Comprehensive View on Architectural Requirements for

Maintenance Information Systems

Andreas Reidt

, Stefan Schuhbäck

and Helmut Krcmar

fortiss GmbH, Guerickestraße 25, Munich, Germany

Chair for Information Systems, Technical University of Munich, Munich, Germany

Keywords: Maintenance Information System, Digitization, Industry 4.0, Industrial Internet, Software Architecture,

Mobile Support System, Predictive Maintenance, Condition Monitoring, Requirement Engineering, Technical

Customer Service.

Abstract: Concepts of Industry 4.0 and digitization are leading to a much higher complexity of production facilities,

related processes, and activities such as maintenance. As a result, the demands made on the process of

maintenance and the maintenance workers involved are getting more demanding. To meet these requirements,

many information systems have been developed with different purposes and technologies. Yet the

development of these systems is occurring in isolation and the main challenges of the integration of these

maintenance solutions and their information with each other are not addressed. The resulting solutions and

their architectures lack a sustainable holistic viewpoint from the start. To solve these challenges and to give

developers a framework in which to develop information systems for maintenance, a holistic view of the

architectural and requirements is needed. Therefore, a framework for the general requirements of all

maintenance information systems has been developed in this paper. To achieve this, a literature review was

conducted where the requirements for a broad set of maintenance information systems were gathered and

compared with each other. Based on this information, general principles for the architectures of maintenance

information systems were derived.

1 INTRODUCTION

Progressive digitization is not only leading to

completely new enterprises, but also poses major

challenges for traditional, established companies as

they are also subject to fundamental change (Horváth,

2017). Trends and technologies like IoT (Gubbi et al.,

2013), CPS (Lee, Bagheri and Kao, 2015), Industry

4.0 (Lasi et al., 2014), and respectively the Industrial

internet (Lin et al., 2015) are driving this change to

lead to major challenges, especially for

manufacturing companies.

On the one hand, the existing processes are

subject to a strong and purely technological change.

On the other, changes to the company's current

business models are not only made possible by

current technologies but they are also necessary, so

that companies must prepare for a disruptive change

(Reidt, Duchon and Krcmar, 2017). Producing

companies often turn into producing service

providers (Daeuble et al., 2015).

Linked to these changes are processes such as

maintenance for production plants, which struggle to

cope with this new resulting environment. This

introduces additional complexities such as changes to

the business models, more complex machinery

throughout the maintenance industry and higher

requirements for the maintenance workforce.

To compensate for these effects, a large variety of

information systems (ISs) exist within the industry to

support maintenance operations and tasks. Besides

such traditional systems as Computerized

Maintenance Management Systems (CMMS)

(Gabbar et al., 2003) that are becoming enhanced

with additional features, completely new support

systems with state-of-the-art technologies such as

mobile support systems (Fellmann et al., 2013), or

systems that use augmented reality (Zhu, Ong and

Nee, 2014) are being introduced to the industry.

The main issue with the current development is

that existing systems have been developed and are

being operated in an isolated manner, and that they

form isolated data islands (Galar, 2014). The

Reidt, A., Schuhbäck, S. and Krcmar, H.

Comprehensive View on Architectural Requirements for Maintenance Information Systems.

DOI: 10.5220/0006698602490257

In Proceedings of the 13th International Conference on Evaluation of Novel Approaches to Software Engineering (ENASE 2018), pages 249-257

ISBN: 978-989-758-300-1

249

information exchange between systems is scarce and

if present often complicated or not completely

automated. This is reinforced because many of these

systems have a specific use case, which is

implemented in isolation from the Big Picture that

current maintenance operation needs. Moreover,

existing systems and requirements for holistic

maintenance systems are not known and their

development is costly. Furthermore, such a

development requires an interdisciplinary view,

which combines knowledge from various fields such

as computer science, maintenance, business and

industrial automation.

To approach these challenges an architectural

requirement framework is proposed in this paper that

emphasizes requirements for maintenance ISs. The

framework will show requirements within different

ISs and pin-point similar and holistic requirements

that must be considered in the development process

of software architectures for maintenance related IS.

