstract

Companies have invested considerable resources in the implementation of enterprise resource planning (ERP) systems. The
results initially expected have rarely been reached. The optimisation (or efficient use) of such information systems is nowadays
becoming a major factor for firms striving to reach their performance objectives. After presenting a synthesis of several studies
on ERP projects, we build on the findings of a French investigation into the assessment and optimisation of ERP performance. A
classification of company positions regarding their ERP use, based on both software maturity and strategic deployment
directions, and an improvement process are proposed. Industrial cases allow validation of this approach.
# 2005 Elsevier B.V. All rights reserved.

Keywords: Integrated information system; ERP; Improvement; Survey; Benchmarking

1. Introduction expectations: 85% of companies consider the ERP

as an investment for more than 5 years, 70% expect no
During recent years, companies have invested more than 25% of return on investment and 50% did
considerable resources in the implementation of not even try to estimate the ROI.
enterprise resource planning (ERP) systems. ERP is Pressure over deadlines due to the ‘‘year 2000’’ or
a software package that attempts to integrate all ‘‘Euro introduction’’, need for revision of initial
departments and functions of a company onto a single ambitions to maintain an acceptable rhythm of
computer system that can serve all different depart- change, difficulty of integration with the peripheral
ments’ needs. Once the system is going on, the applications, obligation to match daily work with the
question of measurement of the operational results required project workload . . .. There are many brakes
obtained remains. But it becomes apparent that on initial objectives. Yet, companies still claim more
nowadays, results do not live up to managers’ value in their tools, in their processes and in their
human resources. At the same time, new needs as well
* Corresponding author. Tel.: +33 4 72 43 60 74;
as new ERP features arrived at maturity: management
fax: +33 4 72 43 85 38. and optimisation of the global supply chain (SCM),
E-mail address: [email protected] (V. Botta-Genoulaz). customer relationship management (CRM), business

0166-3615/$ – see front matter # 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.compind.2005.02.007
574 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587

intelligence. They point to other performance oppor- Calisir and Calisir [19] evaluate specially end-user
tunities. satisfaction factors from the human computer integra-
The optimisation of the information system (which tion point of view. Neither company satisfaction nor
is based on the ERP) becomes a major target for the the efficiency of the use of the new information system
company in reaching its performance objectives. By is subjected to particular study. The aim of this paper is
‘‘optimisation of the information system’’, we under- to study objectives, which correspond to the ‘‘post-go-
stand efficient use of the available technical, human live’’ phase of ERP implementation.
and organisational resources mobilised around the In Section 2, we present a synthesis of three recent
integrated information system. Naturally, there are French surveys and put in light important features
also resulting projects, which are often new applica- about returns of experiment and management issues in
tion implementation projects. This raises several ERP implementation projects. These first conclusions
questions: are completed by the results of some other national
studies (USA, Canada and Sweden). In Section 3, we
 How does an ERP implementation contribute to present the findings of a new qualitative investigation
make the organisation more effective? undertaken in 2003 on ‘‘optimisation and assessment
 Does the company make the most of the potentials of the ERP’’, within a workgroup made up of
of the ERP? academics, manufacturers and consultants and lead
 Is coherence ensured between the information by the Pole Productique Rhône-Alpes (France).
system, the business processes, the management Thirty-five firms were interviewed about their IT
rules, the procedures, and the competency and optimisation strategy concerning ERP. Analysis of the
practices of the users? qualitative answers (indicators and good practices
 Are activity-data and master-data reliable and identified) allows us to propose, in Section 4, a
relevant? classification of company positions regarding their
 Is the ERP well positioned in terms of ‘‘information ERP use. The two axes considered to evaluate the
system urbanisation’’? efficiency of the information system use (software
maturity and strategic deployment) reveal three stages
Despite the wide ERP systems base installed, of optimisation: operational, tactical and strategic.
academic research in this area is relatively new. Like Section 5 introduces some industrial cases that allow
many other new information technology (IT) areas, validation of the proposed classification.
much of the initial literature in ERP consists of articles
or case studies either in the business press or in
practitioner focused journals. More recently, several 2. Synthesis of ERP surveys
academically oriented papers have dealt with various
aspects of ERP. Since 2000, numerous reviews on ERP projects
The first publications dealing with ERP and their have been undertaken in Europe or in USA. Some are
implementations date from the 1990s; they emanated quantitative or qualitative surveys, other are based on
mainly from consulting groups. Since then, numerous case studies. This section presents a synthesis of three
academic researchers have gradually taken an interest recent French surveys about management issues in
in this phenomenon that in the main impacts ERP implementation projects. Preliminary findings
companies [1–13]. They mainly deal with the are followed up by other investigations conducted in
management of ERP implementation projects, their USA, Canada and Sweden. The objective is to bring
impact and success factors. There are also case-based out some relevant elements for the problem of
studies [14–16]. Somers and Nelson [17] were optimisation of ERP use.
interested in ERP project lifecycle including some
‘‘post-go-live’’ phases. Hunton et al. [18] compare 2.1. Survey characteristics
business performance of adopters and non-adopters
from the economic aspect; they suggest that ERP These surveys of ERP implementation in manu-
adoption helps firms gain a competitive advantage. facturing firms aimed to analyse the return on
 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587 575

