Towards a Classiﬁcation Framework for the Digital Twin Tools: A

Taxonomy

Mert Ozkaya

and Alper Turunc

Department of Computer Engineering, Yeditepe University, Istanbul, Turkey

DFDS, Istanbul, Turkey

Keywords:

Digital Twin, Classiﬁcation Framework, Taxonomy, Tools, Technologies.

Abstract:

Digital twins (DTs) have gained an ever-increasing popularity in many industries for the real-time monitoring

of physical systems and performing useful operations such as predictive maintenance and early fault detection.

Many commercial and open-source DT tools are available on the market, which offer services or development

environment for developing and using DTs. However, our research shows that practitioners still use conven-

tional programming technologies for developing DTs in-house rather than using the DT-speciﬁc tools. We

believe that one reason here is to do with the lack of any taxonomy for the determining the DT tools that are of

use for the practitioners. In this paper, we propose our attempt for a classiﬁcation framework that can be used

for analysing and comparing different DT tools. Having reviewed the literature and practitioners’ needs, we

addressed twelve main features for DT tools, which are divided into sub-features. These are (i) domain, (ii) de-

ployment, (iii) type, (iv) maturity, (v) knowledge management, (vi) system integration, (vii) quality assurance,

(viii) reusability, (ix) extensibility, (x) abstraction, (xi) enabling technology, and (xii) development platform.

We believe that our classiﬁcation framework will be useful for different stakeholders: (i) practitioners who

wish to develop and use DTs, (ii) tool vendors who can determine strengths and weaknesses of their tools, (iii)

researchers who address the tool weaknesses.

1 INTRODUCTION

Digital twin (DT) is nowadays considered as one of

the hottest topics of computer science and used in

several industries for the real-time monitoring of any

physical systems remotely and optimising the product

development lifecycle (Rasheed et al., 2020; Jones

et al., 2020; Hu et al., 2021). These include such

industries as aerospace, automotive, manufacturing,

mining, energy, construction, healthcare, agriculture,

smart cities, and oil and gas (Singh et al., 2022).

DTs can offer several advantages for those indus-

tries which include the real-time simulation for the

optimised solutions, personalisation of products, en-

hanced communication among stakeholders, reduced

cost of prototyping, early detection of errors and pre-

dictive maintenance, and remote accessibility (Singh

et al., 2021; Rasheed et al., 2020).

As DTs have gained an ever-increasing popularity

in industries, several different software development

companies offer DT tools whose goal is essentially

to help users solving their business problems effec-

tively and productively. DT development companies

can provide varying solutions for the DT develop-

ment. Some companies release DT tools that actually

offer a development platform with APIs, frameworks,

and libraries which can be used by practitioners in de-

veloping domain-speciﬁc DTs for the speciﬁc needs.

Also, some companies offer DT tools as a service and

such DT tools provide several DT operations that can

be used over cloud licenses (or on-premise) for par-

ticular domains and industries. Besides, some com-

panies offer DT tools with an open-source support for

the development of speciﬁc DTs using re-usable and

customisable generic code under certain constraints.

Table 1 shows some of the well-known DT tools

To understand the use of DT tools in practice, we

focussed on the results of our recent industry survey

that has been conducted online between June-August

2024 and addresses practitioners’ perspectives on the

DT techniques and technologies. Our survey received

131 responses from diverse industries and proﬁles

The full list of DT tools is accessible here: https:

//zenodo.org/records/13956879

The survey questions and the raw response data are ac-

cessible here: https://zenodo.org/records/13628837

Ozkaya, M. and Turunc, A.

Towards a Classiﬁcation Framework for the Digital Twin Tools: A Taxonomy.

DOI: 10.5220/0013239900003896

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 13th International Conference on Model-Based Software and Systems Engineering (MODELSWARD 2025), pages 263-272

ISBN: 978-989-758-729-0; ISSN: 2184-4348

263

Table 1: The list of DT tools.

Digital Twin Tool Website

AnyLogic www.anylogic.com

Tecnomatix

www.plm.sw.siemens.com/en-US/tecnomatix

aPriori www.apriori.com

XMPro App Designer www.xmpro.com

Autodesk Tandem www.intandem.autodesk.com

Azure Digital Twins

www.azure.microsoft.com/en-us/products/digital-twins

Unlearn.AI www.unlearn.ai

Ansys Twin Builder www.ansys.com

Bentley Systems

www.bentley.com/software/infrastructure-digital-twins

Rockwell Automation www.rockwellautomation.com

General Electric www.ge.com

Mimic Simulation

www.mimicsimulation.com/simulation-software

Veerum www.veerum.com

CADMATIC eShare

www.cadmatic.com/en/products/cadmatic-eshare

Archilogic www.archilogic.com

Golem www.golem.at

Value Chain

Resilience GmbH

www.vcresilience.com

Agrilogic www.agrilogicconsulting.com

Twinzo www.twinzo.eu

Eclipse Ditto www.eclipse.dev/ditto

Eclipse BaSyx www.eclipse.dev/basyx

iTwin.js www.itwinjs.org

SewerAI www.sewerai.com

Bosch IoT Suite www.bosch-iot-suite.com

Oracle IoT Production

Monitoring Cloud

www.docs.oracle.com/en/cloud/saas/iot-production-cloud

OpenSpace Group Ltd www.open-space.io

Process Genius Oy www.processgenius.ﬁ

Blackshark www.blackshark.ai

So, we learned that despite many DT tools being

available on the market, practitioners still use the con-

ventional programming languages (e.g., Python and

Java) to develop their DTs in-house from scratch.

Practitioners also face with challenges on such con-

cerns as the system integration, security, performance,

data quality, organisation, environment, development

platform, and abstraction. Despite many commer-

cial and open-source DT tools, no any classiﬁcation

framework is available for analysing and comparing

the DT tools. As discussed in Section 3, while the lit-

erature includes several analytical works about DTs,

none of them provide any lists of the existing DT

tools, any list of criteria for comparing the DT tools,

or any work that provide the comparison of the DT

tools for some speciﬁc requirements.