Furthermore, relevant ISs in a production

environment and support systems for maintenance are

presented that can fulfil these requirements or their

integration are needed in order to extract relevant

information.

Together, the systems and the holistic

requirements form the cornerstone for a framework

that can be the basis for the development of

sustainable and holistic software architectures for

maintenance ISs.

In the following chapter, the existing IS for

production which have a supporting role to

maintenance activities as well as dedicated IS with

the sole purposes to aid maintenance activities on all

levels are presented. Chapter 3 describes the used

method of identifying requirements of ISs and the

subsequent clustering. Lastly, a framework is

generated and the corresponding outcome is

discussed. Furthermore, research outlooks in the

context of creating additional reference architectures

for maintenance ISs are presented.

2 IS FOR MAINTENANCE

The classification of the most relevant ISs for

maintenance requires a definition of the framework

for the concept of maintenance. In this paper, the

concept of maintenance can be understood as the

“combination of all technical, administrative and

managerial actions during the life cycle of an item

intended to retain it in, or restore it to, a state in which

it can perform the required function” (Deutsches

Institut für Normung, 2015).

Maintenance, however, comes in a variety of

types, concepts and strategies, which have been

developed over the last few decades. These influence

the selection and characteristics of maintenance

systems. Therefore, concepts, types and maintenance

strategies relevant to the information systems are

presented below. Subsequently, ISs are described,

from which information is needed in order to carry

out maintenance tasks. These systems form the core

of the IT architecture for most manufacturing

companies. Then, an overview of current systems that

are designed to support maintenance and their terms

in literature and practice is provided.

2.1 Maintenance Types, Concepts and

Definitions

Apart from different maintenance measures, the

literature distinguishes maintenance strategies or

types and concepts. The two main strategies or types

according to Niu et al. (2010) are reactive and

preventive maintenance. Reactive maintenance

describes actions after a unit failure occurs whereas

preventive measures are taken beforehand to mitigate

failures. The preventive measures can be divided into

subcategories: periodical and condition based

maintenance.

Figure 1: Maintenance strategies based on Niu et al. (2010).

With periodical maintenance, tasks are planned

with fixed intervals, regardless of the current state of

the machine. The intervals can be time-driven or

based on other variables such as mileage.

The condition-based approach will schedule

maintenance measures based on the current condition

or on a predicted future condition calculated, based

on statistical models and data mining. This distinction

between predictive maintenance and condition

monitoring is essential in order to distinguish newer

ENASE 2018 - 13th International Conference on Evaluation of Novel Approaches to Software Engineering

250

Figure 2: Overview of relevant systems for maintenance, divided into production systems and support systems.

ISs for maintenance. The entire representation of the

subdivision of maintenance strategies or types is

shown in Figure 1. Within manufacturing industry, all

approaches mentioned are used to some degree and

introduce specific requirements for supporting IS.

Reactive maintenance will require the IS to assist

with failure analysis and repair whereas preventive

maintenance requires planning and scheduling of

maintenance tasks, according to the current or future

state of the machine.

These concepts and types have a significant

impact on the required features of maintenance ISs

and the necessary information a maintenance worker

has to extract.

2.2 Production Systems

In industrial applications, maintenance information is

needed from a variety of ISs to efficiently perform

maintenance tasks. These ISs supply production and

maintenance with information and they are located at

different levels of the automation pyramid

(Bauernhansl, 2014). In the upper section of Figure 2,

a list of the identified most popular ISs relating to

maintenance tasks is presented.

Product Data Management (PDM) systems

manage the data used within the development process

and provide access to this data across every phase of

the product lifecycle. Connected with these kind of

systems are Document Management Systems

(DMSs) which are often used to manage physical

and/or digital heterogeneous documents within

companies. Both systems can be seen as parts of a

generic Knowledge Management Systems (KMSs)

enable the gathering and display of context-

dependent, explicit and implicit knowledge or

information of an organization (Maier, 2007).