experiment from the ERP project, to identify critical Table 1

Major company’s motives
success factors and to investigate future developments.
They concerned with penetration of ERP, motives, A (%) B (%) C (%)
implementation process, functionalities implemented, Solve the Y2K problem 32 88 35
major obstacles and operational benefits. Technical evolution, replace 45 67 45
legacy system
The first survey (denoted A) was carried out by
Restructure company 53 50 n.a.
Canonne and Damret [20] from June 2001 to February organisation/organisation
2002 among 3000 French companies with more than model evolution
100 employees (response rate of the order of 5%). Of
the responses, 54% of the companies have imple-
mented or were in the process of implementing an 2.2. Synthesis of survey results
ERP, 13% were planning to implement one within the
next 18 months and 33% had no plans for an ERP 2.2.1. Company’s motive to implement ERP
system for the near future. Each survey had its own yes/no questions, which
The second one (denoted B) was conducted by makes the synthesis difficult. However, the commonly
Deloitte & Touche consulting group [21] from quoted motivational factors remained: ‘‘solve the Y2K
November 2001 to May 2002 among 347 small and problem’’, ‘‘replace legacy systems’’ and ‘‘restructure
medium-sized companies (mainly situated in the company organisation’’ (see Table 1). All companies
south-east part of France), which already had an ERP had been operating with a patchwork of legacy
(response rate of 16.4%). systems that were becoming harder to maintain and
The third survey (denoted C) was carried out by the upgrade, and the competitive pressures on them
Pôle Productique Rhône-Alpes [22], from January to required increasingly more responsive systems with
April 2002 among 400 medium-sized industrial real-time integrated information that the legacy
companies which had implemented (65%) or were systems could not provide easily. We observe also
in the process of implementing an ERP in the Rhône- some arguments linked to the overall improvement of
Alpes region of France (response rate of 11.3%). the information system or to company willingness to
The findings from these three studies will be have a system able to improve its performance:
completed by three other international investigations. increase performance (B: 67%), simplification and
Mabert et al. [23] between August and October 1999 standardisation of systems (A: 59%), increase
surveyed, 193 manufacturing companies in the US, productivity (B: 54%), ease of upgrading systems
which had adopted an ERP, to study the impact of (A: 42%), improve interactions and communications
organisation size on penetration of ERP, motivation, with suppliers and customers (A: 39%). Other factors
implementation strategies, modules and functional- had lower priority: gain strategic advantage (A: 19%),
ities implemented, and operational benefits from ERP response to market evolution (B: 21%).
projects. Kumar et al. [24] investigated critical Mabert et al. agreed with the former findings:
management issues in ERP implementation projects 85.9% (83.3%) of respondents listed ‘‘replace legacy
in 2002 among 20 Canadian organisations; they systems’’ (respectively, ‘‘simplification and standar-
studied selection criteria (ERP vendor, project disation of systems’’) as either important or very
manager and implementation partners), constitution important. On the other hand, they identified ‘‘gain
of project team, project planning, training, infra- strategic advantage’’ in third position. This last factor
structure development, on-going project management, showed a significant difference between large (91.8%)
quality assurance and stabilisation of ERP. Olhager and small firms (70%): this can be explained by the
and Selldin [25] from November 2000 to January 2001 necessity of strategic financial analysis in large
surveyed ERP implementations in 511 Swedish companies. A surprising result from their survey is
manufacturing firms, concerned with ERP system that the Y2K issue was ranked very low (56%) among
penetration, the pre-implementation process, imple- motivational factors.
mentation experience, ERP system configuration, According to Olhager and Selldin, the two most
benefits and future directions (response rate of 32.7%). commonly quoted motivational factors are ‘‘replace
576 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587

Table 2 versions (33%), limited reporting capabilities

Major module implemented
(25%), some legislated procedures not supported
A (%) B (%) C (%) (15%). In such a case, software was modified (65%),
Financial accounting and control 78 83 72 add-ons were developed (50%) and/or business
Purchasing and material management 90 87 100 processes were re-engineered (30%).
Sales management 84 87 96
Production planning and control 83 62 86
Distribution, logistics 60 61 62 2.2.3. Implementation parameters
The strategy used for the implementation is one of
the most important factors in assessing the impact of
legacy systems’’ (4.11/5) and ‘‘simplify and standar- an ERP system on an organisation. Strategies can
dise systems’’ (3.67/5), whereas ‘‘restructure com- range from a single go-live date for all modules (Big-
pany organisation’’ (2.7/5) and ‘‘solve the Y2K Bang) to single go-live date for a subset of modules
problem’’ (2.48/5) had lower priority. (Mini Big-Bang) to phasing in by module and/or site.
According to B and C surveys, which more concerned
2.2.2. ERP module or functionality implemented small and medium-sized firms, ‘‘Big-Bang’’ is the
The modules the most frequently implemented deal most frequent (more than 60%); this ratio is inverted in
with purchasing and material management, sales, survey A, where the presence of large companies is
production planning and control, and financials (see more important. Both Olhager and Selldin and Mabert
Table 2). To a lesser extent, we find human resources et al. confirmed this finding. The latter authors add that
management (about 40%), quality management (about among the major implemented functionalities, pro-
45%), maintenance management (about 30%) and duction planning is often delayed until later.
R&D management (about 20%). Other functionalities Generally, ERP implementation times are often
were expected but are still absent, such as customer under-estimated, and are exceeded in about 50% of the
relationship or customer service management, and cases. The real duration corresponds on average to
business intelligence. 150% of duration foreseen with one or even two
Surveys show that not all the companies opt for a adjournments of the start-up date. Survey B informs us
unique ERP vendor: according to A, a majority of about the causes of the delays: customisation
companies (85%) opted for a unique software problems (17%), reliability of the tests (16%), data
package; however, about 50% of them keep specific migration (12%), specific developments not ended
tools to cover particular business or functional needs. (13%), elimination of ‘‘bugs’’ (9%), training not
Mabert et al. reveal that small firms prefer a single ended (8%), organisation not ready at the time of ‘‘go-
ERP package (56.6%), while large firms add other live’’ (8%). These findings tend to confirm that while
systems to the single ERP package. Moreover, survey the Big-Bang approach usually results in the shortest
A indicates that nearly half the firms had to adjust the implementation time, it is also the riskiest approach
system on the main functionalities, which generated because it can expose the entire stability of a company
an additional cost, about 12.3% of the budget. in case of any problems.
Olhager and Selldin agreed with these findings. ERP budgets and ERP costs are often confidential
They added information about the customisation effort data and it is difficult to synthesis such information in
needed to match company requirements: the modules a representative manner. Moreover, results about cost
that are related to the time-to-customer process are the drift are generally proportional to implementation
most frequently implemented but also the most time drift.
frequently customised. Kumar et al. clarified that
point: one of the major challenges an adopting 2.2.4. Benefits and obstacles
organisation faces while configuring an ERP system The first survey (A) highlights different rates of
is that software does not fit all their requirements [1]. satisfaction according to modules: the users are rather
About 44% were organisation specific, where the satisfied (rate superior to 75%) by finance/accounting,
software did not support the way the organisation purchasing, materials management and sales manage-
worked; for example, limited industry specific ment modules. Although they are often the subject of
 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587 577