In this paper, we aim to propose a classiﬁcation

framework which describes the important features

that need to be considered for developing and using

DT systems. We believe that our classiﬁcation frame-

work will be useful for many stakeholders. Practi-

tioners can use the framework for analysing and com-

paring the DT tools with regard to their needs. Re-

searchers can identify the key features of any DT tools

and conduct relevant researches. Tool vendors can

identify the weakness and strengths of their DT tools.

Our classiﬁcation framework is depicted in Fig-

ure 1, which consists of twelve different features.

These are (i) domain, (ii) knowledge management,

(iii) system integration, (iv) quality assurance, (v)

reusability, (vi) development platform, (vii) deploy-

ment, (viii) type, (ix) extensibility, (x) abstraction, (xi)

enabling technology and (xii) maturity, and each fea-

ture is further divided into a set of sub-features. In the

rest of the paper, we initially introduce our methodol-

ogy for identifying the DT tool features and discussed

similar works in the literature. Then, we discuss each

tool feature separately in terms of their sub-features.

In the rest of the paper, we initially discuss our

methodology for the DT domain analysis. Then, we

discuss the similar studies in the literature. Next, we

introduce our classiﬁcation framework in terms of its

features and sub-features. Lastly, we discuss a case

study based on the analysis of a well-known DT tool

in terms of our classiﬁcation framework.

2 DOMAIN ANALYSIS

To come up with a set of features for classifying DT

tools, we performed a comprehensive domain analy-

sis. To this end, we ﬁrstly used google scholar and

searched for the papers that discuss (i) the deﬁni-

tions of DTs and their characteristics (Barricelli et al.,

2019; Singh et al., 2021; Mihai et al., 2022; Rasheed

et al., 2020; Jones et al., 2020; Tao et al., 2019; Hu

et al., 2021; Fuller et al., 2020), (ii) the enabling

DT technologies and any challenges in applying those

technologies (Michael et al., 2022; da Rocha et al.,

2022; Qi et al., 2021; Lv and Xie, 2022; TAO et al.,

2024; Newrzella et al., 2022; Shao and Helu, 2020),

(iii) the applications of DTs in different industries

(Ghaboura et al., 2023; Lin et al., 2022; Perno et al.,

2022; Biesinger et al., 2019; Purcell and Neubauer,

2023; Ali et al., 2023), and (iv) the analysis of DT

applications and technologies from different perspec-

tives (Gil et al., 2024; Liu et al., 2022; Alcaraz and

Lopez, 2022; Minerva et al., 2020; Wu et al., 2021;

Lo et al., 2021). Having identiﬁed the relevant pa-

pers, we went through each paper and identiﬁed any

attributes that are discussed in the paper and can be

considered as the DT tool features. A DT tool fea-

ture here can either be a commonly considered func-

tionality by the DT users, a crucial quality concern

for DT tools, a technique or technology needed for

developing DT tools, or some other requirement for

the effective use of the DT tools. All identiﬁed at-

tributes that represent the candidates for being the DT

tool features have been stored on an Excel ﬁle. We

created a separate sheet for each feature in the Ex-

cel ﬁle. We analysed each feature for determining the

sub-features that represent the main concern(s) about

that feature. To this end, we searched the literature

MODELSWARD 2025 - 13th International Conference on Model-Based Software and Systems Engineering

264

Figure 1: The taxonomy for DT tools.

and identiﬁed any well-cited papers that discuss the

feature-related techniques, approaches, and concerns.

By doing so, we were able to determine a set of sub-

features for each feature and stored those sub-features

in the relevant sheets of the Excel ﬁle.

After identifying the DT tool features by means

of the analysis of the relevant papers, we further used

the results of our recent industry survey

for under-

standing practitioners’ perspectives on the DT tech-

nologies. The results of our survey lead to many in-

teresting ﬁndings about the practitioners’ (i) adoption

of DT technologies, (ii) motivations for using and de-

veloping DTs, (iii) development experiences on DTs,

and (iv) their challenges from diverse perspectives.

The survey results revealed many interesting features

that practitioners are concerned with but was not so

explicit in our literature analysis (e.g., the reusability

and abstraction). So, we further extended the Excel

ﬁle with the newly identiﬁed features for the DT tools.

After documenting the features, we performed a

pilot study with one academic and two practitioners

who have considerable experiences on DTs. The aca-

demic and two practitioners have been provided with

the Excel ﬁle that includes the extracted features from

the papers and requested to provide a review based on

their knowledge and experience on DT systems. By

doing so, we were able to remove any redundancies

and inconsistencies, and add any missing features. So,

we ended up with a taxonomy of the DT tool features

that is depicted in Figure 1. Our taxonomy consists

of twelve different features each of which is decom-

posed into a cohesive set of sub-features.

3 RELATED WORK

The literature includes several papers that discuss DTs

from different perspectives, which includes the (i)

characterisations of DTs (e.g., (Barricelli et al., 2019;

VanDerHorn and Mahadevan, 2021; Jones et al.,

2020; Liu et al., 2021)), (ii) applications of DTs in

different domains (e.g., (Cimino et al., 2019) for man-

ufacturing, (Sun et al., 2023) for healthcare, (Ver-

douw et al., 2021) for farming, and (Opoku et al.,

2021) for construction), (iii) challenges that can be

faced with when using and developing DTs (Verdouw

et al., 2021), (iv) the analysis of DTs from particular

aspects (e.g., simulation (Boschert and Rosen, 2016),

fault monitoring (Boﬁll et al., 2023), and maintenance

(Errandonea et al., 2020)), and (v) case studies for di-

verse problems (e.g., (Mendi, 2022; Lu et al., 2020;

Peng et al., 2020)).

