Development of a Semantic Database Model to Facilitate Data

Analytics in Battery Cell Manufacturing

Ozan Yesilyurt

, David Brandt

, Julian Joël Grimm

, Kamal Husseini

Aleksandra Naumann

3,4 e

, Julia Meiners

3,4 f

and David Becker-Koch

Fraunhofer Institute for Manufacturing Engineering and Automation IPA Nobelstraße 12,

70569 Stuttgart, Germany

Karlsruhe Institute of Technology, Kaiserstraße 12, 76131 Karlsruhe, Germany

Technische Universität Braunschweig, Institute of Machine Tools and Production Technology, Langer Kamp 19b,

38106 Braunschweig, Germany

Technische Universität Braunschweig, Battery LabFactory Braunschweig, Langer Kamp 8,

38106 Braunschweig, Germany

The Centre for Solar Energy and Hydrogen Research Baden-Württemberg, Lise-Meitner-Straße 24,

89081 Ulm, Germany

{al.naumann, j.meiners}@tu-braunschweig.de, david.becker-koch@zsw-bw.de

Keywords: Battery Cell Manufacturing, Database Model, Semantic Model Description.

Abstract: The global demand for batteries is increasing worldwide. To cover this high battery demand, optimizing

manufacturing productivity and improving the quality of battery cells are necessary. Digitalization promises

to offer great potential to address these challenges. Through data collection along the manufacturing processes,

hidden correlations can be identified. However, data is highly diverse in battery cell manufacturing,

complicating data analysis. A semantic data storage can increase the understanding of the relationships

between the datasets, facilitating the identification of the causes of defects in manufacturing processes. To

structure heterogeneous data in a semantically understandable and analyzable form, this paper presents the

development of a semantic database model. The realization of this model enables structuring various datasets

for simplified access and usage for increasing productivity and battery cell quality in battery cell

manufacturing.

1 INTRODUCTION

The global demand for batteries for energy storage is

growing due to the continued development of electric

vehicles and other mobile devices (Asif & Singh,

2017). The growing number of battery-electric

vehicles registered illustrates the increasing global

demand for battery cells (Carlier, 2021). To meet this

high demand of the battery cell users, digitalization of

production offers several opportunities to build a

https://orcid.org/0000-0003-3002-3230

https://orcid.org/0000-0003-1781-9541

https://orcid.org/0000-0002-0559-3752

https://orcid.org/0000-0002-7110-4697

https://orcid.org/0000-0002-9193-1294

https://orcid.org/0000-0002-6564-6537

https://orcid.org/0000-0001-5921-5768

flexible, intelligent, adaptable, and efficient

manufacturing system (Zhong, Xu, Klotz, &

Newman, 2017). Data is the key element to realizing

such manufacturing systems, and its availability has

been rapidly growing in the manufacturing industry

(Yin & Kaynak, 2015). Smart manufacturing targets

transforming data towards manufacturing intelligence

to positively affect every manufacturing-related

aspect (O’Donovan, Leahy, Bruton, & O’Sullivan,

2015).

Yesilyurt, O., Brandt, D., Grimm, J., Husseini, K., Naumann, A., Meiners, J. and Becker-Koch, D.

Development of a Semantic Database Model to Facilitate Data Analytics in Battery Cell Manufacturing.

DOI: 10.5220/0011139500003269

In Proceedings of the 11th International Conference on Data Science, Technology and Applications (DATA 2022), pages 13-20

ISBN: 978-989-758-583-8; ISSN: 2184-285X

 2022 by SCITEPRESS – Science and Technology Publications, Lda. All rights reser ved

The use of data-driven manufacturing

technologies is appealing for battery cell

manufacturing to improve scrap and product quality.

Data in battery cell manufacturing is heterogeneous,

resulting from both converging and diverging

material flows, continuous and non-continuous

processes, and single and batch processes (Turetskyy

et al., 2020). On the one hand, it is essential to build

data storage in battery cell manufacturing to solve

heterogeneous data problems. It is challenging to

collect unstructured data from different

manufacturing processes and perform informative

analyses. On the other hand, it is a major gap in

battery cell manufacturing to create and enable

context-based data in a semantically understandable

way because heterogeneous data consists of different

unrelated datasets, which hinders identifying the

cause of manufacturing process issues and

discovering hidden optimization opportunities in the

manufacturing process.