Spare Part Management (SPM) systems contain

information on spare parts, including pictorial

representations, 3D models and information on bills

of materials (BOMs).

Enterprise Resource Planning (ERP) systems

support all business processes while Manufacturing

Execution Systems (MESs) are tailored to technical

production processes.

Customer Relationship Management (CRM)

systems are closely connected to ERP systems with

the goal of increasing shareholder value by managing

customer data and steering business processes to good

customer relationships.

Advanced Planning Systems (APSs) are used to

manage material, personnel, tooling and production

plants across national borders. Many tasks within

APSs can be found in part in the systems described

above. The focus of APSs is mainly on the

mathematical optimization of these complex

problems.

Programmable Logic Controllers (PLCs) directly

control the production process on the lower level of

the automation pyramid within a machine or group of

machines. The real-time data generated is used within

Supervisory Control and Data Acquisition (SCADA)

systems to directly manage the production process.

PLC and SCADA systems contain important

information for real-time condition monitoring.

2.3 Support Systems

In contrast to the ISs mentioned in the previous

section, the following ISs have been identified as

systems dedicated to supporting maintenance

activities directly rather than just serving as an

information source. The ISs are presented in the lower

section of Figure 2.

CMMSs are among the first systems used to

support maintenance tasks. CMMSs are used to

provide support mainly in maintenance planning and

controlling tasks.

Comprehensive View on Architectural Requirements for Maintenance Information Systems

251

Mobile Support Systems (MSSs) assist

maintenance workers on site in remote locations by

providing, for example, billing data or manufacturer

documents to a mobile device. MSSs are mainly used

as data aggregators or presentation devices and for

dispatching tasks to maintenance workers.

Condition Monitoring Systems (CMSs) have the

task of continuously monitoring individual machines

or complete systems. If the measured state has a large

deviation from the theoretical state, the CMS will

generate an appropriate message to inform (for

example) the operator.

E-Maintenance (E-MS) is a term that emerged in

the early 2000s (Iung et al., 2009), linked to the

increasing use of ICT in general and by the Internet

in particular (Li et al., 2005). The term “E-MS

systems” is used to describe a variety of diverse

systems with modern technologies, ranging from

remote maintenance systems (Iung, 2003) to CMMS

with enhanced web technology (Hausladen and

Bechheim, 2004).

Intelligent Maintenance Systems (IMSs) were

developed by (Lee et al., 2006) at the same time as

E-MS. (Ling Wang et al., 2006) defines the IMS core

paradigm of allowing error prediction by means of

sensor analysis and prediction of performance

degradation of a component. With this definition, the

IMS concept is strongly linked to predictive

maintenance and CMS.

The focus of Predictive Maintenance Systems

(PMSs) is similar to that of IMSs but with more

emphasis on error prediction and avoidance, based on

algorithms and data mining techniques. The term and

linked technology are somehow the trending topics

for maintenance IS due to the rise of data mining and

artificial intelligence. Another reason for this

popularity is the increase in big cloud providers who

offer services especially for predictive maintenance.

Decision Support Systems (DSSs) emphasize the

aspect of data analysis and presentation in order to

facilitate the analysis of the current situation by

means of modern algorithms, allowing decision

makers to make informed decisions.

2.4 Architectural Considerations when

Implementing New Maintenance

Systems

The systems presented here outline the challenges

that arise in the development of holistic ISs to support

maintenance. Information from a large number of

systems is required, which is why the architecture of

an IS has to deal primarily with data integration and

interfaces. Furthermore, the number of support

systems shows the many different facets in which

maintenance can be supported. However, the

distinction and inclusion of the relevant requirements,

which lead to the respective systems, is often very

difficult. The development of sustainable

architectures also requires knowledge of the

properties and requirements of the systems and thus

of the overlapping core.

So far, however, no papers are known to us that

relate these systems to one another and compare their

characteristics and requirements. Only overviews of

the requirements for MSSs can be found by

(Matijacic et al., 2013). For this reason, relevant

requirements are subsequently derived, based on

these systems.