The classiﬁcation frameworks in the literature fo-

cus on, e.g., software evolution (Mens et al., 2003),

architecture description languages (Medvidovic and

Taylor, 2000), problem solving techniques in par-

ticular domains (e.g., the taxonomy for schedul-

ing algorithms for wireless mesh networks (Gabale

Towards a Classiﬁcation Framework for the Digital Twin Tools: A Taxonomy

265

et al., 2013)), component-based software engineering

(Crnkovic et al., 2011), software connectors (Mehta

et al., 2000) and software change impact (Lehn-

ert, 2011). The literature also includes classiﬁcation

framework studies about the software tools, such as

the classiﬁcations of software testing tools (Graham,

1991), static code analysis tools (Novak et al., 2010),

software fault monitoring tools (Delgado et al., 2004),

model transformation tools (Kahani et al., 2019), and

manufacturing scheduling tools (Dios and Framinan,

2016). However, no any effort has been made for clas-

sifying the DT tools. While some generic classiﬁca-

tion frameworks have been proposed for classifying

software tools in general such as (Firth et al., 2018),

those studies may not be so helpful due to lacking

support for the DT context.

4 DT TOOL FEATURES

In this section, we discuss each DT tool feature that is

depicted in Figure 1 in terms of its sub-features.

4.1 Domain

As indicated in our survey results

, DTs are used in

many industries for solving diverse problems. There-

fore, we consider the domain as an important fac-

tor for classifying the DT tools. The domain fea-

ture here is concerned with the problem domain for

which any DT tool is developed and used, and con-

sists of two sub-features - i.e., industry and domain-

speciﬁcity. While the industry sub-feature represents

the industry(ies) that a DT tool can support (e.g., man-

ufacturing, defense, and automotive), the domain-

speciﬁcity sub-feature represents the speciﬁc prob-

lem(s) that is(are) addressed in a particular industry.

For instance, (Fahim et al., 2022) shows the use of

predictive modeling in wind turbines, (Peker and Ak-

dur, 2019) shows the real-time simulation for signal

processing in the defense industry, and (Hodavand

et al., 2023) shows the fault detection and diagnosis

for the smart buildings in the construction industry.

4.2 Knowledge Management

Knowledge management for any system is to do with

managing the information and resources, which in-

cludes collecting, analysing information and trans-

forming them into knowledge that can be stored,

shared, and evolved (Despres and Chauvel, 1999).

Knowledge management for a DT tool consists of

several sub-features, which are (i) forum, (ii) mailing

list, (iii) video, (iv) tutorial, (v) certiﬁcation, (vi) white

paper, (vii) case study, and (viii) user guide. Forum

represents an online discussion platform for the DT

tool users to ask questions and get useful responses

from tool experts. Mailing list represents a collection

of e-mails that belong to the subscribed users for a

DT tool, and the DT tool vendor can send broadcast

e-mails for notifying the users about tool-related news

(e.g., version updates). Video represents any visual

presentations that introduce either the features of any

DT tool, usage patterns, user guide, or case-studies

which can be useful for the DT tool users. Tuto-

rial represents any visual presentation or textual doc-

ument that can teach users how to use the DT tool in

a series of stages. Certiﬁcation represents the conﬁr-

mation that can be granted by the tool vendors or any

contracted organisation for any DT tool user so as to

prove their abilities on the DT tool. White paper rep-

resents any report that discusses an innovation, issue

or technique about a DT tool from a scientiﬁc point

of view. Case study represents any document that dis-

cusses the evaluation of a DT tool on a particular case

or real-problem. User guide represents a document

that describes how to use a DT tool and its features.

4.3 Deployment

Deployment is concerned with the making a DT tool

available for the users so that users can use the DT

tool and perform their goals. Here, we consider two

deployment options (i.e., the sub-features), which are

cloud-based license and on-premise license. Cloud-

based license is based on the subscription method

where the users of a DT tool choose any package and

pay its price on a monthly or yearly basis so as to rent

the DT tool which is operated on cloud. On-premise

license prompts a DT tool software to be installed on

the local servers of customers through which users

can access the DT tool.

4.4 Type

Type for a DT tool is concerned with the functional-

ities that can be performed with the DT tool. There-

fore, types can be distinguished with their operations

on the DT model that represent the abstraction(s) over

the physical system. We consider ﬁve types of DTs,

which are visualisation DTs, prediction DTs, simu-

lation DTs, maintenance DTs, and DTs for interop-

erability (Liu et al., 2021; Piroumian, 2021). Note

that any DT tool may be of multiple types at the

same time. A visualisation DT focuses on visualis-

ing any model of the physical system in various nota-

tions (e.g., physical, mathematical, logical) which can

be updated dynamically with the data collected from

MODELSWARD 2025 - 13th International Conference on Model-Based Software and Systems Engineering

266

the physical system. A prediction DT uses AI tech-

niques (e.g., machine learning, deep learning, deci-

sion trees) and makes useful predictions on the mod-

els that represent abstractions over the physical sys-

tem. A simulation DT focuses on executing the mod-

els of the physical system and checking if certain sce-

narios are satisﬁed or not. A maintenance DT fo-

cuses on analysing the models of the physical sys-

tems for some quality properties so as to predict any

future failures and thus planing and scheduling the

system maintenance. A DT for interoperability fo-

cuses on managing the communication and coordi-

nation among a set of physical devices (e.g., sensors

and actuators) and the physical system in a way that

certain standards (e.g., security protocols, simulation

standards, communication protocols) are followed.

4.5 System Integration

System integration is concerned with the support for

developing a DT by means of integrating several het-

erogenous systems in a way that those systems can in-

teract and exchange data without any problems (Perno

et al., 2022). We consider here the system integration

in terms of the support for (i) integration, (ii) inter-

operability, and (iii) compatibility. Integration is con-

cerned with the support for (i) integrating with exter-

nal systems that can provide useful services and (ii)

developing DTs in terms of the composition of exist-

ing systems. Interoperability is concerned with the

support for enabling the heterogenous systems (e.g.,

sensors and actuators) to exchange data with a DT.

