Smart Shop-floor Monitoring via Manufacturing Blueprints and

Complex-event Processing

Michalis Pingos

, Amal Elgammal

, Indika Kumara

, Panayiotis Christodoulou

Mike P. Papazoglou

and Andreas S. Andreou

Department of Computer Engineering and Informatics, Cyprus University of Technology, Limassol, Cyprus

Department of Computers & Information, Cairo University, Egypt

Department of Economics and Management, Tilburg University, Netherlands

Keywords: Product-Service Systems, Smart Product Customization, Smart Shop-floor Monitoring, M2M, H2M.

Abstract: Nowadays, Product-Service-Systems (PSS) are being modernized into smart connected products that target

to transform the industrial scenery and unlock unique prospects. This concept enforces a new technological

heap and lifecycle models to support smart connected products. The intelligence that smart, connected

products embed paves the way for more sophisticated data gathering and analytics capabilities ushering in

tandem a new era of smarter supply and production chains, smarter production processes, and even end-to-

end connected manufacturing ecosystems. The main contribution of this paper is a smart shop-floor

monitoring framework and underpinning technological solutions, which enables the proactive identification

and resolution of shop-floor distributions. The proposed monitoring framework is based on the synergy

between the novel concept of Manufacturing Blueprints and Complex Event Processing (CEP) technologies,

while it encompasses a middleware layer that enables loose coupling and adaptation in practice. The

framework provides the basis for actionable PSS and production “intelligence” and facilitates a shift toward

more fact-based manufacturing decisions. Implementation and validation of the proposed framework is

performed through a real-world case study which demonstrates its applicability, and assesses the usability and

efficiency of the proposed solutions.

1 INTRODUCTION

Nowadays, the trend in manufacturing calls for

increased connectivity and more sophisticated data-

gathering and analytics capabilities empowered by

Cyber Physical Systems (CPS), big data technologies

and the Internet of Things (IoT). These usher in

tandem a new era of smarter supply and production

chains, smarter production processes, and even end-

to-end connected manufacturing ecosystems.

Manufacturing is trying to create a competitive

advantage by not only offering products but

accompanying them with services (Product-as-a-

Service). Product-as-a-Service starts by sensor-based

products that generate data in a continuous manner;

these data can be utilized for delivering preventive

and proactive maintenance. Product-as-a-Service

often called Product/Service Systems (Bustinza et al.,

2015).

However, the current state of practice of

engineering PSSs still suffers from severe drawbacks

(Elgammal et al., 2017; Papazoglou and Elgammal,

2017; Song, 2017). The most noticeable drawback is

that PSS remains at conceptual level considering a

marketing or business perspective and missing solid

IT implementation. There is also complete lack of a

common factory level vision to empower data

sharing, monitoring and cross-correlation. In

addition, PSS do not accommodate evolving user

preferences or product differentiation features to

enable effective customization. PSS are unable to

capture a full view of products and services linking

product structure with product quality, production

processes and services. More importantly, they do not

support analysis of product-related data gathered

along product lifecycles to improve data-driven

decision making.

This demands the use of novel lifecycle,

techniques, and technologies to enable manufacturers

to connect their data, processes, systems, personnel

and equipment. The main contribution of this paper is

a smart shop-floor monitoring framework with

Pingos, M., Elgammal, A., Kumara, I., Christodoulou, P., Papazoglou, M. and Andreou, A.

Smart Shop-ﬂoor Monitoring via Manufacturing Blueprints and Complex-event Processing.

DOI: 10.5220/0007720802290236

In Proceedings of the 21st International Conference on Enterprise Information Systems (ICEIS 2019), pages 229-236

ISBN: 978-989-758-372-8

229

proactive capabilities, which enables the

identification and resolution of execution disruptions.

The proposed monitoring framework realizes the

“Monitoring” process as a part of the novel smart PSS

lifecycle previously introduced in (Papazoglou et al.,,

2018). The lifecycle provides a closed monitoring

feedback loop that enables continuous product and

service improvements on the basis of the novel

concept of knowledge-intensive structures called

Manufacturing Blueprints (Blueprints in short). The

proposed framework is established on the basis of

these tested structures, which semantically capture

product-service and production-related knowledge

(Papazoglou Elgammal, 2017; Papazoglou et al.,

2015). Blueprints integrate dispersed manufacturing

data from diverse sources and locations, which

include and combine business transactional data and

manufacturing operational data to gain full visibility

and control, and provide the basis for production

actionable “intelligence”. A a middleware layer is

introduced that enables loose coupling with

blueprints and adaptation of the proposed framework

in practice. Furthermore, the framework utilizes CEP

technology (Etzion and Niblett, 2010), the latter

offering event processing which combines data from

multiple sources to infer events or patterns that

suggest more complicated circumstances.