3 GATHERING REQUIREMENTS

FOR MAINTENANCE

INFORMATION SYSTEMS

A comprehensive perspective on maintenance ISs is

needed to evaluate whether a requirement can be

considered holistic. This holistic view has to

aggregate all the aforementioned systems to

distinguish special and common requirements and it

needs to identify the most important ones for future

maintenance ISs.

To establish this holistic view, requirements for

individual ISs are gathered in this paper via a

comprehensive literature review that will establish a

holistic and interdisciplinary view of maintenance

ISs.

3.1 Literature Review Approach

The literature review was conducted in 2017 via a

comprehensive database search. The aim of the

literature review was not to show that the literature

search found every relevant article. Instead, the

articles found needed to demonstrate a broad basis of

the diverse systems as viewed from a variety of

research directions, in order to achieve an

interdisciplinary view on requirements. For this

reason, Google Scholar was used as a search

engine/database. Thus, not only could very special

and exact search terms be used, but also very far-

reaching ones.

Keywords were extracted by the above systems,

maintenance concepts, strategies, methods, and

technologies used. These keywords were combined

with the terms ‘system’, ‘requirement’ and/or

‘maintenance’ to form a search string that certainly

had a relation with these terms. If the system names

ENASE 2018 - 13th International Conference on Evaluation of Novel Approaches to Software Engineering

252

varied between German and English, additional

German search terms were employed. A combined

total of 43 search terms were used. Only articles from

2005 to the present were examined.

All the articles resulting for each search term were

investigated unless there were more than 80 results

per search term. In these cases, only the top 80 were

investigated because the relation to the search term

and the relevance decreased with each result.

The results were examined to determine whether

they contained requirements for ISs for maintenance

in general or for a specific IS. These requirements can

be explicit or implicit. In the case of explicit

requirements, these are named exactly. For example,

either by presenting requirements for a system or by

a literature overview of requirements for a system or

framework. Implicit requirements, on the other hand,

can be reliably derived from descriptions of systems

or frameworks. They are not mentioned directly, but

they can certainly be deduced from the functions or

descriptions in the articles.

The results of the respective search terms were

first classified according to the title, keywords, and

abstract only if they were considered relevant. During

this process, 405 relevant papers were identified. The

papers were then further examined and the entire

contents of the paper were analysed. In this step, the

requirements were also extracted. In total, 61 relevant

papers were investigated that dealt with requirements

for maintenance ISs or requirements could be

derived.

In order to avoid a situation where requirements

for a particular prototype, which was treated in

several of the papers found, were included multiple

times in the analysis, these requirements were

normalized. Papers dealing with the same specific

prototype therefore count as one publication and all

the extracted requirements from the individual

publications were summarized in this publication.

However, they only count once.

After removing these papers, a total of 56 articles

remained for inclusion in the requirements analysis

for a holistic maintenance IS. The results are

summarized in Table 1.

Table 1: Summarized results of literature search.

Search strings 43

Overall

esults 3115

Relevant results 405

ers use

Papers used without

duplicates

3.2 Resulting Requirements

Several steps were taken to gather the requirements

from the papers:

 Extracting the explicit and or implicit

requirements in original form

 In the next step, not mentioned requirements that

are prerequisites by already extracted

requirements were added as if they belonged to

the article

 Similar requirements were unified and

standardized. Afterwards, the requirements were

clustered into topics

In total, 135 different requirements were

identified in all the articles. In summary, the 135

requirements were identified 751 times in all articles.

These numbers show that requirements occur

multiple times across different publications in

relation to different systems and domains.

Table 2: Summarized results of literature search.

Condition Monitorin

Order Mana

ement 27

Error/Failure Histor

Overview of plants including specific plant

related information

Remote maintenance with a focus on remote

monitoring

Information on measures and tooling history of

lants

Knowledge management 21

Condition history of parts

lants 19

Spare parts management (storage location,

availability, costs)

Predictive maintenance 18

Provide maintenance plans/schedules for plants 16

Performance assessment and verification of the

lants

Document management (storage, provision,

sharin

of documents

)

Communication and contact with other persons

(

e-mail, messa

es, tele

hone

)

Optimization of maintenance with respect to

orders, inspections etc.