Compatibility is concerned with the support for in-

tegrating heterogenous systems that run on different

platforms and adopt different protocols of interac-

tions.

4.6 Quality Assurance

Quality assurance is concerned with the support for

checking the quality of a DT system that is developed

or used via a DT tool. Here, we consider 3 main qual-

ity properties that are important for the development

of high-quality DTs (Perno et al., 2022). These are (i)

security, (ii) data quality, and (iii) performance. Secu-

rity is concerned with the support for minimising any

security threats for the physical and software compo-

sitions of DTs that can cause issues in terms of the

availability, integrity and conﬁdentiality of system re-

sources (i.e., data and services) (Alcaraz and Lopez,

2022). Data quality is concerned with the support for

accessing the right data from the right data sources

in a way that meets the system quality requirements.

Performance is concerned with the support for the ef-

ﬁcient use of software and hardware resources so that

system data can be exchanged between the physical

and software systems with minimum latency.

4.7 Reusability

In our survey

, reusability (Prieto-Diaz, 1993) has

been observed as one of the top crucial challenges

for the practitioners who develop DTs. Reusability

is concerned with reducing the cost of designing, de-

veloping and operating DT systems and maximising

the quality by means of reusing the tested and proven

solutions and any existing applications. Here, we con-

sider 4 sub-features, which are design reuse, develop-

ment reuse, simulation reuse, and application reuse.

The design reuse sub-feature is concerned with the

support for the reusable design of DT systems and

thus making quality design decisions with the least

cost possible by means of, e.g., design patterns, ar-

chitecture styles, design languages, etc. The develop-

ment reuse sub-feature is concerned with the support

for the re-usable implementation of DT systems by

means of, e.g., programming frameworks, APIs, etc.

The simulation reuse sub-feature is concerned with

the support for the re-usable speciﬁcations of simu-

lation models that can be executed for the real-time

simulation by means of, e.g., model libraries, simula-

tion modeling languages, etc. The application reuse

sub-feature is concerned with the support for re-using

external applications (e.g., COTS) by means of soft-

ware connectors (or adaptors) that handle the adapta-

tion of external systems to the DT systems.

4.8 Development Platform

Some DT tools may provide a development platform

which provides interesting facilities for users and im-

proves the DT development processes by maximising

users’ productiviness and effectiveness. As indicated

in our survey

, practitioners use different DT devel-

opment platforms for diverse problem domains and

indeed face with challenges. Given our analysis of

our survey results, we consider here six different sub-

features for the development platforms, which are (i)

web access, (ii) no-code (or low-code) development

editor, (iii) versioning, (iv) multi-user collaboration,

(v) exporting, and (vi) code analysis. Web access en-

ables users to use the DT development platform over

internet without having to download and install any

tools. No-code (or low-code) development editors

provide graphical editors and enable users to specify

high-level visual models for the DT systems and gen-

erate fully (or partially) executable code. Versioning

enables managing the versions of the DT development

Towards a Classiﬁcation Framework for the Digital Twin Tools: A Taxonomy

267

artifacts (e.g., code) and performing such operations

as comparing, merging, and accessing previous ver-

sions. Multi-user collaboration enables multiple users

to work together on the development of the same DT

system and perform the design and coding activities

at the same time in a synchronised way. Automated

code analysis enables the users to check their DT de-

velopment code automatically and detect any bugs or

quality issues.

4.9 Extensibility

Extensibility is concerned with the ability of extend-

ing the software tools (e.g., adding new functionali-

ties or quality attributes) without breaking their core

architectures and violating the principal design deci-

sions (Kazman et al., 2022). DT tools are expected to

be extensible since the desired requirements can al-

ways change and the DT tools need to be adapted.

Otherwise, any changes lead to unmanageable costs

and cause the DT tools to be discontinued being used.

As revealed in our survey results

, many practition-

ers prefer to use conventional programming technolo-

gies (e.g., Python and Java) in developing DTs in

house despite many DT development platforms that

are available on the market.

We consider the DT tool extensibility in terms of

(i) white-box extensibility, (ii) glass-box extensibil-

ity, and (iii) black-box extensibility (Zenger, 2004).

White-box extensibility enables the DT source-code

to be changed by the users directly such as the open-

source software projects. Glass-box extensibility en-

ables the source-code to be accessed as read-only and

any extensions are performed on a separate copy of

the source-code from the original version. The inher-

itance and dynamic binding mechanisms of object-

oriented programming are the common examples of

glass-box extensibility. Black-box extensibility en-

ables users to extend a DT by means of interfaces that

abstract users from reaching the DT source-code and

prompt any extensions to be performed under certain

rules and constraints using such techniques as APIs,

plug-in development tools, model-driven engineering

tools.

4.10 Abstraction

Abstraction promotes the management of complexity

and solving complex problems by (i) suppressing ir-

relevant details and focusing on important aspects and

(ii) generalising the problems by separating the com-

mon parts from the varying parts (Kramer, 2007). As

DTs are considered as complex systems, abstraction

is a key skill and needs to be adopted throughout he

DT development processes. We consider here the fol-

lowing abstraction techniques that can be supported

by the DT tools, which are (i) model-driven engi-

neering, (ii) software product-line engineering, (iii)

modeling language, (iv) meta-modeling, (v) object-

oriented programming. Model-driven engineering

(MDE) (Schmidt, 2006) promotes the speciﬁcations

of high-level, abstract models of any DTs from differ-

ent viewpoints, the automated transformation of ab-

stract DT models into useful artifacts such as simula-

tor code, programming code, or any useful documen-

tation . Software product-line engineering (SPLE)

(Metzger and Pohl, 2014) promotes the development

of similar products (e.g., similar DTs for a particular

domain) using the minimum time and budget by sep-

arating commonalities of similar products from their

variations. SPLE can be applied in different stages

of the DT development including the analysis, design

and implementation, and maximises the reusable de-

velopment of DTs that derive from the same prod-

uct family which possess common and varied fea-

tures. Meta-modeling (Ozkaya and Akdur, 2021).

promotes the deﬁnition of DT modeling languages in

terms of abstract and concrete notation sets through

which DT models can be speciﬁed (e.g., (Ozkaya,

2024)). Lastly, object-oriented programming (OOP)

promotes the notion of classes that is the unit of ab-

straction which consists of a common set of methods

and attributes and can be specialised using the inheri-

tance mechanism (Stroustrup, 1987).