Table 3 (redundancy of information, choice of the data and

Synthesis of main benefits
messages to be exchanged). It seems that culture
A B C ‘‘management by objective’’ was not extended to ERP
(%) (%) (%)
projects.
Availability of information/quickened 55 71 n.a. Kumar et al. agreed with these findings; they
information response time
present major obstacles that companies faced in the
Increased interaction across the enterprise, 37 n.a. n.a.
integration of business operations/processes ERP implementation projects such as problems in
Improved lead-time 24 74 45 transition to new systems (data migration), unavail-
Improved inventory levels and purchasing 33 74 21 ability of skilled people, high turnover of key project
Improved interaction with customers 18 36 n.a. persons, cost escalations and difficulties in estimating
Improved interaction with supplier 11 59 n.a.
the project requirements. They highlighted on training
Reduced direct operating costs 5 42 35
and ‘‘shakedown’’ challenges.
The first survey (A) investigates further projects;
several developments are operational or planned such
specific developments, production planning and as finite capacity planning (38%), business warehouse
logistics/distribution modules present a rate of weaker (38%), e-business (29%), CRM (27%) and SCM
satisfaction. The second survey (B) measures the (24%). The second survey (B) was interested in the
appropriation of the system by the users: 26% of the evolutions planned by companies; after the develop-
respondents considered it high, 39% satisfactory and ment of new functionalities (23%), the optimisation of
35%weak. the use of their system (10%) arrives in second
Main benefits are synthesised in Table 3. Most of position, followed by the change of version (7%), the
the perceived improvements correspond to the internationalisation of the company thanks to the ERP
expectations, which companies had, but not necessa- (7%), the development of decision-making and the
rily in the same measure: the improvement of business participant implication (7%).
indicators (number of backorders, stock shortage and
customer service rate) is far from being reached, and
the surveys do not allow us to deduce the reasons. 3. An investigation on ERP improvement
Furthermore, the reduction of direct costs (or IT strategy
costs), one of the main objectives of the projects, is not
quoted in the major results obtained. The survey B The previous results show the wide variety of
points out the improvement in traceability (86%), situations that may trigger interest in ‘‘optimisation’’
which was not one of the more expected results. of an ERP in the meaning given in Section 1. These
Olhager and Selldin agreed with these findings. differences can be explained by sample characteristics
In synthesis, ERP improved the global vision of the such as company size and typology, sector or respon-
company and the collaborative work permitting dents.
master-data harmonisation, considerable reduction Ross and Vitale [26] compared the stages of an ERP
of information redundancy and work in real-time. implementation to the journey of a prisoner escaping
The three surveys A, B and C also allowed from an island prison. They identified five stages: (1)
identification of problems encountered by companies ERP design/the approach; (2) ERP implementation/
during ERP implementation. They are mainly related the dive; (3) ERP stabilisation/resurfacing; (4)
to the adaptation of the company to the ‘‘ERP model’’ continuous improvement/swimming; (5) transforma-
or of the ERP to company specific requirements (about tion. Until now, researchers have investigated the ERP
76%), to the resistance to change (membership of the implementation process up to the stabilisation stage in
users, conflicts and social problems), to the resources order to identify the stage characteristics, critical
of the project team (user availability, deficiencies success factors of implementation, best project
of the integration teams, under-estimation of the practices. Fewer authors have worked on the two
resources) and to the problems of data exchanges latter stages, i.e. optimising the use of the information
between the ERP and the existing information system system for company development and performance.
578 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587

