SeGa4Biz: Model-Driven Framework for Developing Serious Games for

Business Processes

Faezeh Khorram, Masoumeh Taromirad and Raman Ramsin

Department of Computer Engineering, Sharif University of Technology, Tehran, Iran

Keywords:

Model-Driven Development, Model Transformation, Serious Game, Business Process.

Abstract:

Organizations look for effective ways to teach their business processes to their employees. The application

of serious games for teaching business processes is getting attraction recently. However, existing works are

by large business-speciﬁc and few of them aim at teaching business processes in general, besides that the

development of such games inherently suffers lack of precise and clear development approaches. This paper

presents SeGa4Biz, a model-driven framework for serious game development for teaching business processes.

Modeling supports different levels of abstraction and hence, increases user involvement throughout the de-

velopment. SeGa4Biz particularly provides metamodels for creating Educational Serious Games (ESG) and

Game-Aware Process (GAP) models, and automates considerable parts of the modeling and development ac-

tivities, via model transformation. The effectiveness and applicability of SeGa4Biz is examined through a

serious game development project in a software development company.

1 INTRODUCTION

A serious game is an interactive computer application

which has a challenging goal, is amusing to play, and

conveys practical skill, knowledge, or attitude to the

player (Cowan and Kapralos, 2017). Potential appli-

cability of serious games for teaching business pro-

cesses is getting attraction recently, since they can

help understand and analyze business processes (San-

torum, 2011). However, the relationship between the

elements of these two contexts is not clear-cut.

Non-technical domain experts have a crucial role

in the design of a serious game since they have knowl-

edge of the target serious domain. Therefore, effec-

tive communication between them and the technical

developers is required to ensure that all the serious ob-

jectives are realized by the game. This increases the

complexity of the serious game development in com-

parison to general game development. Existing game

development approaches are deﬁcient in this regard as

they typically do not balance game design with edu-

cational design (Matallaoui et al., 2015).

Model-driven development (MDD) is recently

used for serious game development as it, speciﬁ-

cally, provides complexity management and effec-

tive user involvement by providing different lev-

els of abstraction and good-level of automation

(e.g., (Thillainathan and Leimeister, 2016)). Exist-

ing MDD approaches, particularly in the context of

business processes, do not cover the essential as-

pects of modeling, such as precise deﬁnition of mod-

eling levels and transformation rules (e.g., (Bancora

et al., 2015; Tang and Hanneghan, 2010)). There are

also few MDD approaches that are domain-speciﬁc

(e.g., Van Broeckhoven and De Troyer (2013) in cyber

bullying) and, hence, are not applicable to the busi-

ness process domain.

This paper presents a model-driven approach,

called SeGa4Biz, for serious game development for

teaching business processes. Central to the proposal is

the deﬁnition of the modeling levels, covering differ-

ent features of the target domains; i.e., game develop-

ment and business processes. SeGa4Biz also provides

a set of transformation rules that supports automation

and so, reduces human intervention. Moreover, it in-

troduces a novel structured design method via speci-

fying how the elements of the game and the business

process domain would relate to each other.

SeGa4Biz is evaluated through a case study: its

prototype implementation is used for developing a

serious game in a company for teaching a business

process to new recruits. The experiment and its out-

come were assessed against a set of evaluation cri-

teria, deﬁned from the perspective of serious game

development in the context of teaching business pro-

cesses. The results show that, among other things,

Khorram, F., Taromirad, M. and Ramsin, R.

SeGa4Biz: Model-Driven Framework for Developing Serious Games for Business Processes.

DOI: 10.5220/0010198801390146

In Proceedings of the 9th International Conference on Model-Driven Engineering and Software Development (MODELSWARD 2021), pages 139-146

ISBN: 978-989-758-487-9

139

SeGa4Biz is applicable in a practical context and ad-

dresses many of the current shortcomings hindering

the use of serious games in this context.

The rest of this paper is structured as follows: Sec-

tion 2 provides an overview of SeGa4Biz ; modeling

levels and transformation rules are explained in Sec-

tion 3; Section 4 presents the evaluation results; an

outline of the related research is presented in Sec-

tion 5; and the paper concludes with a discussion of

the limitations and future work in Section 6.