4.11 Enabling Technology

As aforementioned, DTs are complex systems that re-

quire the interactions of physical and software sys-

tems and perform complex operations including the

real-time data analysis, modeling and simulation,

visualisation, intelligent decision making, and AI-

supported diagnosis and prediction. As revealed

with our survey

, practitioners prefer several differ-

ent technologies for developing DTs. We consider

here the enabling technologies for DTs in terms of

ﬁve main features that represent the distinct categories

of the DT technologies which have been proposed by

Qi et al. (Qi et al., 2021). These are (i) cognising

and controlling the physical world, (ii) modeling, (iii)

data management, (iv) application services, and (v)

connection. The feature for cognising and controlling

the physical world is concerned with any technolo-

gies that enable gathering data from different phys-

ical devices (e.g., sensors and camera), transforming

data into knowledge, and controlling physical systems

using the transformed knowledge. Modeling is con-

cerned with any technologies that enable the model-

MODELSWARD 2025 - 13th International Conference on Model-Based Software and Systems Engineering

268

ing of a product that is monitored by a DT in terms

of different viewpoints including the 3D geometric

modeling of the product shapes, structural modeling

(e.g., the physical and logical structures), behaviour

and interaction modeling, and rule modeling. Data

management is concerned with any technologies that

enable the data management activities which are the

data collection, data transmission, data storage, data

processing, data fusion, and data visualisation. Ap-

plication services are concerned with the technologies

that enable monitoring, simulation, diagnosis and pre-

diction. Connection is concerned with any technolo-

gies that enable the communication, coordination and

complex interactions for the physical and virtual sys-

tems that compose DTs such as the technologies for

the secure data communications, human-computer in-

teraction, and standard communication interfaces.

4.12 Maturity

DT tools can vary depending on their level of matu-

rity which is concerned with the tool vendors’ support

for the tool evolution so as to promote the continu-

ous improvement of their tools with regard to chang-

ing technologies, user demands, rules and regulations

(Becker et al., 2009). We consider the DT tool ma-

turity in terms of ﬁve key sub-features, which are the

(i) version history (ii) ﬁrst release year, (iii) live sup-

port, (iv) technical support, and (v) company founda-

tion year. Version history is concerned with the num-

ber of tool versions that have been released so far and

any details about the versions such as the version re-

lease dates, the changes that are included in each ver-

sion, and any errors corrected. The ﬁrst release year is

concerned with the date on which the ﬁrst, initial ver-

sion of the DT tool has been released. The technical

support team is concerned with such facilities as live

chat, e-mail service, and help desk through which the

tool users can be supported in resolving any technical

issues that they face with while using the DT tools.

Lastly, company foundation year represents how long

the company has been working on the relevant ares

and their level of experience.

5 CASE STUDY

In this section, we demonstrate the use of our classi-

ﬁcation framework that is discussed in Section 4 for

AutoDesk Tandem

, which is one of the most well-

known development platforms for developing DTs.

To collect data about each feature and their sub-

AutoDesk Tandem: https://intandem.autodesk.com

features for our classiﬁcation framework, we used

AutoDesk Tandem’s free version and searched over

their comprehensive web-site.

Concerning the domain feature, AutoDesk Tan-

dem is speciﬁc for the buildings domain and promotes

for smarter buildings. AutoDesk Tandem is used by

the architecture, engineering, construction, and oper-

ations industry.

Concerning the knowledge management, Au-

toDesk Tandem supports community forums for the

users to share their ideas with each other. AutoDesk

Tandem provides several e-learning resources such as

online courses and tutorials, which can be categorised

depending on the DT software, industry and users’ job

positions. Moreover, AutoDesk Tandem provides cer-

tiﬁcations for the students and practitioners to prove

their skills on the use of AutoDesk Tandem for the

DT development. AutoDesk Tandem provides a com-

prehensive blog page through which several white pa-

pers and case studies have been published. Lastly,

AutoDesk provides several user guide documents for

different aspects of the DT development platform.

Concerning the deployment, AutoDesk Tandem

can only be used over a cloud platform and it is not

possible to install the DT development software on a

local machine.

Concerning the type, AutoDesk Tandem provides

a visualisation support for the 3D graphics modeling

of the buildings and simulation support for analysing

the building performance.

Concerning the system integration, AutoDesk

Tandem provides several facilities for managing the

integration, interoperability, and compatibility issues.

These facilities include streams, services, connectors,

converters, data mappers, ﬁlters, channels and gate-

ways, which can be used over a no-code development

platform without having to write any code.

Concerning the quality assurance, AutoDesk Tan-

dem provides an authentication API for the DT secu-

rity. AutoDesk Tandem provides a data dashboard for

managing the data quality such as checking the data

correctness and completeness for different elements

of the buildings. However, we could not reach any

resources for improving the performance of DT sys-

tems.

Concerning the reusability, AutoDesk Tandem

provides development reusability via the API libraries

and application reusability via the integration facili-

ties (e.g., plugins, connectors, etc.).