Implementation and validation of the proposed

framework is demonstrated through a real-world case

study sourcing from the H2020 ICP4Life EU Project,

while the validation process assess the applicability,

usability and efficacy of the proposed solutions.

The remainder of the paper is structured as

follows: Section 2 discusses related work efforts in

the areas of servitization and shop-floor monitoring.

Section 3 presents the smart manufacturing

framework and discusses its main components. This

is followed by presenting the current implementation

efforts in section 0. Finally, section 5 concludes the

paper and highlights future work directions.

2 RELATED WORK

Related work efforts found in the literature are mostly

focused on separately addressing aspects of the two

converging research directions in this paper, namely

Servitization and Shop-floor Monitoring. These are

discussed in the next two sub-sections.

2.1 Servitization

Servitization is the innovation of an organization’s

capabilities and processes to shift from selling

products to selling integrated products and services

that deliver value in use (Howard et al., 2013; Tim et

al., 2017). Different approaches in the literature build

on a distinction between products and services, and

demonstrate how a change in the balance between

these can result in different levels of servitization

(Tim and Howard, 2013).

An approach with focus on value proposition that

distinguishes between "base", "intermediate" and

"advanced" services is proposed in (Howard et al.,

2013; Tim and Howard, 2013). The base services

focus on the product provision; intermediate services

are based on exploitation of production competences

to also maintain the condition of products; finally, the

advanced services concentrate on the capability

delivered through performance of the product

(Howard et al., 2013; Tim and Howard, 2013).

Another frequently addressed approach for PSS

classification proposes distinguishing between three

main categories (Baines et al., 2007): (i) product-

oriented, (ii) use-oriented, and (iii) result-oriented. In

the product-oriented PSS, the product is offered in a

traditional sale model, but also includes the sale of

additional services (Baines et al., 2007). In the use-

oriented and the result-oriented PSS, customer

satisfaction is achieved by the functions provided by

the products or the result of services rather than the

product ownership (Chou et al., 2015).

Unfortunately, existing literature provides little or

no guidance on how to successfully tackle the

servitization challenges. Baines et al. (2009) discuss

the scarcity of previous studies “that provide

guidance, tools or techniques, that can be used by

companies to servitize”, pointing out that “guidance

in the literature on how to approach organizational

strategy (for servitization) is largely limited to

anecdotal evidence from case studies that suggest

good practices and processes for implementation”.

Sai et al. (2016) add to the discussion that most of the

existing servitization studies remain at a theoretical

level, limiting the applicability of the findings.

Unlike efforts in the literature, which lack

concrete IT solutions to realize the vision of

servitization (Howard et al., 2013; Tim et al., 2017),

the proposed monitoring framework in this paper

provides a native support to couple products and

services for their continuous monitoring and

improvement through a closed feedback loop.

2.2 Shop-floor Monitoring

Papazoglou et al. (2018) presented a novel PSS

customized lifecycle approach that includes

technological solutions aiming to enable PSS

ICEIS 2019 - 21st International Conference on Enterprise Information Systems

230

customization. The proposed methodology utilizes

blueprints (Papazoglou and Elgammal, 2017;

Papazoglou et al., 2015), which provide the root for

actionable PSS and production intelligence. As

previously mentioned, the proposed monitoring

framework relies on blueprints as the basis of

manufacturing intelligence; blueprints are briefly

discussed in Section 3.

In addition, manufactures today are moving into a

different direction that targets in fulfilling orders on

demand by negotiating value-adding processes in

real-time, taking at the same time into consideration

quality, time, price etc. The growing demand of

customized production results is considered a major

challenge to traditional manufacturing businesses

(Zhang et al., 2017). Wan et al. (2018) proposed a

customized version of a Smart Factory for

pharmaceutical manufacturing that was tested on a

demand-based drug packing production. That work

also introduced a Manufacturing's Semantics

Ontology knowledge in the perception layer that

aimed to plan the scheduling of the pharmaceutical

production, thus the new plans are directly created

from the production demand and the data collected

from machines. Similarly, the work in (Zhang et al.,

2017) outlined a framework of an intelligent shop-

floor to allocate resources based on the production

requirements. The proposed structure consists of

three models: (i) the smart machine agent model, (ii)

the self-organizing model and (iii) the self-adaptive

model. A prototype cyber-physical system that

includes the aforementioned models was developed

to test the proposed methodology and assess the

flexibility of configuring resources to deal

disturbances.