Component overview of plants 13

Presentation of the state of the plant on

dashboards

Remote access (changing settings, controlling

com

onents

)

Intelligent planning and scheduling of

ees

Spare parts, material and possible tool lists per

orde

Display of an overview of internal and external

erts for maintenance

Error message with detailed description 12

Comprehensive View on Architectural Requirements for Maintenance Information Systems

253

Table 2 summarizes all requirements found in the

literature that occur at least 12 times. As a result, the

requirements are seen in this article as top

requirements. The names of the requirements have

been partially truncated or aggregated.

We can see that the “Condition Monitoring”

requirement or the requirement block is stated in

more than half of all publications. Despite the fact

that only 13 publications dealt with condition

monitoring in detail, it can be said that condition

monitoring is important for a broad range of

maintenance ISs.

Order management and failure history are two

requirements that are also broadly represented in the

publications. An order management system, which

manages incoming orders and tasks while monitoring

ongoing ones, is essential for most ISs for

maintenance if they want to be productive. In many

cases, order management was performed by external

systems that were not specially stated.

Supervision and monitoring were other important

requirements. An overview of plants with information

on spare parts, functionality and location was a top

requirement. Linked to this, the ability to monitor the

plant via remote maintenance is an essential

requirement for many publications. Yet there is a

difference between remote maintenance with a focus

on monitoring and remote maintenance enabling

direct access to the plant. This requirement was also

a top requirement, but had only a value of 12.

Other important requirements with more than 17

occurrences were the provision of information on

measures, tools and the history in general of a plant,

the requirement for knowledge management

(especially for knowledge of how to repair certain

plants) and general knowledge for maintenance

personnel. Another important requirement was the

need to save and maintain access to the condition

history of the plant and machinery. These are

especially important for analysis and failure analysis.

Connected with this requirement is the requirement

for predictive maintenance, counted 18 times, which

in fact was in most cases separated from condition

monitoring as predictive maintenance is a forecast

and data mining techniques are used sometimes.

Spare part management was also a very important

issue, which should be handled via IS. This

requirement includes the integration of spare part

ordering and its connection to failures or orders, so

that maintenance personnel knows beforehand which

spare parts are needed for a job. Additionally, the

need to automatically mark used spare parts was

mentioned in this requirement.

These requirements, only 22 in all, have a total

count of 388 of 751 of all requirements and cover

almost ~52% of all mentions. Thus, one can say that

some of these requirements or parts of them are the

basis of nearly all the systems investigated.

Considering the impact these requirements can

have on the architectural consideration of a holistic

maintenance information system, the need arise to

integrate them into planning as early as possible. A

framework will be presented in the next section to do

this and to integrate the requirements not explicitly

mentioned.

4 FRAMEWOK FOR DEDUCING

THE ARCHITECTURAL

REQUIREMENTS

The top requirements, like the rest of the 135

requirements, can be classified into specific

categories.

These categories, in addition to the top requirements,

are designed to facilitate architectural decisions on

new ISs for maintenance. The relevant requirements

are known and can be classified more easily through

the categories and the top requirements. Furthermore,

software architectures and experts can compare their

own requirements with those in the framework,

identify their own and recognize interactions. Besides

this, the architecture and the development of the

categories can assume an early form of

modularization. Interactions on a higher level can

also be investigated.

Figure 3: Clustered categories for requirements.

All the requirements were initially grouped into

diverse categories. This classification was refined

systematically through an iterative process. Finally,

Technical Customer

Service

Remote Maintenance

Ordermanagement &

Information

Service Management

Maintenance Planning

and Optimization

Plant Overview and

Information

Spare Parts

Management

Employee

Management

Communication

Mobile System

(Historical)

Evaluations

Fault Management

CM and PM

Document

Management

Knowledge

Management

Recommendations/

Guidance

External Services

System

Administration

ENASE 2018 - 13th International Conference on Evaluation of Novel Approaches to Software Engineering

254

the following categories were identified in which the

requirements for a maintenance IS can be assigned.