2 OVERVIEW OF SeGa4Biz

In general, a game development process involves

three main phases: 1) Pre-production, focusing on

the preliminary design of the game and high-level

decisions, commonly documented in the Game Con-

cept Document (GCD); 2) Production, covering the

detailed design, development, and testing of the

game, commonly resulting in the Game Design Doc-

ument (GDD); and 3) Post-production, concerning

the deployment and acceptance testing of the game.

SeGa4Biz focuses on the detailed design of the game,

so covers the ﬁrst and parts of the second phases.

At the heart of SeGa4Biz is a modeling frame-

work that supports modeling the target business pro-

cess and the desired serious game at four levels of ab-

straction. It also provides a set of model transforma-

tions that supports semi-automatic construction of the

target models, describing the detailed design of the

game. As depicted in Figure 1, it speciﬁes the intra-

and inter-level relationships between the models. In

the next section, all the modeling levels are described

thoroughly.

Tool Support. A prototype implementation of the

proposed framework is implemented using ATL De-

velopment Tools (ADT), which are built on top of the

Eclipse Modeling Framework (Jouault et al., 2006).

All the code for SeGa4Biz, and also the case study’s

artefacts, are available online (khorram, 2020).

3 MODELING LEVELS

This section introduces the four modeling levels of the

proposed framework.

3.1 Level 1: Scope Modeling

During scope modeling, the goals and requirements

of the game are identiﬁed and modeled. This level in-

volves two main, yet totally different, domains: busi-

ness processes and serious games. Scope model-

Business Process Model

Game-Aware Process

Model (GAP)

Game Concept

Document (GCD)

Game Design

Document (GDD)

Level 1

Scope Modeling

Level 2

Structural and

Behavioral

Modeling

Game Class Diagram

Game Sequence Diagram

Level 3

State-based

Modeling

State-Machine

Level 4

Game

Engine-based

Modeling

Custom Finite State Machine

Modeling

Level

Model Document

(semi-) automatic

transformation information flow

Figure 1: SeGa4Biz framework.

ing includes three classes of artefacts which have to

be implemented manually: business process models,

game documents, and Game-Aware Process (GAP)

models (which deﬁne the relationships between the

two domains).

3.1.1 Business Process Modeling

The context of the target business process is elabo-

rated to identify the information that should be trans-

ferred to the player during the game.

Business Process Models mainly depict the pro-

cess activities, their order, and the roles involved.

They are modeled using BPMN as the employees of

organizations are typically familiar with it, and this al-

lows effective communication with them throughout

game development, particularly during requirements

elicitation and for receiving feedback.

A process may involve various roles that are not

necessarily interrelated; hence, each role needs to be

skillful in some parts of the process, instead of the

whole. Assuming that roles in the process model are

mapped to players in the game, each role requires its

speciﬁc game story (or stories), so that the ﬁnal game

meets the speciﬁc needs of individual roles. In this

context, we may create several partial process mod-

els by slicing the complete model with respect to a

particular role.

Moreover, a (partial) business process may con-

tain several decision points, and hence, represent dif-

ferent concrete paths. Throughout a game, a path may

be selected based on a players runtime decisions, au-

tomatically based on situational conditions, or even

MODELSWARD 2021 - 9th International Conference on Model-Driven Engineering and Software Development

140

randomly. In the random case, a separate BPMN

model must be created for each possible path to en-

sure that all the paths are covered in the game, and

the learning is therefore complete. Whereas in other

cases, the player or the system chooses the correct

path by examining the conditions, and thus no addi-

tional process model is required. Accordingly, a sin-

gle (partial) business process may be represented with

a set of, so-called, prime models; each corresponding

to only one simple path in that process.

Complementary Information. Certain informa-

tion is required for designing the game, but is not

covered by BPMN models. We have identiﬁed the

following essential parameters, which should be spec-

iﬁed for each (prime) process:

The process level indicates the importance and

complexity of the process in the organization, so de-

termines the difﬁculty of the game level correspond-

ing to the process.

The knowledge resources specify the available or

required resources for completing the process. The

information is either asked from the players during the

game to challenge their knowledge or given to them

to perform their tasks properly.

The required knowledge and skills explains the ca-

pabilities of each speciﬁc role to perform their tasks.

We map roles to game characters, and tasks to game

activities. Thus, this data is essential for the spec-

iﬁcation of the capabilities of game characters, the

conditions of progress through game levels, and the

rewarding mechanism.

The task weight indicates the importance of a task

in the process, and is required for calculating the task

completion score at a game level.