A possible solution to address these challenges is

creating a semantic database with a model that

facilitates identifying, accessing and processing the

appropriate data. Semantically structured data

enables data scientists to track and evaluate the

manufacturing processes and find optimization

potentials in the battery cell manufacturing with

various analysis tools.

This paper aims to present the requirements for a

specific battery manufacturing scenario and present a

possible implementation of such a semantic database

model as a solution.

The following chapter describes the data storage

transformation. It also presents current data

challenges, introduces developed approaches to solve

them, and derives the scientific gap. Chapter 3

introduces the specific scenario for which the

semantic database model is created. The following

two chapters illustrate the model’s underlying

concept and its realization. The paper concludes with

a summary and highlights the future work with the

developed semantic database model.

2 RELATED WORK

Data is a key element for smart manufacturing to meet

manufacturing needs and inform manufacturing

decision-making areas (O’Donovan et al., 2015).

Relational databases have been used for decades as

regular database systems to store manufacturing data.

A relational database is a digital database used for

electronic data management in computer systems and

is based on a table-based relational database model as

proposed by (Codd, 1970). With the introduction of

IoT technologies, cloud computing, big data analytics,

and AI integrated into manufacturing systems, a high

level of multi-source and heterogeneous data is

generated (Tao, Qi, Liu, & Kusiak, 2018). Therefore,

more and more new database technologies were

integrated into the existing data storage architectures,

such as NoSQL. With these, the challenges associated

with storing the great amount of manufacturing data

could be adressed. NoSQL databases became widely

used around 2009, which process data faster than

relational databases because their data models are

built more simply (Leavitt, 2010). They can be

categorized into five groups (Column-based,

document-based, key-value-based, graph-based,

time-series-based) (Cui, Kara, & Chan, 2020; Yen,

Zhang, Bastani, & Zhang, 2017). The same big data

challenge is also observed in battery cell

manufacturing. Various systems (e.g. equipment,

controls and simulation models) are involved and

generate heterogeneous (e.g. time series, discrete)

data in large volumes. Data is distributed in several

heterogeneous datasets and databases that need to be

linked to enable in-depth analysis for identifying and

addressing issues in battery cell manufacturing

processes.

Some solution approaches are developed to

address this need. A hybrid framework for industrial

data storage to utilize zero-defect manufacturing was

introduced by (Grevenitis et al., 2019), where

unstructured data generated by the IoT devices are

processed in a NoSQL database (Apache Cassandra);

at the same, time the structured data is stored in a SQL

database (MySQL). Then, the filtered data from both

databases is converted into knowledge and stored in a

triplestore database to be used by experts.

Furthermore, (Hildebrand, Tourkogiorgis,

Psarommatis, Arena, & Kiritsis, 2019) developed a

generic algorithm for the automated conversion of

different data types into RDF. With this solution, the

researchers seek to enable data mapping without hard

coding. This solution is reusable across various data

schemas and ontologies, which can be easily

modified to fit other data formats. In addition,

(Wessel, Turetskyy, Wojahn, Abraham, & Herrmann,

2021) presented and implemented a methodology to

develop an ontology-based traceability system in

battery cell manufacturing so that the relations

between the data sources along the manufacturing

chain can be determined. Moreover, (Grimmel,

Wessel, Mennenga, & Herrmann, 2022) introduced

an ontology-based data processing that enables the

creation and distribution of knowledge from

decentralized and unstructured data such as

DATA 2022 - 11th International Conference on Data Science, Technology and Applications

warehouse static data, stock exchange data, energy

demand data, machine data and the production plan

of heterogeneous battery processes in a learning

factory. Furthermore, (Malburg, Klein, & Bergmann,

2020) developed 70 semantic web services based on

different ontologies for intelligent manufacturing

control to enable a near real-time verification for

executing cyber-physical workflows. Last, (Kalaycı

et al., 2020) created a framework in that

manufacturing data from the machines of the Bosch

company in Salzgitter for placing electronic

components and automated optical inspection are

semantically integrated to be used in quality analysis

tasks. However, the semantic representation (e.g.

relations between process parameters and mapping of

various discrete and time-series datasets from

equipment, simulation models, and controls) to

enable linked and structured datasets, which can be

used to analyze optimization potentials in the battery

cell manufacturing, has not yet been considered. The

next chapter presents the battery cell manufacturing

use case scenario adressed in this paper.