The categories are shown in Figure 3. They are

displayed, from the top left to the bottom right,

according to the sum of the mentions of the individual

requirements and the number of top requirements.

Grey blocks contain top requests, while white blocks

do not.

The individual categories are presented below

with reference to exemplary requirements:

 (Historical) Evaluations: The category with the

highest number of mentions of requirements is

the (historical) evaluation. Four of the top

requirements are in this block that deals with

failure, order and condition histories, as well as

with the evaluation of performance connected

with these aspects.

 Plant Overview and Information: This category

contains all requirements that provide

information about plants. This includes

requirements such as an overview of all plants

and their properties, as well as a component

overview for every plant. In addition, technical

documents, such as manuals, for example,

should be directly accessible.

 Order Management & Information: Order

management for incoming orders and the

management of tasks is necessary in a broad

variety of systems. In this category,

requirements for order management and the

connected information about the order can be

found. Furthermore, the section includes

employee-oriented features, such as an

individual overview of own tasks and the

possibility to accept and reject tasks. It also

includes aspects such as connecting an order

with the required skills, and the ability to

prioritize orders. Considering architectural

implications, this category comprises a

managing core for most maintainer-oriented

systems because most information from other

categories have to be connected with the

corresponding order.

 Maintenance Planning and Optimization: In

this category, features for optimization can

mostly be found. The optimization of the order

management or an intelligent disposition of the

workforce considering their locations, skills

and failure priorities are examples. In addition,

requirements for a risk classification or

integrated cost management are also assigned

to this category.

 CM and PM: In this category, requirements that

deal with condition monitoring and predictive

maintenance can be found. On the one hand,

the most frequently mentioned requirement,

namely to determine the condition of a

machine, is present in this category, together

with further requirements for the visualization

of this state. On the other hand, there are

requirements in this category that are related to

the prediction of errors and the use of data-

mining techniques.

 Knowledge Management: Requirements for

knowledge management can be found in this

category. The most frequently cited

requirement in this category is the knowledge

management as an aggregation. This

requirement is specialized in other

requirements, which mainly deal with the

storage, restoration and linking of knowledge

with the concrete task.

 Spare parts management: This category of

requirements is all about the management of

spare parts. This includes the display of

required spare parts, their availability, costs

and storage location. Further requirements for

the (automatic) ordering, forecasting and links

to current orders are among other things

grouped in this category.

 Remote Maintenance: The category of remote

maintenance covers primarily the read access

and thus contains remote maintenance as a top

requirement. In addition, the requirements

dealing with writing access and remote access

and control for plants are illustrated here. There

are also special requirements for the audio and

video transmission of the machine data.

 Document management: The document

management category includes requirements

for managing documents, for the partially

automated documentation of completed orders,

and for fill-in assistants for specific forms and

reports.



Fault management: Failure management

includes requirements for the fault messages of

the systems and the system. Above all, a

detailed error message is required. Moreover, a

fault report directly sent to the device used by

the responsible person or error messages with

cause-effect relationships and the possibility to

prioritize errors and report them manually are

requirements in this category.

 Employee Management: Requirements in this

category are used to manage and overview

employees in maintenance. This includes the

provision of an overview of the internal and

external competency providers with an

Comprehensive View on Architectural Requirements for Maintenance Information Systems

255

availability schedule, a potential system for

measuring working time, provision of skill

profiles or semi-automatic travel cost

recording.

 Communication: Communication

requirements include the ability to

communicate with other employees and

competencies for example via e-mail, SMS, or

telephone. Additionally, message functionality,

Web 2.0 capabilities for commenting solutions

and feedback functions are generally

mentioned as requirements in this category.

 Service Management: Maintenance

management includes only three requirements.

In particular, the provision of maintenance

plans for the systems and their components, the

possibility to manually define maintenance

limits and to compile maintenance

documentation.