According to (Theorin et al., 2015) future

manufacturing systems must be flexible in order to

adapt easily in the continuously changing market

demands, but at the same time they need to make a

better use of source data, thus low-level data should

be transformed to real-time information for decision-

making support. (Theorin et al., 2015) presented a

Line Information System Architecture, called LISA,

to enable factory integration and data exploitation.

LISA is an event-driven framework with a prototype-

oriented model which combines international

standards and well-known off-the-shelf technologies

aiming to be mechanically applicable. The work of

(Christ, et al., 2016) introduced a different

methodology based on Complex Event Processing

(CEP), a technology to analyse event streams. The

limitation of traditional CEP is that it cannot consider

events that have not taken place yet, thus this paper

introduces the concept of Conditional Event

Occurrence Density Estimation (CEODE) to CEP.

Christ et al. (2016) outline a structure for merging

CEP engines with predictive analytics using CEODEs

and demonstrates how CEP can change from a

waiting process to predictive and prescriptive, to be

able to deal with the production line challenges.

In a nutshell, this paper proposes a monitoring

framework based on manufacturing blueprints and

Complex Event Processing (CEP) technologies; to

the best of our knowledge, it constitutes one of the

very few, if any, initial studies that combine the

aforementioned approaches under the same structure.

The proposed framework aims to enable the proactive

identification and resolution of shop-floor distribu-

tions in order to help businesses to connect their data,

systems, equipment and personnel, developing at the

same time a user-friendly environment for customers

to customize products where resources are allocated

based on the production requirements.

3 METHODOLOGY AND

APPROACH

The approach of the present paper relies on a

production planning/engineering middleware which

is placed between the process of product design

customization and the actual execution of the

production steps performed at the shop-floor,

following the novel smart product lifecycle

introduced in (Papazoglou et al., 2018). The aim of

the middleware is to provide the missing link

connecting a conceptualized co-creation process

based on which a user/customer designs the desired

product using a graphical environment - for the

purposes of this paper we assume that this design

definition (product request) is already available and

has been performed in the Unreal Engine

environment (https://www.unrealengine.com) - with

the processes and actions executed by machines

during the actual making of the product.

In this context a series of steps are followed which

start by transforming the product customization

information inserted by the user into a standardized

representation to facilitate comparison with existing

knowledge stored in a repository as regards properties

of the product to be developed and the sequence of

operations at the machine level to actually construct

it. To this end, the concept of Blueprints is adopted,

to take advantage of their formal, standardized

representation of product properties, events and

workflows for the conceptual description of the

details for building a product as analyzed earlier.

Smart Shop-ﬂoor Monitoring via Manufacturing Blueprints and Complex-event Processing

231

Figure 1: UML Meta-model OWL ontology in Protégé.

The existing knowledge, as expressed through

Blueprints, essentially describes product and

production related information, which consists of

product events knowledge and product emergency

events knowledge.

The latter is based on extending certain Blueprint

types, namely Production Process Blueprint (PPB)

and Production Service Blueprint (PSB), something

which constitutes part of the novelty of the present

paper. This extension to abovementioned Blueprints

(see figure 1) provides refined details for the

machines at the shop-floor involved in the specific

product creation, and more specifically for the type,

frequency of sampling (timing) and thresholds of

sensors these machines include, as well as a list of

actions to handle each emergency event according to

sensor values (thresholds). Thresholds are defined by

the control room operator or the shift manager, and in

this work we consider these values as given.

In the next step the new knowledge is pushed to

the shop-floor to complement and extend the normal

workflow of machine actions throughout the process

of building the product with actions that handle

undesired cases (i.e., when alerts are raised). During

this step the product request is compared with the

existing knowledge stored and the closest match is

used as the backbone to define refinements and revise

the sequence of events for building the product and

perform emergency actions. This revision creates a

new Blueprint instance. The dedicated middleware

receives this Blueprint instance and follows the cycle

of normal execution and emergency actions described

therein by translating them into a series of events that

will take place at the shop-floor. To do so it parses

and queries the RDF/OWL (https://www.w3.org/

OWL/) images of the Blueprints involved to retrieve

this information.