Concerning the development platform, AutoDesk

Tandem provides a web application for developing

and using DTs. No-code/low-code development plat-

forms are provided for the system integration facili-

ties. The past data used by DTs can be stored and

Towards a Classiﬁcation Framework for the Digital Twin Tools: A Taxonomy

269

accessed at any time, which can further be used for

reverting any changes on the DT systems. AutoDesk

Tandem enables multi-users to collaborate while de-

veloping and using DT systems. No any code analysis

support is provided.

Concerning the extensibility, AutoDesk Tandem

supports the black-box extensibility via their APIs

and plugin tools.

Concerning the abstraction, AutoDesk Tandem

supports model-driven engineering only. That is, 3D

building models can be speciﬁed, executed for sim-

ulation purposes and managed in terms of different

viewpoints.

Concerning the enabling technologies, AutoDesk

Tandem supports cognising and controlling the physi-

cal world thanks to its APIs and system integration fa-

cilities. AutoDesk Tandem also supports the 3D mod-

eling of buildings, management of data via its data

dashboard features, application services (i.e., moni-

toring, simulation and diagnosis), and the connection

technologies.

Concerning the maturity, AutoDesk Tandem’s ﬁrst

release has been announced on July 12, 2021 and sev-

eral releases have been published so far. AutoDesk

Tandem does not provide any live support. Technical

support is provided via the customer representatives

upon request. Lastly, the the AutoDesk company has

been founded in 1982.

6 CONCLUSION

Many DT tools are available on the market, which can

be used by the practitioners to obtain DT solutions

for their business problems. DT tools can either offer

a DT development platform (e.g., APIs, frameworks,

and libraries), any DT products that can be rented for

solving speciﬁc problems, or an open-source environ-

ment for extending an existing DT tool. Despite many

DT tools existing, practitioners develop their DTs in-

house using the conventional programming technolo-

gies. We believe that an important reason here is that

no any classiﬁcation framework for comparing the DT

tools have been proposed so far. Therefore, it is not

easy for the practitioners to determine the DT tools

available on the market, compare the tools among

each other, and choose the DT tool(s) that best ﬁt their

requirements.

In this paper, we aimed to propose a classiﬁcation

framework for the DT tools. To this end, we ﬁrstly

analysed the DT literature and further conducted a

practitioner survey among 131 practitioners from di-

verse industries. Performing a pilot study on our anal-

ysis results on the DT domain, we proposed a tax-

onomy which consists of twelve different features.

These are knowledge management, domain, deploy-

ment, system integration, type, extensibility, quality

assurance, abstraction, reusability, enabling technol-

ogy, development platform, and maturity. Each fea-

ture is further decomposed into a set of sub-features.

Moreover, we demonstrated our classiﬁcation frame-

work using one of the most well-known DT develop-

ment platforms called AutoDesk Tandem.

We are currently working on validating our clas-

siﬁcation framework with some well-known DT tools

besides AutoDesk Tandem. We will further conduct

a survey and interviews among the practitioners who

work for the defense industry and frequently use and

develop DTs so as to validate our classiﬁcation frame-

work. Besides, we will analyse 80 different DT tools

with regard to the features of our classiﬁcation frame-

work. By doing so, we aim to provide a really useful

guide that can be used by the practitioners for identi-

fying the existing DT tools, analysing the features of

interest, and choosing the DT tool(s) that best meet

their needs.

REFERENCES

Alcaraz, C. and Lopez, J. (2022). Digital twin: A compre-

hensive survey of security threats. IEEE Communica-

tions Surveys & Tutorials, 24(3):1475–1503.

Ali, W. A., Fanti, M. P., Roccotelli, M., and Ranieri,

L. (2023). A review of digital twin technology for

electric and autonomous vehicles. Applied Sciences,

13(10).

Barricelli, B. R., Casiraghi, E., and Fogli, D. (2019). A

survey on digital twin: Deﬁnitions, characteristics,

applications, and design implications. IEEE Access,

7:167653–167671.

Becker, J., Knackstedt, R., and P

oppelbuß, J. (2009). Devel-

oping maturity models for it management. Business &

Information Systems Engineering, 1(3):213–222.

Biesinger, F., Kraß, B., and Weyrich, M. (2019). A survey

on the necessity for a digital twin of production in the

automotive industry. In 2019 23rd International Con-

ference on Mechatronics Technology (ICMT), pages

1–8.

Boﬁll, J., Abisado, M., Villaverde, J., and Sampedro, G. A.

(2023). Exploring digital twin-based fault monitoring:

Challenges and opportunities. Sensors, 23(16).

Boschert, S. and Rosen, R. (2016). Digital Twin—The Sim-

ulation Aspect, pages 59–74. Springer International

Publishing, Cham.

Cimino, C., Negri, E., and Fumagalli, L. (2019). Review of

digital twin applications in manufacturing. Computers

in Industry, 113:103130.

Crnkovic, I., Sentilles, S., Vulgarakis, A., and Chaudron,

M. R. (2011). A classiﬁcation framework for software

MODELSWARD 2025 - 13th International Conference on Model-Based Software and Systems Engineering

270

component models. IEEE Transactions on Software

Engineering, 37(5):593–615.

da Rocha, H., Pereira, J., Abrishambaf, R., and Espir-

ito Santo, A. (2022). An interoperable digital twin

with the ieee 1451 standards. Sensors, 22(19).

Delgado, N., Gates, A., and Roach, S. (2004). A taxon-

omy and catalog of runtime software-fault monitoring

tools. IEEE Transactions on Software Engineering,

30(12):859–872.

Despres, C. and Chauvel, D. (1999). Knowledge man-

agement(s). Journal of Knowledge Management,

3(2):110–123.

Dios, M. and Framinan, J. M. (2016). A review and classi-

ﬁcation of computer-based manufacturing scheduling

tools. Computers & Industrial Engineering, 99:229–

249.

Errandonea, I., Beltr

an, S., and Arrizabalaga, S. (2020).

Digital twin for maintenance: A literature review.

Computers in Industry, 123:103316.