3 THE VIPRO PROJECT

The concept of the semantic database model for smart

battery cell manufacturing is embedded in the project

ViPro – “Virtual Production Systems in Battery Cell

Manufacturing for cross-process production control”.

The project’s objective is to develop and validate a

concept of cross-process control with a virtual

production system. The concepts’ envisioned benefits

are an increase in battery cell manufacturing

productivity and an improvement in the quality of the

produced cells through an efficient operation.

Realistic and low-risk testing of optimization

measures shall be conducted in the virtual space.

Once satisfactory results are achieved, these measures

can be implemented seamlessly in the battery

manufacturing process.

The overall system architecture of ViPro consists

of different components shown in

Figure

. The superordinate systems intelligent

operation control and cross-process control are

connected and perform data processing. The

intelligent operation control system includes operator

interfaces, where target conditions can be entered and

relevant data for decisions is displayed. The cross-

process control contains machine learning algorithms

evaluating process control values based on

intermediate product features of preceding process

steps.

Figure 1: ViPro overall system architecture.

Product features are evaluated within the single

process models of the virtual production system based

on the respective process input parameters. The

virtual production system includes coating, stacking,

electrolyte filling, and formation quality prediction

models. All of the virtual process representations

have a corresponding physical system: the coating

machine of the ZSW “research platform for the

industrial production of large lithium-ion cells”, the

stacking machine of the KIT “Battery Technology

Center”, the electrolyte filling unit of the TU

Braunschweig “Battery LabFactory Braunschweig”,

and the formation equipment of the Fraunhofer IPA

“Center for Battery Cell Manufacturing”.

To enable communication between the different

ViPro components, Virtual Fort Knox (VFK) is

implemented as a cloud IoT platform together with

the communication middleware Manufacturing

Service Bus (MSB) (Schel et al., 2018). The

connection from the physical equipment to the VFK-

platform is carried out with Station Connector

(Defranceski, 2021), which enables a control-

independent communication.

To enable and ensure communication between the

different services models, the intelligent operation

control and cross-process control systems, a

representation of the information structure of the

underlying complex process behavior and

relationships, as well as the heterogeneous data, is

needed. This representation of the information

structure builds up a network between the different

services enabling operable communication and direct

data exchange. Furthermore, continuous data

exchange from the physical production equipment is

needed to evaluate the potential for cross-process

control. To address these challenges the semantic

database presented in this paper is developed and

integrated into the ViPro overall system architecture.

Development of a Semantic Database Model to Facilitate Data Analytics in Battery Cell Manufacturing

Figure 2: Semantic database concept.

4 SEMANTIC MANUFACTURING

DATABASE CONCEPT

The concept described in this paper aims to develop a

semantic database that structures heterogeneous data

in a semantically understandable and analyzable form,

enabling a solid data foundation to analyze and

optimize battery cell manufacturing. First, the

requirements for developing this database concept are

identifieed, and then the designed concept is

introduced. After that, the semantic database model is

described, consisting of the semantic description of

the four simulation models considered in the ViPro

project. Then, it is explained why the considered

simulation models require this semantic database

model and which quality and general parameters of

this database model are relevant for the simulation

models. Last, the requirements of the intelligent

operation control and cross-process control systems

are introduced. These requirements define what kind

of datasets they need from the semantic database and

which features the semantic database should

additionally have.

The following general requirements have been

identified collaboratively with the project´s process

engineers for the semantic database in the ViPro

scenario:

• Different heterogeneous data from the models,

equipment, and control systems such as

experiment, equipment’s process data

(measurement data), and recipe data should be

stored digitally.

• The data should be easily accessible and

trackable.

• The relationships between different process

parameters of the simulation data should be

representable.

• The data from the equipment should be stored

in a time frame of one-second intervals.

• The data from the equipment should be linked

to the data from the simulation models.

• The database system should be centrally

provisioned and enable access for all partners

involved (e.g. via virtual private networks)

and the possibility to operate across processes.