A project has now been launched in the Rhône- Table 4

Sample description
Alpes region (France) to identify best practices of ERP
‘‘optimisation’’ in companies, and their application No. of
firms
context. The aim is to propose a typology of these
‘‘post-go-live’’ situations for small and medium-sized Branch of industry
Metallurgy, mechanics, 12
firms.
construction of machines
Components electric, electronic, 6
3.1. Research methodology data processing
Manufacture of pieces of furniture, 5
A working group lead by Pole Productique Rhône- articles of sports and various
Textile industry, clothing, leather and shoes 4
Alpes has been created bringing together researchers,
Industry of rubber and the plastics 3
consultants, and industry to undertake a qualitative Chemical and pharmaceutical industry 2
survey. The study was carried out between January and Others 3
March 2003 among 217 manufacturing companies in Size (no. of employees)
the Rhône-Alpes region that have an ERP ‘‘stabilised’’ Small (<100) 4
for at least 1 year. Medium (100–500) 16
The survey questionnaire asked for information on Large (>500) 15
ERPimplementationandcurrent usein the company: the Typology of production
respondent’s and the company’s characteristics, the MTS (make-to-stock) 74 a
ERP project characteristics and their initial contri- MTO (make-to-order) 74 a
ATO (assembly-to-order) 54 a
bution (motives, timelines, budgets, functionalities, a
benefits, user satisfaction, . . .), organisational charac- Percentage of firms.
teristics (during and after stabilisation), needs of
improvement/evolution and ‘‘post-go-live’’ diagnostic.
It was 10 pages long. Most questions in the survey annual revenue between s15 million and s300
required multiple responses. The responses were million, and belong to a group (76.7%). The number
encoded using a mix of check boxes, open-ended of employees ranged from 130 to 1400; however, the
answers and a binary scale with ‘yes’ or ‘no’ responses. sample also includes large corporations and large
The amount of open-ended questions allowed us divisions within corporations.
to appreciate numerous details but caused some
difficulties for the analysis.
After the initial development of the survey ques- Table 5
Project characteristics depending on implementation strategy
tionnaire, it was pre-tested with two respondents to
Implementation strategy Big-Bang By stage
check its validity: the primary objective was to test
whether the instrument provided consistent and Average implementation 20.56 26.14
time (month)
accurate information. The mailing was sent out in the
middle of January 2003 to IT managers or CEOs Level of satisfaction (%)
(managing director). Fourteen qualitative interviews Very good 81 25
Good 9.5 37.5
were conducted in parallel to guarantee response inter- Fair 9.5 37.5
pretation. By the end of March 2003, 35 responses had Insufficient 0 0
been received, for an overall response rate of 16%. Given
Process formalisation (%)
the length and comprehensive nature of the survey, this Yes 54.5 25
response rate was concluded to be reasonable. Partially 45.5 75
No 0 0
3.2. Enterprise characteristics Recurring IT audit (%) 50 25
User documentation (%) 86.4 100
The enterprise characteristics are summarised in Updated documentation (%) 68.4 50
Table 4. The majority of the responding firms have an Existence of user group (%) 50 75
 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587 579

3.3. Investigation results Table 6

Differences between installed and used functions

3.3.1. Diagnosis of ERP projects and use Function No. of firms No. of firms
who install who use
These projects, introduced by the head office in 73%
of the cases, were characterised by an average budget of Forecasting/marketing 17 16
SOP/MPS 24 22
s2.57 million, of which s1.37 million of external
MRP 28 27
services. The average implementation timewas about 22 Planning/scheduling 27 25
months, while it was estimated as 17 months: 63% of the Dispatching and transportation 28 27
firms under-estimated this parameter. The Big-Bang Warehouse management 20 19
strategy was used in 74% of the cases. Several other Quality 18 16
Maintenance 15 14
characteristics have been studied according to the
After sales service 14 12
implementation strategy (see Table 5). Controlling 27 26
It emerges that user satisfaction level is better in Business intelligence 13 12
case of Big-Bang, as well as the degree of formalisa-
tion of the business processes. More than 75% of
companies consider themselves (very) satisfied with
the project. Companies that implemented an ERP system are
The main functions installed deal with production, experiencing improved performance mainly from the
sales and logistics: sales management, material information perspective. Information is more easily
management, shop floor management, material re- accessible and the interaction across the enterprise has
quirement planning, dispatching and transportation, improved, as indicated in Table 7. Even if they were
planning and scheduling. It has been noticed that of part of the initial motives, some issues like financial
the 23 available functions, 12 are not always used (see flows control, lead-time control, cost control and
Table 6). For 52% of them, the ERP encompasses inventory control have been affected to a lesser extent.
more than 75% of their entire information system, but Furthermore, another unexpected benefit emerged: the
most have at least one functionality covered by a clarification of the organisation.
specific development: 56% (respectively, 40%) of Firms were also questioned about the traps that can
the respondents interface their ERP with external appear during ERP projects. This was an open
software for the payroll (respectively, human resource question and we collected more than a thousand
management). responses, which were classified in nine categories

Table 7
Benefits
% Expected % Realised % Expected/% realised
Direct benefits
Control of flows of goods 70.0 70.0 100
Information flows control 76.7 73.3 96
Financial flows control 73.3 63.3 86
Services/department opening up 50.0 50.0 100
Information reliability 83.3 66.7 80
Uniqueness of information 86.7 80.0 92
Organisation clarification 36.7 43.3 118
Common view across the company 60.0 46.7 78
Process benefiting from ERP
Cost control 76.7 56.7 74
Lead-time control 83.3 43.3 52
Inventory control 80.0 66.7 83
Customer service improvement 70.0 70.0 100
Supplier relationship improvement 46.7 40.0 86
580 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587