The level of a player character with respect to the

others is essential in deﬁning the rules for proceeding

in the game and designing the rewards for the game

levels. As mentioned above, each game player char-

acter corresponds to an organizational role, and thus,

its level is deﬁned based on its position in the organi-

zations roles hierarchy.

3.1.2 Serious Game Modeling

Game development activities start by preparing game-

speciﬁc documents. We work with textual documents

in early development stages, and in particular, pre-

scribe the use of a Game Concept Document (GCD)

as well as a Game Design Document (GDD). The

GCD speciﬁes the high-level requirements and con-

text of the target game, such as its premise, player

motivation (win and lose conditions), target mar-

ket, genre, target platform, license, risk analysis and

goals (Minaei, 2017).

The GDD contains the information required for ﬁ-

nalizing the pre-production phase and being ready for

the game production. This study jointly uses the tem-

plates presented in (Miles, 2016) and (Minaei, 2017),

and collects the following information: game story,

game characters (player and non-player), game au-

dio, game world, mechanics, UI, technologies, and

production plan. We suggest two formats for repre-

senting the information in the GDD: 1) Descriptive,

providing the information in natural language; and 2)

Logical, associating each GDD item with one or more

models (i.e., a model represents or complements the

information).

Game-aware Process Models (GAP). are intro-

duced to systematically specify the relationships be-

tween the elements of the business process (BPMN)

and the serious game, such that a serious game can

be designed for a given business process in a (semi-

)automatic and structured fashion. These models are

later used as the basis for model transformations.

To deﬁne the GAP metamodel, we ﬁrst introduced

a metamodel for Educational Serious Games(ESG)

(brieﬂy presented in Figure 2), and then we linked its

elements to those of BPMN, driven from OMG meta-

model (OMG, 2014). For instance, for each prime

process, a game level is deﬁned, in which the level

number and the needed score are determined based on

the process’s complementary information. Each lane

of the process model can be an organizational role, a

system, or an organizational unit, that are assigned to

different game elements (e.g., an organizational role

is a game character which could be player or non-

player). Tasks and gateways are respectively mapped

to states and knowledge challenges; they are assigned

to the player character deﬁned for the related role. An

excerpt of GAP metamodel is depicted in Figure 3.

We suggest a two-step process for creating a GAP

model from a BPMN model: 1) Specifying the basic

game elements that are directly linked to the BPMN

elements in the GAP metamodel (e.g., game chal-

lenges for the process gateways) 2) Deﬁning addi-

tional game elements linked to those of initially speci-

ﬁed (e.g., schoring rules for a game challenge deﬁned

in the ﬁrst step). While the two-step process facili-

tates modeling, it particularly provides iterative and

incremental game design.

3.2 Level 2: Structural and Behavioral

Modeling

The second modeling level of SeGa4Biz provides

the structural and behavioral aspects of the whole

SeGa4Biz: Model-Driven Framework for Developing Serious Games for Business Processes

141

Figure 2: An excerpt of ESG Metamodel.

Figure 3: An excerpt of GAP Metamodel.

game, through UML class and sequence diagrams,

respectively. Both diagrams are generated semi-

automatically by applying vertical model transforma-

tion to GAP models. The transformation rules imple-

mented for generating the sequence diagram, respect

those for generating class diagram, since system be-

havior indicates the run-time interaction between the

system’s structural elements. Transformation rules

are grouped with respect to the granularity and impor-

tance of the target element and hence, are not applied

all at once; a transformation is carried out in a se-

ries of consecutive steps: each step is applied on the

output of the previous step and augments the target

diagram. For example for the class diagram, assum-

ing that the target product follows the three-tier ar-

chitecture, the ﬁrst group of transformation rules pro-

duces Game Logic, Game Object, and Database com-

ponents, followed by rules that generate the related

classes under these components. Then, the game el-

ements of the GAP model (i.e., the input model), are

transformed into classes within the Game Logic and

Game Object components, while a new component is

generated that contains the classes corresponding to

the BPMN elements. The connections between the

generated classes are determined according to the re-

lations between their matching elements in the input

GAP model. It has to be noted that, this work involves

the use of combined fragments, the most complex

parts in sequence diagrams, which results in complex

and challenging transformation rules.