Fahim, M., Sharma, V., Cao, T. V., Canberk, B., and Duong,

T. Q. (2022). Machine learning-based digital twin for

predictive modeling in wind turbines. IEEE Access,

10:14184–14194.

Firth, R., Mosley, V., Pethia, R. D., Gold, L. R., and Wood,

W. (2018). A Guide to the Classiﬁcation and Assess-

ment of Software Engineering Tools.

Fuller, A., Fan, Z., Day, C., and Barlow, C. (2020). Digi-

tal twin: Enabling technologies, challenges and open

research. IEEE Access, 8:108952–108971.

Gabale, V., Raman, B., Dutta, P., and Kalyanraman, S.

(2013). A classiﬁcation framework for scheduling al-

gorithms in wireless mesh networks. IEEE Communi-

cations Surveys & Tutorials, 15(1):199–222.

Ghaboura, S., Ferdousi, R., Laamarti, F., Yang, C., and Sad-

dik, A. E. (2023). Digital twin for railway: A compre-

hensive survey. IEEE Access, 11:120237–120257.

Gil, S., Mikkelsen, P. H., Gomes, C., and Larsen, P. G.

(2024). Survey on open-source digital twin frame-

works–a case study approach. Software: Practice and

Experience, 54(6):929–960.

Graham, D. R. (1991). Software testing tools: A new clas-

siﬁcation scheme. Software Testing, Veriﬁcation and

Reliability, 1(3):17–34.

Hodavand, F., Ramaji, I. J., and Sadeghi, N. (2023). Dig-

ital twin for fault detection and diagnosis of building

operations: A systematic review. Buildings, 13(6).

Hu, W., Zhang, T., Deng, X., Liu, Z., and Tan, J. (2021).

Digital twin: a state-of-the-art review of its enabling

technologies, applications and challenges. Journal

of Intelligent Manufacturing and Special Equipment,

2(1):1–34.

Jones, D., Snider, C., Nassehi, A., Yon, J., and Hicks, B.

(2020). Characterising the digital twin: A systematic

literature review. CIRP Journal of Manufacturing Sci-

ence and Technology, 29:36–52.

Kahani, N., Bagherzadeh, M., Cordy, J. R., Dingel, J., and

Varr

o, D. (2019). Survey and classiﬁcation of model

transformation tools. Software & Systems Modeling,

18(4):2361–2397.

Kazman, R., Echeverr

ıa, S., and Ivers, J. (2022). Extensi-

bility. Technical Report CMU/SEI-2022-TR-002. Ac-

cessed: 2024-Oct-30.

Kramer, J. (2007). Is abstraction the key to computing?

Commun. ACM, 50(4):36–42.

Lehnert, S. (2011). A taxonomy for software change im-

pact analysis. In Proceedings of the 12th International

Workshop on Principles of Software Evolution and the

7th Annual ERCIM Workshop on Software Evolution,

IWPSE-EVOL ’11, page 41–50, New York, NY, USA.

Association for Computing Machinery.

Lin, Y., Chen, L., Ali, A., Nugent, C., Ian, C., Li, R., Gao,

D., Wang, H., Wang, Y., and Ning, H. (2022). Human

digital twin: A survey.

Liu, M., Fang, S., Dong, H., and Xu, C. (2021). Review of

digital twin about concepts, technologies, and indus-

trial applications. Journal of Manufacturing Systems,

58:346–361. Digital Twin towards Smart Manufactur-

ing and Industry 4.0.

Liu, Y. K., Ong, S. K., and Nee, A. Y. C. (2022). State-of-

the-art survey on digital twin implementations. Ad-

vances in Manufacturing, 10(1):1–23.

Lo, C., Chen, C., and Zhong, R. Y. (2021). A review of digi-

tal twin in product design and development. Advanced

Engineering Informatics, 48:101297.

Lu, Q., Parlikad, A. K., Woodall, P., Ranasinghe, G. D.,

Xie, X., Liang, Z., Konstantinou, E., Heaton, J., and

Schooling, J. (2020). Developing a digital twin at

building and city levels: Case study of west cam-

bridge campus. Journal of Management in Engineer-

ing, 36(3):05020004.

Lv, Z. and Xie, S. (2022). Artiﬁcial intelligence in the dig-

ital twins: State of the art, challenges, and future re-

search topics [version 2; peer review: 2 approved].

Digital Twin, 1(12).

Medvidovic, N. and Taylor, R. N. (2000). A classiﬁcation

and comparison framework for software architecture

description languages. IEEE Trans. Software Eng.,

26(1):70–93.

Mehta, N. R., Medvidovic, N., and Phadke, S. (2000). To-

wards a taxonomy of software connectors. In Pro-

ceedings of the 22nd International Conference on

Software Engineering, ICSE ’00, page 178–187, New

York, NY, USA. Association for Computing Machin-

ery.

Mendi, A. F. (2022). A digital twin case study on automo-

tive production line. Sensors, 22(18).

Mens, T., Buckley, J., Zenger, M., and Rashid, A. (2003).

Towards a taxonomy of software evolution.

Metzger, A. and Pohl, K. (2014). Software product line en-

gineering and variability management: achievements

and challenges. In Herbsleb, J. D. and Dwyer, M. B.,

editors, Proceedings of the on Future of Software En-

gineering, FOSE 2014, Hyderabad, India, May 31 -

June 7, 2014, pages 70–84. ACM.

Michael, J., Pfeiffer, J., Rumpe, B., and Wortmann,

A. (2022). Integration challenges for digital twin

systems-of-systems. In Proceedings of the 10th

IEEE/ACM International Workshop on Software Engi-

neering for Systems-of-Systems and Software Ecosys-

Towards a Classiﬁcation Framework for the Digital Twin Tools: A Taxonomy

271

tems, SESoS ’22, page 9–12, New York, NY, USA.

Association for Computing Machinery.