Table 8 Table 10
Traps in ERP projects Considered evolutions
Traps % Considered evolutions %
Lack of re-engineering before 40 Deployment of new functions 43.3
ERP project Optimisation of existing tools utilisation 20.0
Lack of project planning 30 Upgrade of version 16.7
Gap in the requirement definition 30 Implementation of business intelligence solutions 16.7
Under-estimation of the 23 Geographic deployment on multi-site companies 16.7
importance of the choice ERP
Specific software development, 20
too much customisation The importance of these actions must be underlined
Lack of training 19
because of the great involvement of top management
Lack of planning post-go-live, 13
gap in stabilisation phase management in these projects in 65.38% of the cases. After the
Lack of communication and 13 implementation of the ERP, the organisation of the
implication of the management company was adapted by the creation of an ERP centre
Under-estimation of data-migration risk 13 of competence (66.7% of the cases), the formalisation
of owners of process (78.3%) and the definition of the
operational roles on the processes (60.9%).
(see Table 8). It emerges that the major trap is linked The detailed list of motives and of actions
with business process definition or re-engineering. undertaken leads us to the classification proposed in
The survey was not focused on the project phase Section 4.
but on the post-go-live phase; nevertheless, these
elements may explain the improvement actions that
are undertaken and presented in the following section.
4. Company position classification regarding
3.3.2. Projected improvement ERP use
Ninety percent of the respondents consider it
necessary to optimise the conditions of use and 4.1. Analysis of qualitative answers
functioning of their ERP system. Near one on two
have an outstanding project or a project in preparation. Reasons for improvement as described in Table 9
Their motives are presented in Table 9. are of different kinds. Some arise from the need to
The evolutions considered by the respondents are master an application, which does not give satisfac-
presented in Table 10. Among new functions, we find tion, when corrective action is required due to
CRM, maintenance (MRO), product lifecycle man- unsatisfactory results on the initial project. Others
agement (PLM), manufacturing execution system arise from internal improvement objectives or con-
(MES), quality management (TQM), warehouse secutive to external changes. This diversity of reasons
management, engineering change order, . . .. The for optimisation leads to a classification of companies
optimisation projects mentioned mainly deal with according to their ERP use. Existing quantitative
planning integration, archiving, access security, data answers are always relative to a situation, a line of
reliability and reporting. business, a maturity of the company . . . and no
reference norms can be used to compare and segment
Table 9 these results. On the contrary, qualitative answers like
Motives for optimisation the reasons for improvement seem to be more
Motives % significant in a ‘‘profile’’ regarding the ERP use.
Better use and exploitation of the ERP 30.0
Whatever the volume of data or the line of business,
Expected results not reached 23.3 when a manager identifies ‘‘poor data reliability’’ as a
Insufficient knowledge of the system installed 16.7 reason for an optimisation project, it identifies clearly
Evolution of needs 13.3 a situation, which can occur in all enterprise
Evolution of the environment 13.3 typologies. The current section proposes a typology
 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587 581

Table 11
Classification of identified traps
Traps with ERP as a tool Traps with ERP as an enterprise model
 Too much customisation  Ill defined requirements
 Lack of training  Lack of communication and implication of the management
 Under-estimated work and weak planning method  Lack of business process re-engineering (BPR) before project
 Lack of planning post-go-live, lack of stabilisation phase  Confusions between ERP, BPR and management (ERP choice
before/after BPR, change management approach/BPR)
 Purge and make data consistent before migration  Under-estimation of the importance of the choice ERP

Table 12
Classification of expected improvements
ERP tool improvement Enterprise model improvement
 Version/release migration  New applications: MRO, CRM, TQM, PLM/PDM
 Authorisations control, data archiving,  New ERP modules: analytics accounting, engineering
quality and reliability of data change order, master plan schedule
 Data flow automation: B2B, WebEDI and bar-coding  Decision-making support: supply chain planning,
performance indicators, business intelligence

of ERP use situation built on a classification of Finally, as is noted in Table 7, the achievement rate
qualitative answers to this survey. for direct benefits is better than that for process related
Regarding traps identified in the survey, some are benefits, which allows us to differentiate maturity of
clearly related to the application as a tool, which has to software mastery from process control.
be appropriate to enterprise users. Others are related to
the application as a model of the enterprise, embedded 4.2. Proposition of a two-axis classification
standard business processes inside the application, as
an implicit model of running the enterprise. Table 11 All of this leads us to identify two axes to measure
summarises the proposed classification. perfect command and control of the information
The improvements planned by the respondents can system; each is split into three levels. The first axis
also be differentiated between technical improvements/ entitled ‘‘software maturity’’ relates to the good use
complements and functional extensions to meet enter- of these systems from the point of view of their
prise needs, as presented in Table 12. ERP project are proper efficiency, and is separated into software
generally long journey and in a majority of enterprises, mastery, improvement and evolution. The second,
deployment in other subsidiaries or on other sites are still entitled ‘‘strategy deployment’’, relates to the
planned even after 2 or 3 years of ERP use. contribution of the information system to the
Regarding the reasons for optimisation, some are performance of the company itself, to its global
typically relevant to malfunctioning; some, on the efficiency; it is separated into master-data control,
contrary, mark efficient use, allowing working for process control and strategic support. Fig. 1 illus-
better results; others result from technical or con- trates the different possible stages depending on the
textual evolution, as indicated in Table 13. two axes.

Table 13
Classification of optimisation reasons
Malfunctioning Improvement Evolution
 Insufficient knowledge of the system installed  Better use and exploitation of the ERP  Evolution of needs
 Incapacity in the definition of the working procedures,  Expected results not reached  Technical evolution
the roles of the ERP
 Incapacity in the definition of needs/specifications
582 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587

Fig. 1. Improvement process matrix.

Table 14
Axis of software maturity
Level Alerts Actions
Software mastery  Non-appropriation of the system by the users  Additional training of the users
 Unsatisfactory operational execution  Create a competence centre
 Insufficient speed/ability to react  Empowerment of the users on their role and
on their duty (user’s charter, quality indicators)
 Insufficient system response time  Stabilisation of the execution (indicators with
 The users create parallel procedures follow-up of objectives)
 No documentation on parameters, data,
data management procedures
Improvement  The full ERP potential is not used  Definition of performance indicators, business indicators
 Results not reached, expectations unsatisfied  Improvement and automation of the reporting
 The standard system installed does not fit  Rethink the roles to simplify the procedures
all requirements
 The number of office automation  Implement the functions that are not yet used
utilities increases
 The procedures are too heavy
Evolution  Context ‘‘multi-activities’’, international firm  Standardisation on several sites/activities
 Reorganisations, technological changes  Version upgrade
 Need of (analytics) reporting  Address the ERP/environment technological
evolution (EAI)
 Outside integration: B to B  Develop business intelligence systems
 Bar-code integration  Implementation of enterprise architecture
 Version upgrade (application mapping)
 Software maturity depreciation
 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587 583