• The database should provide the interfaces so

that required data can be processed via these

interfaces by other IT systems in the ViPro

project

To cover the requirements, the proposed solution

is a combination of several databases for the different

types of data and a service that manages the

referencing of the data (see Figure 2). Since the

relationships between the different types of data and

their underlying knowledge need to be stored, a

knowledge graph is used to store data. A knowledge

graph represents a self-describing knowledge base

that stores data and its schema in a graph format and

illustrates their relationship (Fensel et al., 2020). A

graph database is used in this paper to store the

knowledge graph. Due to the requirements of storing

DATA 2022 - 11th International Conference on Data Science, Technology and Applications

and accessing data in real-time, a time-series database

(TSDB) is used. The TSDB stores the data with an

additional tag such as experiment ID to reference the

experiment data, stored in a graph database. The

graph database stores this reference key (experiment

ID) to describe the simulation data semantically. With

this design, the service can store different forms of

data and respond to other ViPro IT systems’ requests.

A middleware enables fast and low-overhead

integration of smart objects and IT services (e.g.,

equipment, simulation models, and database service).

Therefore, a middleware that understands different

protocols is used to access the data and to connect to

the database service.

After introducing the designed concept for the

semantic database, the semantic database model is

described below. A semantic database model is

required to identify the relations between the

heterogeneous stored datasets in databases so that

they can be easily exported from databases with their

linked information to analyzing tools.

Communication between the different services

models, the intelligent operation control, and cross-

process control systems must be ensured. Therefore,

data has to be transferred fast from and to the database.

Figure 3: Semantic database model description.

Furthermore, the simulation models' heterogeneous

input and output data need to be considered. That is

time-series data from the process execution and

control data from the cross-process control and the

intelligent operation control systems. To do so, a

representation of the information structure of the

simulation models concerning behavior and

relationship has to be implemented in the semantic

database. This representation of the simulation

models’ information structure is the semantic

database model. It is developed based on the

description of the four simulation models considered

in the ViPro architecture and described in the

following paragraphs.

Figure 3 shows the structure of the semantic

database model description. Various experiments are

carried out in which data of input and output

parameters are stored in the graph database. Each

experiment contains the considered battery cell

manufacturing processes. The input data are divided

into general input parameters and control parameters.

The general input parameters include, for example,

material parameters and specifications regarding the

cell format, whereas the control parameters include

the setting parameters on the respective production

machine. The output parameters are further

subdivided into general output parameters and quality

parameters. The general output parameters include

values that each process step contains. These are, for

example, statements about the energy requirements of

the process and the throughput. The quality

parameters provide information about the respective

intermediate product properties or process quality.

The structure described can be transferred into a data

exchange format such as JSON or XML.

Four processes of battery cell manufacturing are

considered for the semantic description of the models,

which need to be integrated into the overall system of

cross-process control: Electrode coating, assembly,

electrolyte filling, and formation. The models are

based on different approaches, and each of them

focuses on specific key quality parameters.

The model of the electrode coating process is

based on historical data. A Kernel density estimation

is used to determine the relationships between input

and output parameters and tested via cross validation

and optimized with a grid search (Hasilová & Vališ,

2018). The key parameter of this model is the coating

weight per unit area, which has a major impact on

final cell performance. The next model considers the

process of cell assembly. For this purpose, a

simulation is implemented using Simcenter Amesim,

which depicts the separation of electrodes and the

stacking process. In particular, the target web tension

Development of a Semantic Database Model to Facilitate Data Analytics in Battery Cell Manufacturing

Figure 4: Semantic database components.

and the web speed are the parameters that determine

the final dimensional accuracy of sheets. Furthermore,

the electrolyte filling of battery cells is considered.

Due to the lack of quality data on the filling process,

known historical relationships of various publications

are used to map the process. The amount of

electrolyte filled in is a decisive factor in determining

the quality parameter wetting degree. The last model

represents the process of formation. It is implemented

as a grey-box model in MATLAB/Simulink and

consists of a discrete-event-simulation and a database

model part. The key parameters of the formation

process equal the parameters for final cell

performance, which are cell capacity and efficiency.