Focusing on the emergency actions, the

middleware supports the monitoring and control of

the execution of normal tasks by the machines and

their actuators. The “product monitoring and

actuating knowledge” is a process that is invoked in

parallel with normal execution and monitors the

threshold values in contrast with sensor readings so

that in cases where anomalies are detected the proper

actions are initiated, again in the form of events, as

these were previously defined by the user (e.g. the

shop-floor manager). This initiation process is

handled by a tool called WSO2 (https://wso2.com/),

which is essentially an event-driven framework that

supports event-driven systems. Therefore, our

approach follows the integration of WSO2 and

Blueprints in terms of translating the conceptual

Blueprint actions into actual steps/events executed

through WSO2.

WSO2 is an open-source enterprise platform that

enables integration of application programming

interfaces (APIs), applications and web services both

at local level and across the Internet. WSO2 offers a

platform of middleware products for agile integration,

API management, identity and access management,

and smart analytics.

WSO2 essentially monitors in real-time the

execution steps at the shop-floor by receiving real-

time events. The detection of a violation of any of the

defined monitoring rules (and/or thresholds), apart

ICEIS 2019 - 21st International Conference on Enterprise Information Systems

232

Figure 2: Functional representation of the proposed middleware

from the alerting process, fires the appropriate

response action(s) defined in the corresponding rules

of the Blueprints. An action in its simplest form could

be the generation of an alert sent to the shop-floor

manager, and in a more advanced case, the initiation

of a series of signals/actions/controls passed directly

to the shop-floor to instruct and drive the actuators of

the machines to execute a process. For example, a

possible action as a response to the detection of a rise

in temperature for a welding machine (as compared

to the predefined threshold), is to send a signal to the

actuators at the shop-floor to turn on specific

ventilation or air-conditioning machinery to cool the

place and lower the temperature.

Figure 1 shows the extension of the service

Blueprint in order to reveal alerts in WSO2 CEP tool

by parsing the new RDF file of the extended

Blueprint. Machine and sensors are part of a factory.

Sensors is located in machines and are part of

machines such as CO2 LASER, Laser Cutter and

Drilling Machine. Machines perform normal actions

in order to produce a requested product. If an

abnormal scenario occurs, then emergency actions,

such as alerts and healing actions, are executed.

Certain types of sensors, such as for temperature,

pressure, humidity etc., offer the ability to a set-up

threshold values (e.g. min, max).

Figure 2 shows the whole concept of the

workflow with the combination of the existing tools.

When a customer builds a product request in a

GUI environment (this step is beyond the scope of the

present paper and it is assumed to be available), the

request is transformed into a Blueprint image. Τhis

request-image is then pushed to the middleware in

order to be compared with the existing production

Blueprint monitoring repository images. Shop-floor

managers or engineers have already defined the

thresholds of sensors on the machines which produce

the requested product (e.g., Min Temperature: 50°C

and Max Temperature: 250 °C). In order to push

events and details of the whole production, normal

operation and emergency cases, the latter being based

on threshold values of the sensors at the shop-floor,

PPB and PSB must first be parsed and then queried in

order to obtain this information (see manufacturing

Blueprint - BL images with light red color in figure 1

– second and third from left). PPB consists of

Production Workflow (solution), Process Event and

Data Collection, and Resources Devices and

Equipment. PSB consists of Service Type, Service

Sensors, Service Metrics and Service Schedule

(Papazoglou and Elgammal 2017). What is actually

executed is parsing and querying the images of these

blueprints (PPB, PSB) expressed in RDF/OWL form.

An RDF data model

is similar to classical

conceptual modelling approaches (such as entity–

relationship or class diagrams), but it provides a

semantic support. It is based on the idea of making

Smart Shop-ﬂoor Monitoring via Manufacturing Blueprints and Complex-event Processing

233

Figure 3: Stream example of the attributes of sensors.

statements about resources (in particular web

resources) in expressions of the form subject–

predicate–object, known as triples. The subject

denotes the resource, while the predicate denotes traits

or aspects of the resource, and expresses a relationship

between the subject and the object. RDF is an abstract

model with several serialization formats (i.e. file

formats), so the particular encoding for resources or

triples varies from format to format. A collection of

RDF statements intrinsically represents a labelled,

directed multi-graph. This in theory makes an RDF

data model better suited to certain kinds of knowledge

representation than other relational or ontological

models. However, in practice, RDF data is often stored

in relational database or native representations. OWL

is a computational logic-based language such that

knowledge expressed in OWL can be exploited by

computer programs, e.g., to verify the consistency of

that knowledge, or to make implicit knowledge

explicit. OWL documents, known as ontologies, can be

published in the World Wide Web and may refer to or

be referred from other OWL ontologies. OWL is part

of the W3C’s Semantic Web technology stack, which

includes RDF, RDFS, SPARQL, etc.