Table 15
Axis of strategy deployment
Level Alerts Actions
Master-data control  Numerous erroneous technical data  Cleaning of the migrated data
 Messages of ERP not relevant (stock shortages,  Define responsibility for data
rescheduling in/out MRP, purchase proposals)
 Numerous manual inventory corrections  Assert the uniqueness of the data in the
whole company
 Product lifecycle not improved,  Indicators of data control
not integrated in the IS (revision . . .)  Maintain a business project team with
a plan of action to master-data
Process control  Conflicts between services on procedures  Revise management rules in the company
 Contradictions between local and  Verify the appropriateness of the tool to
global indicators the organisation
 Results not reached, needs unsatisfied  Rethink the roles to simplify the procedures
 Demands for improvement, for roles  Define responsibility for processes
redefinition by the users
 Higher expectations of customers and  Strengthen the transverse responsibilities:
top management indicators, communication
 No return on investment calculated
Strategy support  Business objectives not reached  Modelling and optimisation of the supply chain
 Higher expectations of customers and  External integration: B to B
top management
 Changes of markets, of customer expectations  Implementation of application mapping
 International extension  Business Process Management (modelling,
process performance measure)
 Management expectation concerning  IT associated to business strategies
follow-up consultancy

We propose on this matrix a synthetic vision of the These three optimisation stages allow us to character-
process of optimisation. It underlines the constraint of ise three situations (numbered 1–3), which are defined
coherence between both axes. The information system below.
cannot support company strategy without being Situation 1 is described as a result of an operational
mastered as a ‘‘tool’’. Certain situations are conse- optimisation centred on the good use of what exists
quently impossible (control of the processes without (‘‘master the tools to master the data’’). To reach
mastery of the software). Every level is defined by situation 1, the information system is considered as a
alert criteria allowing recognising it, and by the typical tool of production and broadcasting of data.
actions of improvement to be implemented at this Situation 2 is described as a result of a tactical
level. These alert criteria and improvement actions are optimisation centred on the best integration of what
presented in Table 14 for the software maturity axis exists to allow a more effective use (improve ERP use
and in Table 15 for the strategy deployment axis. for better control on the processes). To reach this
situation 2, the information system is considered as a
4.3. Towards a guideline for ERP use support for the control of company operational
improvement processes.
Situation 3 is defined as the maximal use of the
On this basis, we can bring out a process of information system focused on a strategic optimisa-
optimisation (in the sense of a better use of the tion leading to modification the positioning of the
information system) in three stages, which produce an existing ERP in the information system strategy. The
information system contributing to the strategy of the information is then a real component in defining the
company (situation number 3 on the matrix, Fig. 1). strategy of the company.
584 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587

This approach matches the last three stages defined The technical data were considered as unreliable and
by Ross and Vitale [26]: stabilisation, continuous the procurement manager did not understand how the
improvement and transformation. Activities observed master plan forecasts, which in theory feed the MRP,
for the stabilisation stage are typically operational were established.
optimisation as defined in situation 1 (cleaning up data In that case, we see that the real problem is not to
and parameters, resolving bugs in the software, improve the use of the ERP, nor of course the
providing additional training). During the continuous performance of the company. In fact, the ERP is used
improvement stage, firms focus on implementing in a degraded mode. Its full functionality as a tool is
adding functionality such as bar coding, EDI, sales not reached. The possibilities of integration through
automation . . . generating significant operating ben- the standard processes were gradually abandoned for
efits, which fit with situation 2. Finally, situation 3 the benefit of regulation modes based on the direct
corresponds to the transformation stage, which aims to relations between users. In this kind of situation, the
gain increased agility, organisational visibility and ERP is only used for its transactional functions of
customer responsiveness. The process of optimisation orders follow-up to guarantee their accounting
proposed agree with the taxonomy designed by Al- integration. The stabilisation phase has reduced the
Mashari et al. [8], which illustrates that ERP benefits initial perimeter, excluding all functions not mastered.
are realised when a tight link is established between Obtaining a full function ERP control requires
implementation approach and business process per- reviewing initial set-up for planning and control
formance measures. functions, incorrectly mastered and consequently not
Some audits, presented in Section 5, have been used. This entails first the correction of erroneous data
realised in several companies on the use of the MRP and then analysis of the causes of those MRP
modules within the ERP. They allow validation of the messages that users find abnormal.
currently proposed classification: optimisation stages
and situations. These elements represent the process of 5.2. ‘‘Tactical’’ improvement
improvements, which is the core of the proposed
guideline. This process has been checked against an 5.2.1. Company C2
industrial case, described in Section 5.4. Middle size industrial company. Components for
the construction industry. Batch production with
numerous options and variants. ERP project of type
5. Validation from return of experiments ‘‘Big-Bang’’ in 1997–1998. Audit carried out in 2000.
The MRP planned purchase orders are printed and
5.1. ‘‘Operational’’ improvement sent to purchase agents, but they are evaluated as
incomplete and often seized again. The cause
5.1.1. Company C1 determined by the operators was the absence of full
International group of electro-technical materials. pegging in the MRP. It should be noted however that
Numerous sites of production in the world with the company is typified by make-to-stock production
deployment of ERP between 1993 and 1997 always as with a lead-time two to three times superior to the
‘‘Big-Bang’’. average commercial delay, but still does not use a
An audit and a training plan were organised in 1999 master production schedule (MPS) nor a sale and
after group reorganisation for make-to-order divi- operation planning (SOP). Numerous reporting rea-
sions. The audit revealed that the procurement lised on spreadsheet from the ERP data exist and are
managers manually size every purchase order because used for planning and control.
the MRP and project requirement planning (PRP)
proposals were not reliable. Their decisions were more 5.2.2. Company C3
often based on a make-to-stock strategy and on Medium size industrial company. Parts manufac-
knowledge of the in-progress business orders. The turer for the automotive industry. Mass production
numerous exception messages and rescheduling project of type ‘‘Big-Bang’’ in 1996–1997. Audit
messages in/out from the MRP were not handled. carried out in 1999.
 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587 585