To access data fast, the described structure of the

database needs to be built up logically and according

to the intelligent operation control and cross-process

control systems that work together closely. They both

need similar individually compiled datasets for

different applications. One important requirement is

that certain data types can be extracted from other

experiments. For the human-machine interface, the

control and machine learning modules, default

datasets of process parameters, and historical and

real-time data are needed from the database and the

cross-process control system, respectively. The

datasets are used for monitoring and decision support.

Since the intelligent operation control system should

be able to access and present all types of data related

to the whole manufacturing process, including

different analyses, the interfaces must be defined and

designed accordingly.

5 IMPLEMENTATION

This chapter introduces the selected database

software systems first. Then, the realization of the

database systems for ViPro is described. At last, it is

examined whether this realization fulfills the

predefined requirements and can be implemented in

all battery cell manufacturing processes.

Neo4j is used as a graph database for storing the

model data and its references to the control data and

equipment data in a synchronized manner. The

determining factor was Neo4j’s numerous advantages

in points of performance, flexibility, and

interoperability.

The process data of four different ViPro equipment

should be stored in one-second intervals in a database.

According to the executed benchmark and the test

results of (Hao et al., 2021), InfluxDB has the best

compression performance, the highest performance at

writing data at high concurrency and handles queries

faster compared to the other three TSDBs.

The realization of the database systems for ViPro

is illustrated in Figure 4. A database service with

InfluxDB and Neo4j connectors which can process

data according to the relevant database, was

developed and implemented. It is called Semantic

Database Query Engine (SDQE). Using the Neo4j

connector, the SDQE can store the simulated model

data in a knowledge graph, as shown in Figure 2. The

InfluxDB connector was developed in SDQE for

storing the control and equipment data. The SDQE

controls the underlying data with the service layer and

decides which data structure is stored in which

database. An application programming interface

(API) layer was defined that can process the

requests from the MSB middleware and the databases.

According to the defined data structure in Figure

3, a schema was created in Neo4j. The structure in

Figure 5 shows the different information and their

relationships. It shows that the "Experiment" node is

connected to the "Process" node, which in turn may

have the "InputParameter" and "OutputParameter"

nodes. The parameter nodes then store the value and

its reference to the TSDB. The "Unit" node is not

DATA 2022 - 11th International Conference on Data Science, Technology and Applications

stored directly in the parameter to make the nodes as

atomic as possible for better scalability.

As described above, the implementation solution

consists of a Neo4j graph database and Influx TSDB.

Neo4j graph database contains the data of simulation

models, while InfluxDB allows storing the data from

the equipment and control systems. To query these

databases, two database interfaces with InfluxQL and

GraphQL were developed in the SDQE. Furthermore,

every stored data is identified with a reference key in

both databases so that the data from the databases can

be accessed and used more efficiently. The data

stored in the Neo4j graph database is built after the

semantic database model description. This design

aims to visualize the relationships of the simulation

data with each other and to create a fundamental

database for further analysis.

Figure 5: Data schema in Neo4j.

Moreover, both databases are deployed in a cloud

platform so that only certain services and users have

access to these systems via VPN. In addition, the first

tests are performed with the equipment. The

equipment's process data (measurement data) can be

read out in a one-second time cycle and stored with a

tag (experiment ID, which is created by the Neo4j

graph database) in the InfluxDB.

The proposed semantic database model is a generic

model for all battery cell manufacturing processes.

Only the input and output parameters should be

adjusted for the new manufacturing processes. Thanks

to the Neo4j IT architecture, new processes can be

easily added to the database with new input and output

parameters. Additionally, the equipment data of the

new processes can be stored in the InfluxDB as new

measurements. With a new reference key, which

SDQE creates, the data from Neo4j can be linked to the

equipment data from the InfluxDB.

6 CONCLUSIONS

A semantic database with a model is developed

considering the ViPro use case for smart battery cell

manufacturing. First, requirements that enable

managing heterogeneous data in a semantically

comprehensible and analyzable format are identified.

Then, the designed concept is introduced. Last, the

realization of two different database technologies and

a connecting middleware was presented.