3.3 Level 3: State-based Modeling

The behavior of individual objects is modeled in the

third modeling level of SeGa4Biz using UML state

machines. State machines are generated by applying

a vertical transformation on the previously generated

sequence diagrams. Several research efforts have fo-

cused on this area. The earliest work introduces an

algorithm using OCL to resolve the conﬂicts and the

similarities between sequence diagrams (Whittle and

Schumann, 2000). Using this algorithm, an approach

is presented in (Graaf, 2007) to establish compatible

behavioral models, and thereby generate a set of state

machines from a set of sequence diagrams using ATL.

However, the main shortcoming is the lack of sup-

port for combined fragments and the extra complexity

caused by OCL constraints. In (Grønmo and Møller-

Pedersen, 2010), an algebraic graph transformation

method is used to transform sequence diagrams into

state machines; this method supports combined frag-

ments, but its implementation is incomplete and only

partial automation is provided.

Aiming to address these shortcomings, our pro-

posed model transformation particularly supports

combined fragments without using OCL. For each

lifeline of the sequence diagram, a distinct state ma-

chine is generated. Transformation rules are priori-

tized and then applied, from the highest to the low-

est, on the elements of each lifeline. Doing so, the

sequence of the states is derived from the source se-

quence diagram.

3.4 Level 4: Game Engine-based

Modeling

At the last level of modeling, the platform-speciﬁc

models are generated. In the context of this study, the

game engine is the most important (if not the only)

part of the platform. Due to Unity’s popularity, we

chose it along with its Playmaker (Miles, 2016) plu-

gin, which provides a powerful editor, debugging tool,

and run-time library. Playmaker automatically gener-

ates the game logic code from a set of custom Finite

State Machines (FSMs).

MODELSWARD 2021 - 9th International Conference on Model-Driven Engineering and Software Development

142

Hence, the last modeling activity aims at gener-

ating Playmaker FSMs. Playmaker FSMs and UML

state machines are similar in principle, the distinction

is in a set of executable actions and events, provided

by Playmaker, to facilitate the implementation of the

game object’s behaviors.

The models that are automatically generated in

this level are closer to the required format, yet are

not immediately supported by Playmaker. Mapping

the actions and events of the UML state machines to

the system actions and events in Playmaker requires

considering their semantics. Therefore, the generated

models are completed by the developers, w.r.t. the

available models and documents, to be imported into

Playmaker and used for the next development tasks.

4 EVALUATION

To assess the applicability and effectiveness of

SeGa4Biz, we conducted an evaluation that addresses

the following research questions: RQ1. How logical

and accurate are the modeling levels and model trans-

formations? RQ2. Does SeGa4Biz facilitate/ease the

serious game design and development tasks? RQ3.

Are the game-related concerns well covered?

4.1 Case Study

To answer the research questions, SeGa4Biz was ap-

plied to a case study, designed based on the guidelines

provided in (Runeson et al., 2012), which are spe-

ciﬁc to software engineering experiments. The case

was a real-world project in a software development

company that specializes in using and tailoring MDD

methodologies and developing software solutions for

medium to large businesses. The production manager

of their software product line was actively involved

in the study, both as the business expert and the cus-

tomer. SeGa4Biz was used to develop a serious game

for teaching the Leave-of-Absence Request process

to new recruits, so only those parts of the process that

relate them (i.e., employees) were considered.

One of the authors was involved in the project and

carried out the activities prescribed by SeGa4Biz, in

close collaboration with the product manager. The de-

velopment took about 30 working days, and all the

models and documents were delivered to the product

manager (as the business expert and customer), to ac-

quire his feedback and conﬁrmation. Data gathering,

analysis, and evaluation were performed iteratively

to achieve a ﬂexible design reﬂecting the experience

gained during the study.

4.2 Evaluation Criteria

In order to evaluate SeGa4Biz in a structured way,

particularly with respect to the research questions,

we introduced a set of evaluation criteria focusing

on the MDD features (Asadi and Ramsin, 2008), and

support for serious game development in the context

of business processes (Roungas and Dalpiaz, 2015;

Thillainathan, 2013); they are shown in Tables 1 and

2. Accordingly, the participants involved in the case

study were interviewed using these criteria, and the

outcome is shown in the last column of the Tables.

Also, the strengths and limitations of the proposed

framework were discussed with the participants for

future improvement.