The MRP planned orders are used and generally up, . . .) and calculate an available-to-promise on the
confirmed by the procurement agents, but the critical resources justifying such groupings. The
production manager considers that the compromise existing ERP allows this feature in a later version
ability-to-react/productivity do not fit his objectives in but this functionality supposes a complete integration
the production program proposed by the MRP. He of the commercial function in the ERP.
underlines a capacity utilisation calculated by the In that case, the ERP functioning is quite optimised
ERP, which goes up and down, not realistic for in the perimeter initially defined, but the information
operational activities. Consequently, he is led to system strategy adapted to a relatively segmented
numerous manual modifications. organisation forbids developing the ERP towards a
In case C3, we see that the ERP ‘‘works’’ and complete and efficient integration of delivery time.
allows control of the company based on the More precisely, functions like available-to-promised
information system data. The existence of automated (ATP) or capable-to-promised (CTP) allowing defini-
indicators and reporting to support control decision- tion of a delivery time taking into account manu-
making differentiates this case from company C2. facturing constraints, actual inventories and work in
However, the ERP project did not allow progress progress, cannot be used without considerable re-
towards a complete integration of the planning engineering of the software and IT architecture.
functions. The causes lie in the difficulties in using A strategic choice is thus proposed to the company:
a standard MRPII approach (no MPS in both cases a first hypothesis is the pursuit of a ‘‘best of breed’’
while the study of total cycle makes it in theory approach (urbanisation) which will necessitate inte-
compulsory). gration tools like enterprise application integration
In these two cases, the improvement requires an (EAI) and an advanced planning system (APS)
analysis of the business processes to implement a software dedicated to the planning integration
planning method, which did not succeed in the initial between source, make and deliver processes. The
project. The difficulty did not concern the mastery of second hypothesis defines a wider perimeter for the
the tool itself but rather the implementation of a ERP by leading a project of ERP extension (using the
classic MRPII approach in the company. new version) integrating the commercial function (at
least for order management). Unlike the cases C2 and
5.3. ‘‘Strategic’’ improvement C3, the scope, architecture and aims with which the
ERP was implemented, can be redefined. The choice
5.3.1. Company C4 arises from a strategic decision. How the information
Furniture manufacturing company with strong system could best contribute to the strategy of the
image, make-to-order oriented. company and specially the type of customer relation-
The ERP project was positioned as a manufacturing ship.
project with interfaces to a commercial software
(totally specific solution internally developed) and a 5.4. Process improvement validation
financial management software (standard package).
The strategy is more oriented towards a ‘‘best of The classification proposed seems to be efficient in
breed’’ solution, but essentially for historical and characterising a specific case and to orientate actions
organisational reasons. The deployment was made of staff involved in ERP optimisation. One of the
gradually by activity between 1994 and 1999. All the qualitative interviews conducted in the survey
flows of goods were then managed in the ERP and represents a direct validation of this proposed process
strongly interfaced with the specific commercial with a wide experience in a large enterprise.
software. The pressure over both the commercial In this enterprise (nuclear components producer
deadline and the manufacturing efficiency require- and distributor), the ERP has been used since 1999
ment, lead the manufacturing manager to look for a after an important business BPR project. The project
better integration of commercial and industrial dead- leader was previously an operational manager and
lines. He worked to optimise the necessary groupings became the group IT Manager. He has conducted a
for manufacturing efficiency (lot size, machines set- large number of optimisation actions, which can be
586 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587

presented through key performance indicators. The Audits realised within certain companies allow
first generation of indicators were considered as validation of this approach.
‘‘strictness school’’ and were based on data validation Future trends seem oriented towards a repository of
(quality rules not defined, purchased items without ERP use and an optimisation guideline intended for
supplier data, items without lead-time, work-centres middle-sized companies. Identifying clear causes of
without capacity, manufactured items without cost malfunctioning as proposed in Section 4 could help to
price, . . .). These indicators are synthesised in an offer a benchmarking approach to ERP use. It
‘‘ERP quality indicator’’ measuring the rate of supposes to go into alert criteria definitions in depth
abnormal event per entity and type. Corrective actions with quantitative indicators. The next step could be to
are essentially training complements. The second evaluate corrective action priorities depending on the
generation were based on dynamic data and more characteristics of a specific situation. This would lead
oriented to measuring interaction between functions to a configurable method for defining a customised
(incorrect bill of material noticed in production order way of continuously improving ERP use.
picking, production order released without stock
components checked, . . .). The third generation is
evolving towards process measurement (expected on References
actual delivery date, expected on actual inventory
level, stock shortage per item groups, service level [1] T.H. Davenport, Putting the enterprise into the enterprise
defined as partial on complete delivery). Improvement system, Harvard Business Review (1998) 121–131.
[2] S. Buckhout, E. Frey, J. Nemec, Making ERP succeed: turning
actions are based on parameter control in relation to
fear into promise, Journal of Strategy and Business, Technol-
mean time objectives (waiting time for work-centre, ogy 15 (1999) 60–72 (Reprint No. 99208).
stock service level, . . .). At this stage, data control [3] H. Klaus, R. Rosemann, G. Gable, What is ERP? Information
allowed more than 50% of management scorecard to Systems Frontiers 2 (2000) 141–162.
be determined directly from the ERP. [4] V. Botta-Genoulaz, P.-A. Millet, G. Neubert, The role of
These three generations of improvement indicators enterprise modeling in ERP implementation, in: International
Conference on Industrial Engineering and Production Manage-
correspond well to the three stages defined previously. ment (IEPM’01), Proceedings Book I August 20–23, 2001,
Quebec, Canada, 2001, pp. 220–231.
[5] R. Poston, S. Grabski, Financial impacts of enterprise resource
6. Conclusions planning implementations, International Journal of Account-
ing Information Systems 2 (2001) 271–294.
[6] P. Rajagopal, An innovation—diffusion view of implementa-
The stakes in control of integrated systems cannot tion of enterprise resource planning (ERP) systems and devel-
be limited to the phases of implementation or opment of a research model, Information & Management 40
deployment. Better use of these information systems (2002) 87–114.
drives companies to new organisations and to [7] G. Saint Léger, G. Neubert, L. Pichot, Projets ERP: incidence
des spécificités des entreprises sur les Facteurs Clés de Succès,
continuous adaptation of company strategy. It should
in: Proceedings of AIM 2002, Hammamet, Tunisia, 30 Mai–1
help in the (re)evaluation of the positioning of the ERP Juin, 2002.
in the information system to identify relevant [8] M. Al-Mashari, A. Al-Mudimigh, M. Zairi, Enterprise
improvement actions in a given situation. resource planning: a taxonomy of critical factors, European
From practices noticed in companies and from Journal of Operational Research 146 (2003) 352–364.
results of various surveys, we propose a classification [9] M. Bradford, J. Florin, Examining the role of innovation
diffusion factors on the implementation success of enterprise
of ERP use characterised by three stages of resource planning systems, International Journal of Account-
optimisation described on two axes. This leads to a ing Information Systems 4 (2003) 205–225.
three-stage optimisation process: operational (the [10] V.-A. Mabert, A. Soni, M.-A. Venkataramanan, Enterprise
information system is considered as a production resource planning: managing the implementation process,
and data broadcasting ‘‘tool’’), tactical (control of the European Journal of Operational Research 146 (2003) 302–
314.
operational processes for a better integration between [11] P.-A. Millet, V. Botta-Genoulaz, Differences between service
functions) and strategic (contributing to company and manufacturing organizations regarding ERP, in: Interna-
strategy in response to changes and evolutions). tional Conference on Industrial Engineering Production
 V. Botta-Genoulaz, P.-A. Millet / Computers in Industry 56 (2005) 573–587 587