The developed concept shows that the

heterogeneous data from the simulation models,

equipment, and controls can be stored in a

semantically understandable way. Various datasets

are linked and structured for all manufacturing

processes, which can be used later to analzye and

exploit battery cell manufacturing optimization

potentials. In future work, the communication

solution is built for the ViPro use cases so that the

ViPro IT-components exchange data with each other

and retain the data as in the created concept. Future

developments can include the inclusion of further

processes and models or a transfer to different use

cases and industries.

ACKNOWLEDGEMENTS

This research was conducted within the scope of the

project “Virtuelle Produktionssysteme in der

Batteriezellfertigung zur prozessübergreifenden

Produktionssteuerung” (ViPro). The authors

gratefully acknowledge the financial support of the

German Federal Ministry of Education and Research

(BMBF). D.B.-K. would like to thank Steffen

Stökler-Thurn for the introduction to the subject and

the support.

REFERENCES

Asif, A., & Singh, R. (2017). Further Cost Reduction of

Battery Manufacturing. Batteries, 3(4), 17.

https://doi.org/10.3390/batteries3020017

Development of a Semantic Database Model to Facilitate Data Analytics in Battery Cell Manufacturing

Carlier, M. (2021). Worldwide number of battery electric

vehicles in use from 2016 to 2020: (in millions).

Retrieved from https://www.statista.com/statistics/270

603/worldwide-number-of-hybrid-and-electric-vehicle

s-since-2009/

Codd, E. F. (1970). A relational model of data for large

shared data banks. Communications of the ACM, 13(6),

377–387. https://doi.org/10.1145/362384.362685

Cui, Y., Kara, S., & Chan, K. C. (2020). Manufacturing big

data ecosystem: A systematic literature review. Robotics

and Computer-Integrated Manufacturing, 62, 101861.

https://doi.org/10.1016/j.rcim.2019.101861

Defranceski, M. (2021). Auslesen und Nutzen von

Maschinendaten einfach gemacht. Wt Werkstattstechnik

Online, 111(03), 159–160. https://doi.org/10.37544/

1436–4980–2021–03–67

Fensel, D., Şimşek, U., Angele, K., Huaman, E., Kärle, E.,

Panasiuk, O., Toma, I., Umbrich, J., Wahler, A. (2020).

Why We Need Knowledge Graphs: Applications. In D.

Fensel, U. Şimşek, K. Angele, E. Huaman, E. Kärle, O.

Panasiuk, Toma, I., Umbrich, J., A. Wahler (Eds.),

Springer eBook Collection. Knowledge Graphs:

Methodology, Tools and Selected Use Cases (1st ed., pp.

95–112). Cham: Springer International Publishing;

Imprint Springer. https://doi.org/10.1007/978-3-030-

37439-6_4

Grevenitis, K., Psarommatis, F. [F.], Reina, A., Xu, W.,

Tourkogiorgis, I. [I.], Milenkovic, J., Cassina, J., Kiritsis,

D. [D.] (2019). A hybrid framework for industrial data

storage and exploitation. Procedia CIRP, 81, 892–897.

https://doi.org/10.1016/j.procir.2019.03.221

Grimmel, P., Wessel, J., Mennenga, M., & Herrmann, C.

(2022). Potentials of ontology-based knowledge

discovery in data bases for Learning Factories. SSRN

Electronic Journal. Advance online publication.

https://doi.org/10.2139/ssrn.4073026

Hao, Y., Qin, X., Chen, Y., Li, Y., Sun, X., Tao, Y., Zhang,

X., Du, X. (2021). TS-Benchmark: A Benchmark for

Time Series Databases. In 2021 IEEE 37th International

Conference on Data Engineering (ICDE) (pp. 588–599).

IEEE. https://doi.org/10.1109/ICDE51399.2021.00057

Hasilová, K., & Vališ, D. (2018). Non-parametric estimates

of the first hitting time of Li-ion battery. Measurement,

113, 82–91. https://doi.org/10.1016/j.measurement.20

17.08.030

Hildebrand, M., Tourkogiorgis, I. [Ioannis], Psarommatis, F.

[Foivos], Arena, D., & Kiritsis, D. [Dimitris] (2019). A

Method for Converting Current Data to RDF in the Era

of Industry 4.0. In F. Ameri, K. E. Stecke, G. von

Cieminski, & D. Kiritsis (Eds.), IFIP Advances in

Information and Communication Technology. Advances

in Production Management Systems. Production

Management for the Factory of the Future (Vol. 566, pp.