4.3 Analysis of Results

The results are presented in order of the research

questions. Answering RQ1- The accuracy level of

the framework from the MDD perspective: The val-

ues given by business expert to the evaluation criteria

of Table 1 showed that the modeling levels are de-

ﬁned accurately and distinguishably, and all the re-

quired abstraction levels are provided. The main lim-

itations are the medium automation support and low

automatic code generation (i.e., particularly in gen-

erating Playmaker FSMs). SeGa4Biz has low sup-

port for features that are general to any MDD ap-

proach (e.g., round-trip engineering), herein, we have

focused on domain-speciﬁc features and have planned

to work on rest as future work.

Answering RQ2- SeGa4Biz ease of use: The case

study demonstrated that SeGa4Biz improves the un-

derstandability and quality, from the design and mod-

eling perspectives. These features along with the

provided automation, signiﬁcantly increases the ease

of use compared to similar approaches, which ulti-

mately, motivates companies to move towards using

serious games for teaching their business processes.

However, we were once again acknowledged that the

medium level of automation, especially at the last

modeling level, was the main concern.

Answering RQ3- The coverage level of game-

related concerns: It is evident from the values given to

the criteria of Table 2, that the game-related concerns,

especially those required for the context of business

process education, are well-satisﬁed by SeGa4Biz.

The main deﬁciency is the minimal attention given to

artistic features, that is however out of our scope.

SeGa4Biz: Model-Driven Framework for Developing Serious Games for Business Processes

143

Table 1: Evaluation criteria for assessing support for Model-Driven Development.

Criterion

Description of possible values

Result

Transparency between modeling levels

Boundary between levels: A: is accurately detectable, B: has relative transparency, C: cannot be distinguished.

Classification of the modeling level's data

A: Precise classification, B: Relative classification, C: Lack of classification

Support for abstraction levels

(CIM, PIM, PSM)

A: Full support for abstraction levels and transitions, B: All abstraction levels are defined, but transition

between them is not supported, C: Some abstraction levels are not supported.

Structural, Behavioral, Functional modeling

A: All the system's aspects are modeled, B: Some aspects of system are not modeled.

Model Transformation type

Vertical: Abstraction levels of the source and target models are different.

Horizontal: Source and target models are at the same level of abstraction.

Vertical

Automation level of the transformations

Low: Manual, Medium: Semi-automated, High: Fully-automated.

Medium

Automatic code generation

A: All parts of the code are generated automatically, B: Most parts of the code, C: Some parts of the code.

Tool support (for model validation, meta-

date management, automatic test, traceability

between models)

A: A complete toolset is provided, or precise guidelines are defined to select alternative tools.

B: A complete toolset is not provided, but general guidelines are defined to select alternative tools.

C: A specific tool, or guidelines to select an appropriate one is not defined.

Round-trip engineering

Synchronization of source & target models

Verification/ Validation

A: Detailed procedures are specified for the task in the methodology.

B: Only general guidelines are provided for the task.

C: The task is not covered by the methodology

5 RELATED WORK

Little research is done on serious games, whereas

gamiﬁcation is considered relatively more, especially

in the context of business processes. Despite the dif-

ferences between these two domains, we will provide

a discussion herein on the related efforts in both areas.

5.1 MDD Approaches for Serious Game

Development

In (Pﬂanzl and Vossen, 2018), a descriptive gamiﬁ-

cation modeling language called GaML is proposed,

based on which a model-driven architecture for the

design and development of serious games is intro-

duced in (L

ofﬂer et al., 2018). In (Roungas and

Dalpiaz, 2015), a web-based model-driven knowledge

management environment is introduced based on a

conceptual model. In (Thillainathan and Leimeister,

2016), an MDD framework consisting of a visual pro-

gramming environment (VIPEr), a domain-speciﬁc

modeling language (GLiSMo), and an MDD tool-

chain is introduced to enable non-technical people to

get involved in serious games development. In (Tang

and Hanneghan, 2010) an MDD framework for de-

velopment of learning serious games is provided, in-

troducing three models: Game Technology Model,

Game Content Mode, and Game Software Model.

In (Van Hoecke et al., 2015), an MDD framework

is deﬁned to build serious game production environ-

ments for non-technical users. In (Van Broeckhoven

and De Troyer, 2013), a graphical modeling language,

ATTAC-L, and an MDD framework are proposed for

building cyber-bullying games. In (Calder

on et al.,

2018) an approach and a graphical tool (MEdit4CEP-

Gam) is introduced that can be used by non-technical

users to design and model gamiﬁcation strategies that

can be automatically transformed into code.

Most of the above approaches are domain-speciﬁc

and are not appropriate for our intended domain.