(https://jena.apache.org/documentation/query/).

The current paper is not involved with data

visualization; it deals with rules and implements the

definition of the actions to be taken when emergency

cases arise according to some threshold values set,

while it assumes that the latter are already available

as their definition is the subject of another research

work by the authors (reference omitted for blind

review and will be given upon acceptance).

To sum-up, when a request for building a product

is received (see figure 2), the middleware, according

to the request, firstly produces an extended

production profile, which includes both the normal

events that must be executed to produce the desired

customized product and the emergency events, along

with the threshold values of the sensors in each

machine at the shop-floor that, when exceeded,

trigger the execution of emergency actions. Secondly,

it converts this profile into a series of events using

WSO2 and transferring these events to the machines.

At this stage it is also assumed that the details of the

machines required for production, as well as the

number and type of the sensors each machine

includes, have already been defined and are available

prior to the request sent to the middleware

4 IMPLEMENTATION AND

DEMONSTRATION

This section presents a demonstration example where

the methodology described in section 3 is applied on

a real-world use-case. Firstly, an extension in the

Blueprints is performed to refine details for the

machines and the sensors at the shop-floor, as well as

a list of alerts and actions (see table 1) that are used to

handle each emergency event according to sensor

values (thresholds). As presented in figure 3, in our

demonstration example we used three types of

machines, a CO2 Laser, a Laser Cutter and a Drilling

Machine. In addition, to handle the emergency events,

we have utilized various types of sensors that have

been considered as an integral part of the machines,

ICEIS 2019 - 21st International Conference on Enterprise Information Systems

234

such as Temperature, Humidity and Light. Sensors

were defined as presented in figure 3 and their real-

time readings were presented as shown in figure 4.

As already mentioned in the methodology, upon

occurrence of an abnormal event, emergency actions,

alerts and/or healing actions that were defined by the

shift manager upfront are executed/produced

automatically.

Table 1: Demonstration alerts and actions.

Alerts Actions

Temperature value overcame

upper threshold

Turn on A/C

Humidity value overcame

upper threshold

Stop the execution

Switch-off machine

Following our methodology, the new knowledge

acquired from the previous step is pushed to the shop-

floor. This extends the usual workflow of machines

and aids in the building of a customized product by

creating a new Blueprint instance.

Figure 4: Real-time snapshot of sensors’ readings.

Figure 5: JAVA pseudocode for handling and querying

Blueprint images.

Figure 6: JSON results for the proposed example.

The proposed methodology utilizes Apache Jena

to handle the Blueprint images and SPARQL for

accessing Blueprint images. SPARQL takes the

description in the form of a query, and returns that

information, in the form of a set of bindings or an

RDF graph. The JAVA code/pseudocode for handling

and querying the Blueprint images of the

demonstrated example is shown in figure 5 and the

outcomes of the JSON file that were imported into

WSO2 are presented in figure 6.

In summary, the proposed methodology

supported the identification of the most meaningful

events and patterns from multiple data sources, the

analysis of their impact, and the decisions to be taken

for resolving any problems occurred in real-time.

5 CONCLUSIONS AND FUTURE

WORK

The present paper introduces a smart shop-floor

monitoring framework that supports the identification

and resolution of execution disruptions with proactive

capabilities. The framework adopts a smart PSS

lifecycle to offer a monitoring feedback loop that

enables continuous product and service

improvements based on knowledge-intensive

structures called Manufacturing Blueprints. The latter

integrates data from diverse sources and locations in

the manufacturing environment and facilitate

production actionable “intelligence”. The

middleware layer of the proposed framework

connects it with the Blueprints offering at the same

time adaptation capabilities in order to be fully

operational in practice. In addition, the framework

utilizes Complex Event Processing technology to

combines data from the multiple sources present at

the shop floor and infer events or patterns according

to the circumstances. The proposed framework is

being implemented and validated using a real-life

case study demonstrating its applicability, usability

and efficiency.

Smart Shop-ﬂoor Monitoring via Manufacturing Blueprints and Complex-event Processing

235

Future research is ongoing in a number of parallel

and complementary directions, which include: (i)

design and development of a user-friendly graphical

domain-specific language to enable the product

engineer/designer to specify and interpret monitoring

rules in a user-friendly and intuitive manner, (ii)

(semi-) automating recovery actions by seeding self-

adaptiveness and self-healing capabilities, moving

towards the vision of self-autonomous smart factory,

and (iii) augmenting the dashboard with sophisticated

visualization features by supporting augmented and

virtual reality.