307–314). Cham: Springer International Publishing.

https://doi.org/10.1007/978-3-030-30000-5_39

Kalaycı, E. G., Grangel González, I., Lösch, F., Xiao, G., ul-

Mehdi, A., Kharlamov, E., & Calvanese, D. (2020).

Semantic Integration of Bosch Manufacturing Data

Using Virtual Knowledge Graphs. In J. Z. Pan, V.

Tamma, C. d’Amato, K. Janowicz, B. Fu, A. Polleres, O.

Seneviratne, L. Kagal (Eds.), Springer eBook Collection:

Vol. 12507. The Semantic Web – ISWC 2020: 19th

International Semantic Web Conference, Athens, Greece,

November 2–6, 2020, Proceedings, Part II (1st ed., Vol.

12507, pp. 464–481). Cham: Springer International

Publishing; Imprint Springer. https://doi.org/10.1007/

978-3-030-62466-8_29

Leavitt, N. (2010). Will NoSQL Databases Live Up to Their

Promise? Computer, 43(2), 12–14.

https://doi.org/10.1109/MC.2010.58

Malburg, L., Klein, P., & Bergmann, R. (2020, November 2

- 2020, November 4). Semantic Web Services for AI-

Research with Physical Factory Simulation Models in

Industry 4.0. In Proceedings of the International

Conference on Innovative Intelligent Industrial

Production and Logistics (pp. 32–43). SCITEPRESS -

Science and Technology Publications.

https://doi.org/10.5220/0010135900320043

O’Donovan, P., Leahy, K., Bruton, K., & O’Sullivan, D. T. J.

(2015). An industrial big data pipeline for data-driven

analytics maintenance applications in large-scale smart

manufacturing facilities. Journal of Big Data, 2(1).

https://doi.org/10.1186/s40537-015-0034-z

Schel, D., Henkel, C., Stock, D., Meyer, O., Rauhöft, G.,

Einberger, P., Stöhr, M., Daxer, M.A., Seidelmann, J.

(2018). Manufacturing Service Bus: An Implementation.

Procedia CIRP, 67, 179–184. https://doi.org/10.1016/

j.procir.2017.12.196

Tao, F., Qi, Q., Liu, A., & Kusiak, A. (2018). Data-driven

smart manufacturing. Journal of Manufacturing Systems,

48, 157–169. https://doi.org/10.1016/j.jmsy.2018.01.006

Turetskyy, A., Thiede, S., Thomitzek, M., Drachenfels, N.

von, Pape, T., & Herrmann, C. (2020). Toward Data‐

Driven Applications in Lithium‐Ion Battery Cell

Manufacturing. Energy Technology, 8(2), 1900136.

https://doi.org/10.1002/ente.201900136

Wessel, J., Turetskyy, A., Wojahn, O., Abraham, T., &

Herrmann, C. (2021). Ontology-based Traceability

System for Interoperable Data Acquisition in Battery

Cell Manufacturing. Procedia CIRP, 104, 1215–1220.

https://doi.org/10.1016/j.procir.2021.11.204

Yen, I.‑L., Zhang, S., Bastani, F., & Zhang, Y. (2017). A

Framework for IoT-Based Monitoring and Diagnosis of

Manufacturing Systems. In Sose 2017: 11th IEEE

International Symposium on Service-Oriented System

Engineering: Proceedings: 6-9 April 2017, San

Francisco, California (pp. 1–8). Piscataway, NJ: IEEE.

https://doi.org/10.1109/SOSE.2017.26

Yin, S., & Kaynak, O. (2015). Big Data for Modern Industry:

Challenges and Trends [Point of View]. Proceedings of

the IEEE, 103(2), 143–146. https://doi.org/10.1109/J

PROC.2015.2388958

Zhong, R. Y., Xu, X., Klotz, E., & Newman, S. T. (2017).

Intelligent Manufacturing in the Context of Industry 4.0:

A Review. Engineering, 3(5), 616–630.

https://doi.org/10.1016/J.ENG.2017.05.015

DATA 2022 - 11th International Conference on Data Science, Technology and Applications