Management (IEPM’03), Proceedings Book I May 26–28, mentations in the US manufacturing sector, Omega 31 (2003)
2003, Porto, Portugal, 2003, pp. 358–365. 235–246.
[12] S. Sarker, A.S. Lee, Using a case study to test the role of three [24] V. Kumar, B. Maheshwari, U. Kumar, An investigation of
key social enablers in ERP implementation, Information & critical management issues in ERP implementation: empirical
Management 40 (2003) 813–829. evidence from Canadian organizations, Technovation 23
[13] E.J. Umble, R.R. Haft, M.M. Umble, Enterprise resource (2003) 793–807.
planning: implementation procedures and critical success fac- [25] J. Olhager, E. Selldin, Enterprise resource planning survey of
tors, European Journal of Operational Research 146 (2003) Swedish manufacturing firms, European Journal of Opera-
241–257. tional Research 146 (2003) 365–373.
[14] J. Motwania, D. Mirchandani, M. Madan, A. Gunasekaran, [26] J.W. Ross, M.R. Vitale, The ERP revolution: surviving vs.
Successful implementation of ERP projects: evidence from thriving, Information Systems Frontiers 2 (2) (2000) 233–241.
two case studies, International Journal of Production Econom-
ics 75 (2002) 83–96.
[15] P. Kræmmergaard, J. Rose, Managerial competences for ERP Valérie Botta-Genoulaz is Assistant
journeys, Information Systems Frontiers 4 (2) (2002) 199–211. Professor in the Industrial Engineering
[16] H.H. Chang, A model of computerization of manufacturing Department of the National Institute of
systems: an international study, Information & Management Applied Sciences (INSA) of Lyon,
39 (2002) 605–624. France. Her main courses deal with pro-
[17] T.M. Somers, K.G. Nelson, A taxonomy of players and duction management and enterprise
activities across the ERP project life cycle, Information & information systems. 5 years experience
Management 41 (2004) 257–278. in industry, PhD in Computer Sciences,
[18] J.E. Hunton, B. Lippincott, J.L. Reck, Enterprise resource application consultant ‘Production Plan-
planning systems: comparing firm performance of adopters ning’ for SAP R/3, her research interests
and non-adopters, International Journal of Accounting Infor- are oriented on planning and management of supply chains (SCM),
mation Systems 4 (2003) 165–184. ERP and SCM project management (including process modeling
[19] F. Calisir, F. Calisir, The relation of interface usability char- and organisational dimensions).
acteristics, perceived usefulness, and perceived ease of use to
end-user satisfaction with enterprise resource planning (ERP)
systems, Computers in Human Behavior 20 (4) (2004) 505– Pierre-Alain Millet is Associate Profes-
515, Available online 30 October 2003. sor in the Information Technology
[20] R. Canonne, J.-L. Damret, Résultats d’une enquête sur Department of the National Institute of
l’implantation et l’utilisation des ERP en France, Revue Applied Sciences (INSA) of Lyon, France
Française de Gestion Industrielle 21 (2002) 29–36. where his main courses concerned infor-
[21] E. Labruyere, P. Sebben, M. Versini, L’ERP a-t-il tenu ses mation system and ERP. 20 years experi-
promesses? Deloitte & Touche report, 45 p., Juin 2002. ence in industry and services, ERP
[22] C. Moulin, Enquête ERP 2002: les principaux résultats, 21ème consultant since 1991 on SSA solutions
Journée régionale de la Productique, Pôle Productique Rhône- (BAAN), his research interests are
Alpes, France, 2002. focused on integration in ERP and
[23] V.-A. Mabert, A. Soni, M.-A. Venkataramanan, The impact of SCM systems, enterprise modeling and organisational dimension
organization size on enterprise resource planning (ERP) imple- of IT.