Some are dependent on technical knowledge, whereas

in SeGa4Biz, different techniques are used to address

this shortcoming.

5.2 Approaches for Serious Game

Development and Gamiﬁcation for

Business Processes

In (Herzig et al., 2013), a serious game-based

method for business process management and a role-

playing game simulation tool are proposed to dis-

play, improve, and evolve existing business processes.

In (Matallaoui et al., 2015) a business process gam-

iﬁcation model (GameLog) for connecting game el-

ements to a business process is introduced. To im-

prove process sustainability, a gamiﬁcation tool is

presented in (Mancebo et al., 2017) that analyzes the

events of business process management systems and

encourages users to work more sustainably by us-

ing gamiﬁcation mechanisms. Kaleidoscope of Ef-

fective Gamiﬁcation is a gamiﬁcation design model

and an analysis tool (Kappen and Nacke, 2013), that

present guidelines for designing gamiﬁed commer-

cial applications based on the layers of the design

model. A novel conceptual framework for gamiﬁ-

cation design in collaborative and online work en-

vironments is deﬁned in (Rosmansyah et al., 2016).

PierSim is a 3D business process simulation environ-

ment for teaching the basics of BPM to students in

a gamiﬁed manner (Craven, 2015). The game ideas

that are applicable to business process modeling are

discussed in (Santorum, 2011) addressing issues like

low quality of models and low motivation of model-

ers. In (Klevers et al., 2016) a framework for anal-

ysis and classiﬁcation of requirements for the devel-

MODELSWARD 2021 - 9th International Conference on Model-Driven Engineering and Software Development

144

Table 2: Evaluation criteria for assessing support for serious game development in the context of business processes.

Criterion

Result

Status of SeGa4Biz based on criterion

Precise business process modeling

BPMN, the most popular standard for business process modeling is used.

Game elements customization for business processes

GAP model customizes the game elements for the target context.

Design of progress path for the player

General information of business processes is extracted for this purpose.

Player-centric game design

Game design task is defined based on considering employees as players.

Player interaction with the game

This feature is embedded in the framework.

Similarity of the game world to the real world

Game is designed based on the BPMN model (GAP model), which simulates the real world.

Game basis on precise rules

Game rules are defined based on business process rules

Art features such as Audio, Video, Animation

These features are considered, but they should be defined manually by art designers.

Educational goals elicitation

Guidelines are provided at the first modeling level.

Game goals elicitation

This task is performed along with educational goals elicitation

Mapping Between different goals

The mapping between the two categories of goals is defined relatively.

Resolution of the knowledge to be transferred to the

player throughout the game and how to do so

Realized by BPMN modeling and extracting complementary information; game design is based on

BPMN models, which helps find the position in the game to which knowledge can be transferred.

Definition of game levels based on the player's level

Player's level and level of the business processes that should be simulated are extracted.

Encouraging elements (e.g., scoring system, challenges)

Various rewards, scoring rules, and challenges are defined in the GAP metamodel.

Legend- A: Precise guidelines are prescribed for the feature; B: Importance of the feature is mentioned, but details are not specified; C: Feature is not embedded.

opment of a simulation game is proposed, aiming at

training business process changes in digital transfor-

mations. A combination of MDD, BPM techniques,

and gamiﬁcation mechanisms is introduced in (Ban-

cora et al., 2015) focusing on the impacts on indi-

vidual and social work management. In (Zribi et al.,

2016), a gamiﬁcation model for interactive learning

is described and applied on the PAd business process

simulator, an online model-based learning environ-

ment.

Most of the above research efforts focus on gam-

iﬁcation of business processes and hence, are not di-

rectly useful for serious game development. Whereas

our approach focuses on serious game development,

and GAP models providing precise mapping between

game elements and business processes.

6 CONCLUSION

In this paper, we introduced SeGa4Biz, a MDD

framework for developing serious games for teaching

business processes. It includes models at four levels

of abstraction. GAP metamodel is proposed in the

ﬁrst level to connect two different domains, business

process and serious game. For providing smooth tran-

sition between modeling levels as well as supporting

automation, we implemented transformation rules to

generate models demonstrating the structural and be-

havioral aspects of the ﬁnal product. Experimental

results showed that SeGa4Biz is effective and easy-

to-use by non-technical customers. As future work,

we plan to improve the medium degree of automation

at the last modeling level by supporting Playmaker

FSM semantics. We also aim to increase SeGa4Biz ’s

scalability by adding more complex game elements,

such as AI and networking features.