ACKNOWLEDGEMENTS

This paper is part of the outcomes of the Twinning

project Dossier-Cloud. This project has received

funding from the European Union’s Horizon 2020

research and innovation programme under grant

agreement No 692251.

REFERENCES

Baines, T., Lightfoot, H. W., Evans, S., Neely, A.,

Greenough, R., Peppard, J., Wilson, H., 2007. State-of-

the-art in product-service systems. Proceedings of the

Institution of Mechanical Engineers, Part B: Journal of

Engineering Manufacture, 221(10).

Baines, T. S., Lightfoot, H. W., Benedettini, O., & Kay, J.

M., 2009. The servitization of manufacturing: A review

of literature and reflection on future challenges. Journal

of Manufacturing Technology Management.

Bustinza, O. F., Bigdeli, A. Z., Baines, T., & Elliot, C.,

2015. Servitization and Competitive Advantage: The

Importance of Organizational Structure and Value

Chain Position. Research-Technology Management.

Chou, C.-J., Chen, C.-W., & Conley, C., 2015. An approach

to assessing sustainable product-service systems.

Journal of Cleaner Production, 86, 277-284.

Christ, M., Krumeich, J., & Kempa-Liehr, A. W., 2016.

Integrating Predictive Analytics into Complex Event

Processing by Using Conditional Density Estimations.

Paper presented at the 2016 IEEE 20th International

Enterprise Distributed Object Computing Workshop

(EDOCW).

Elgammal, A., Papazoglou, M., Krämer, B., &

Constantinescu, C., 2017. Design for Customization: A

New Paradigm for Product-Service System

Development. Paper presented at the The 9th CIRP

IPSS Conference: Circular Perspectives on

Product/Service-Systems, Denmark.

Etzion, O., & Niblett, P., 2010. Event Processing in Action:

Manning Publications Co.

Howard, L., Tim, B., & Palie, S., 2013. The servitization of

manufacturing: A systematic literature review of

interdependent trends. International Journal of

Operations & Production Management, 33(11/12).

Papazoglou, M., & Elgammal, A., 2017. The

Manufacturing Blueprint Environment: Bringing

Intelligence into Manufacturing. Paper presented at the

IEEE International Conference on Engineering,

Technology and Innovation (ICE), Portugal.

Papazoglou, M., Heuvel, W. J. v. d., & Mascolo, J., 2015.

Reference Architecture and Knowledge-based

Structures for Smart Manufacturing Networks. IEEE

Software, PP(99), 1-1.

Papazoglou, M. P., Elgammal, A., & Krämer, B. J., 2018.

Collaborative on-demand Product-Service Systems

customization lifecycle. CIRP Journal of

Manufacturing Science and Technology.

Sai, S. N., David, L., Gillian, W., & Nick, Y., 2016. Eight

challenges of servitisation for the configuration,

measurement and management of organisations.

Journal of Service Theory and Practice, 26(6).

Song, W. 2017., Requirement management for product-

service systems: Status review and future trends.

Computers in Industry, 85, 11-22.

Theorin, A., Bengtsson, K., Provost, J., Lieder, M.,

Johnsson, C., Lundholm, T., & Lennartson, B. 2015.

,An Event-Driven Manufacturing Information System

Architecture. IFAC-PapersOnLine, 48(3), 547-554.

Tim, B., Ali, Z. B., F., B. O., Guang, S. V., James, B., &

Keith, R., 2017. Servitization: revisiting the state-of-

the-art and research priorities. International Journal of

Operations & Production Management, 37

(2).

Tim, B., & Howard, W. L., 2013. Servitization of the

manufacturing firm: Exploring the operations practices

and technologies that deliver advanced services.

International Journal of Operations & Production

Management, 34(1), 2-35.

Wan, J., Tang, S., Li, D., Imran, M., Zhang, C., Liu, C., &

Pang, Z., 2018. Reconfigurable Smart Factory for Drug

Packing in Healthcare Industry 4.0. IEEE Transactions

on Industrial Informatics, 1-1.

Zhang, Y., Qian, C., Lv, J., & Liu, Y., 2017. Agent and

Cyber-Physical System Based Self-Organizing and

Self-Adaptive Intelligent Shopfloor. IEEE

Transactions on Industrial Informatics, 13(2), 737-747.

ICEIS 2019 - 21st International Conference on Enterprise Information Systems

236