Mihai, S., Yaqoob, M., Hung, D. V., Davis, W., Towakel,

P., Raza, M., Karamanoglu, M., Barn, B., Shetve, D.,

Prasad, R. V., Venkataraman, H., Trestian, R., and

Nguyen, H. X. (2022). Digital twins: A survey on

enabling technologies, challenges, trends and future

prospects. IEEE Communications Surveys & Tutori-

als, 24(4):2255–2291.

Minerva, R., Lee, G. M., and Crespi, N. (2020). Digital

twin in the iot context: A survey on technical features,

scenarios, and architectural models. Proceedings of

the IEEE, 108(10):1785–1824.

Newrzella, S. R., Franklin, D. W., and Haider, S. (2022).

Three-dimension digital twin reference architecture

model for functionality, dependability, and life cy-

cle development across industries. IEEE Access,

10:95390–95410.

Novak, J., Krajnc, A., and

Zontar, R. (2010). Taxonomy of

static code analysis tools. In The 33rd International

Convention MIPRO, pages 418–422.

Opoku, D.-G. J., Perera, S., Osei-Kyei, R., and Rashidi, M.

(2021). Digital twin application in the construction

industry: A literature review. Journal of Building En-

gineering, 40:102726.

Ozkaya, M. (2024). Towards an architecture modeling lan-

guage for specifying digital twin architectures using

C4. In Fujita, H., Meana, H. M. P., and Hernandez-

Matamoros, A., editors, New Trends in Intelligent

Software Methodologies, Tools and Techniques - Pro-

ceedings of the 23rd International Conference on

New Trends in Intelligent Software Methodologies,

Tools and Techniques (SoMeT

24), Cancun, Mexico,

September 24-26, 2024, volume 389 of Frontiers in

Artiﬁcial Intelligence and Applications, pages 151–

164. IOS Press.

Ozkaya, M. and Akdur, D. (2021). What do practitioners

expect from the meta-modeling tools? A survey. J.

Comput. Lang., 63:101030.

Peker, O. and Akdur, D. (2019). A method for elimination

of false iff target reports by using isls and rsls tech-

niques. In 2019 Signal Processing Symposium (SP-

Sympo), pages 315–318.

Peng, Y., Zhang, M., Yu, F., Xu, J., and Gao, S. (2020). Dig-

ital twin hospital buildings: An exemplary case study

through continuous lifecycle integration. Advances in

Civil Engineering, 2020(1):8846667.

Perno, M., Hvam, L., and Haug, A. (2022). Implementation

of digital twins in the process industry: A systematic

literature review of enablers and barriers. Computers

in Industry, 134:103558.

Piroumian, V. (2021). Digital twins: Universal interoper-

ability for the digital age. Computer, 54(1):61–69.

Prieto-Diaz, R. (1993). Status report: software reusability.

IEEE Software, 10(3):61–66.

Purcell, W. and Neubauer, T. (2023). Digital twins in agri-

culture: A state-of-the-art review. Smart Agricultural

Technology, 3:100094.

Qi, Q., Tao, F., Hu, T., Anwer, N., Liu, A., Wei, Y., Wang,

L., and Nee, A. (2021). Enabling technologies and

tools for digital twin. Journal of Manufacturing Sys-

tems, 58:3–21. Digital Twin towards Smart Manufac-

turing and Industry 4.0.

Rasheed, A., San, O., and Kvamsdal, T. (2020). Digital

twin: Values, challenges and enablers from a model-

ing perspective. IEEE Access, 8:21980–22012.

Schmidt, D. C. (2006). Guest editor’s introduction: Model-

driven engineering. Computer, 39(2):25–31.

Shao, G. and Helu, M. (2020). Framework for a digital twin

in manufacturing: Scope and requirements. Manufac-

turing Letters, 24:105–107.

Singh, M., Fuenmayor, E., Hinchy, E. P., Qiao, Y., Murray,

N., and Devine, D. (2021). Digital twin: Origin to

future. Applied System Innovation, 4(2).

Singh, M., Srivastava, R., Fuenmayor, E., Kuts, V., Qiao,

Y., Murray, N., and Devine, D. (2022). Applications

of digital twin across industries: A review. Applied

Sciences, 12(11).

Stroustrup, B. (1987). What is “object-oriented program-

ming”? In B

ezivin, J., Hullot, J.-M., Cointe, P., and

Lieberman, H., editors, ECOOP’ 87 European Con-

ference on Object-Oriented Programming, pages 51–

70, Berlin, Heidelberg. Springer Berlin Heidelberg.

Sun, T., He, X., and Li, Z. (2023). Digital twin in

healthcare: Recent updates and challenges. DIGITAL

HEALTH, 9:20552076221149651.

TAO, F., SUN, X., CHENG, J., ZHU, Y., LIU, W., WANG,

Y., XU, H., HU, T., LIU, X., LIU, T., SUN, Z., XU, J.,

BAO, J., XIANG, F., and JIN, X. (2024). maketwin:

A reference architecture for digital twin software plat-

form. Chinese Journal of Aeronautics, 37(1):1–18.

Tao, F., Zhang, H., Liu, A., and Nee, A. Y. C. (2019). Dig-

ital twin in industry: State-of-the-art. IEEE Transac-

tions on Industrial Informatics, 15(4):2405–2415.

VanDerHorn, E. and Mahadevan, S. (2021). Digital twin:

Generalization, characterization and implementation.

Decision Support Systems, 145:113524.

Verdouw, C., Tekinerdogan, B., Beulens, A., and Wolfert, S.

(2021). Digital twins in smart farming. Agricultural

Systems, 189:103046.

Wu, Y., Zhang, K., and Zhang, Y. (2021). Digital twin net-

works: A survey. IEEE Internet of Things Journal,

8(18):13789–13804.

Zenger, M. (2004). Programming language abstractions for

extensible software components. PhD thesis, EPFL,

Lausanne.

MODELSWARD 2025 - 13th International Conference on Model-Based Software and Systems Engineering

272