The next two categories were unique to some

special kind of systems and have no top requirement:

 Technical Customer Service: In this section the

only requirement that is necessary for IS which

is used to manage a technical customer service

and the interaction with external customers. It

contains the customer management,

connections with a CRM, the display of

customer information for the maintenance

personnel on site and several interfaces for

reporting faults.

 Mobile System: This category contains

requirements for MSS or IS that support mobile

devices. It is an optional block of requirements

that only contains specific mobile features such

as barcode scanning, offline mode, photo

features or a wireless diagnosis with a mobile

device.

 Recommendations/Guidance: The

recommendations/guidance block mainly

includes the feature to provide

recommendations for specific errors,

maintenance, or error diagnoses, and they are

automatically attached to the job/order.

The last two blocks consist of a few requirements

that deal with specific system administration

requirements and specific integrations of external

services. All of the requirements have a low value and

seem not very important for most authors. Yet the

functions seem to be necessary for the most

productive systems.

5 CONCLUSION

Contrary to the assumption that the many existing

systems for maintenance are very different, the result

of this article shows that maintenance systems have a

generic core. This should be considered in future

developments and the framework presented can

support this.

The framework consists of 18 categories, which

together contain 135 requirements. Of these 135

requirements, 22 requirements are defined as top

requirements covering 52% of all claims of

requirements. In addition to the requirements in the

framework, the most important potentially existing

systems for the maintenance activity were presented

in this article. A distinction was made between

systems from which data is primarily extracted and

systems that directly support maintenance.

By providing this information, the framework helps

to simplify the development and the selection of

maintenance ISs of various types.

Through the structured overview of a holistic

view on requirements for a maintenance IS impacts

and future developments can be better assessed.

Further, architecture requirements can be compared

and derived. Building on this knowledge of existing

ISs combined with the overview of requirements, the

selection and development of ISs for maintenance can

be facilitated by early interface considerations.

The framework offers an abstract basis for

designing the architecture and deriving architectural

components from the categories and examining

existing architectures for the extension with regard to

different requirements.

The various requirements and their

categorizations also reveal the problem of the

fragmentation of the individual services for

maintenance, since data must be obtained from a

variety of individual systems and services to form a

holistic entity.

In the future, a reference architecture for holistic

maintenance systems needs to be developed based on

these findings, which will not only enable the core of

a maintenance system to be displayed technically. It

should also enable various existing services and

systems to be combined in such a way that synergies

are created.

ACKNOWLEDGEMENTS

Thanks are due to the Bundesministerium für Bildung

und Forschung (BMBF) for financial support of this

ENASE 2018 - 13th International Conference on Evaluation of Novel Approaches to Software Engineering

256

work within the project PRODISYS (FKZ

02K16C050).

REFERENCES

Bauernhansl, T. (2014) Die Vierte Industrielle Revolution

– Der Weg in ein wertschaffendes

Produktionsparadigma, in Bauernhansl, T., ten

Hompel, M., and Vogel-Heuser, B. (eds) Industrie 4.0

in Produktion, Automatisierung und Logistik.

Wiesbaden: Springer Vieweg, pp. 5–36.

Daeuble, G. et al. (2015) Information Needs of the Mobile

Technical Customer Service -- A Case Study in the

Field of Machinery and Plant Engineering, in 2015 48th

Hawaii International Conference on System Sciences.

Kauai, HI, USA: IEEE, pp. 1018–1027.

DIN 13306:2015-09 (2015) Instandhaltung - Begriffe der

Instandhaltung; Deutsche und Englische Fassung prEN

13306:2015 (Entwurf). Edited by DIN Deutsches

Institut für Normung e.V. Berlin: Beuth Verlag.

Fellmann, M. et al. (2013) Towards a Mobile Technical

Customer Service Support Platform, in Daniel, F.,

Papadopoulos, G. A., and Thiran, P. (eds) Mobile Web

Information Systems. Berlin Heidelberg: Springer, pp.

296–299.

Gabbar, H. a. et al. (2003) Computer-aided RCM-based

plant maintenance management system, Robotics and

Computer-Integrated Manufacturing, 19(5), pp. 449–

458.

Galar, D. (2014) CONTEXT DRIVEN MAINTENANCE:

An eMAINTENANCE APPROACH, Management

Systems in Production Engineering, 3(15), pp. 112–

120.

Gubbi, J. et al. (2013) Internet of Things (IoT): A vision,

architectural elements, and future directions, Future

Generation Computer Systems, 29(7), pp. 1645–1660.

Hausladen, I. and Bechheim, C. (2004) E-maintenance

platform as a basis for business process integration, in

2nd IEEE International Conference on Industrial

Informatics, 2004. INDIN ’04. Berlin, Germany: IEEE,

pp. 46–51.

Horváth, P. (2017) Geschäftsmodellinnovationen durch

Digitalisierung – Neue Herausforderungen an den

Controller, in Technologie, Strategie und Organisation.

Wiesbaden: Springer Fachmedien Wiesbaden, pp. 113–

125.

Iung, B. (2003) From remote maintenance to MAS-based e-

maintenance of an industrial process, Journal of

Intelligent Manufacturing, 14(1), pp. 59–82.

Iung, B. et al. (2009) Conceptual framework for e-

Maintenance: Illustration by e-Maintenance

technologies and platforms, Annual Reviews in Control,

33(2), pp. 220–229.

Lasi, H. et al. (2014) Industry 4.0, Business & Information

Systems Engineering, 6(4), pp. 239–242.

Lee, J. et al. (2006) Intelligent prognostics tools and e-

maintenance, Computers in Industry, 57(6), pp. 476–

489.

Lee, J., Bagheri, B. and Kao, H.-A. (2015) A Cyber-

Physical Systems architecture for Industry 4.0-based

manufacturing systems, Manufacturing Letters. Society

of Manufacturing Engineers (SME), 3(January), pp.

18–23.

Li, Y. et al. (2005) An Agent-based Platform for Web-

enabled Equipment Predictive Maintenance, in

Proceedings of the 2005 IEEE/WIC/ACM International

Conference on Intelligent Agent Technology (IAT’05).

Compiègne, France: IEEE, pp. 132–135.

Lin, S.-W. et al. (2015) Industrial Internet Reference

Architecture, Technical Report. tr. 001. Industrial

Internet Consortium. Available at:

http://www.iiconsortium.org/IIRA.htm (Accessed: 3

September 2015).

Ling Wang et al. (2006) Advanced Maintenance Strategy

for Power Plants - Introducing Intelligent Maintenance

System, in Proceedings of the 6th World Congress on

Intelligent Control and Automation. Dalian, China:

IEEE, pp. 7444–7448.

Maier, R. (2007) Knowledge Management Systems. 3.

Berlin, Heidelberg: Springer.

Matijacic, M. et al. (2013) Elicitation and Consolidation of

Requirements for Mobile Technical Customer Services

Support Systems, in Proceedings of the 34th

International Conference on Information Systems

(ICIS), Milan (Italy), December 15-18, 2013. Milan,

Italy, pp. 1–16.

Niu, G., Yang, B. S. and Pecht, M. (2010) Development of

an optimized condition-based maintenance system by

data fusion and reliability-centered maintenance,

Reliability Engineering and System Safety. Elsevier,

95(7), pp. 786–796.

Reidt, A., Duchon, M. and Krcmar, H. (2017) Erstellung

einer Referenzarchitektur anhand von individuellen

Unternehmensanforderungen, in Bullinger-Hoffman,

A. C. (ed.) Abschlussveröffentlichung S-CPS:

Ressourcen-Cockpit für Sozio-Cyber-Physische

Systeme. Chemnitz: Verlag aw&I - Wissenschaft und

Praxis, pp. 23–42.

Zhu, J., Ong, S. K. and Nee, A. Y. C. (2014) A context-

aware augmented reality system to assist the

maintenance operators, International Journal on

Interactive Design and Manufacturing, 8(4), pp. 293–

304.

Comprehensive View on Architectural Requirements for Maintenance Information Systems

257