ACKNOWLEDGEMENTS

This project was partially funded by the European

Union’s Horizon 2020 research and innovation pro-

gram under the Marie Skłodowska Curie grant agree-

ment No 813884.

REFERENCES

Asadi, M. and Ramsin, R. (2008). Mda-based methodolo-

gies: an analytical survey. In European Conference on

Model Driven Architecture-Foundations and Applica-

tions, pages 419–431. Springer.

Bancora, M., Ripamonti, D., Vaccarella, A., and Brambilla,

M. (2015). Model-driven development and business

process modeling applied to personal productivity in

the consumer mobile app market. In 2015 2nd ACM

International Conference on Mobile Software Engi-

neering and Systems, pages 174–175. IEEE.

Calder

on, A., Boubeta-Puig, J., and Ruiz, M. (2018).

Medit4cep-gam: A model-driven approach for user-

friendly gamiﬁcation design, monitoring and code

generation in cep-based systems. Information and

Software Technology, 95:238–264.

Cowan, B. and Kapralos, B. (2017). An overview of serious

game engines and frameworks. In Recent Advances

in Technologies for Inclusive Well-Being, volume 119,

pages 15–38. Springer.

Craven, D. (2015). Gamiﬁcation in virtual worlds for learn-

ing: a case study of piersim for business education. In

Gamiﬁcation in education and business, pages 385–

401. Springer.

Graaf, B. (2007). Model-driven evolution of software archi-

tectures. In 11th European Conference on Software

SeGa4Biz: Model-Driven Framework for Developing Serious Games for Business Processes

145

Maintenance and Reengineering (CSMR’07), pages

357–360. IEEE.

Grønmo, R. and Møller-Pedersen, B. (2010). From se-

quence diagrams to state machines by graph transfor-

mation. In International Conference on Theory and

Practice of Model Transformations, pages 93–107.

Springer.

Herzig, P., Jugel, K., Momm, C., Ameling, M., and Schill,

A. (2013). Gaml-a modeling language for gamiﬁca-

tion. In 2013 IEEE/ACM 6th International Confer-

ence on Utility and Cloud Computing, pages 494–499.

IEEE.

Jouault, F., Allilaire, F., B

ezivin, J., Kurtev, I., and Val-

duriez, P. (2006). Atl: a qvt-like transformation lan-

guage. In Companion to the 21st ACM SIGPLAN

symposium on Object-oriented programming systems,

languages, and applications, pages 719–720.

Kappen, D. L. and Nacke, L. E. (2013). The kaleidoscope of

effective gamiﬁcation: deconstructing gamiﬁcation in

business applications. In Proceedings of the First In-

ternational Conference on Gameful Design, Research,

and Applications, pages 119–122.

khorram, f. (2020). Sega4biz: Model-driven

framework for developing serious games

for business processes. Mendeley Data.

http://dx.doi.org/10.17632/5gmhm9cc6d.3.

Klevers, M., Sailer, M., and G

unthner, W. A. (2016). Imple-

mentation model for the gamiﬁcation of business pro-

cesses: a study from the ﬁeld of material handling. In

Simulation and gaming in the network society, pages

173–184. Springer.

ofﬂer, A., Prifti, L., Knigge, M., Kienegger, H., and Krc-

mar, H. (2018). Teaching business process change in

the context of the digital transformation: A review on

requirements for a simulation game. Multikonferenz

Wirtschaftsinformatik (MKWI).

Mancebo, J., Garcia, F., Pedreira, O., and Moraga, M. A.

(2017). Bpms-game: tool for business process gamiﬁ-

cation. In International Conference on Business Pro-

cess Management, pages 127–140. Springer.

Matallaoui, A., Herzig, P., and Zarnekow, R. (2015).

Model-driven serious game development integration

of the gamiﬁcation modeling language gaml with

unity. In 2015 48th Hawaii International Conference

on System Sciences, pages 643–651. IEEE.

Miles, J. (2016). Unity 3D and playmaker essentials: game

development from concept to publishing. CRC Press.

Minaei, B. (2017). Preproduction: From concept to pro-

posal. University Lecture Notes. Iran University of

Science and Technology.

OMG (2014). Uml proﬁle for bpmn processes.

Pﬂanzl, N. and Vossen, G. (2018). What do business pro-

cess modelling and super mario bros. have in com-

mon? a games-perspective on business process mod-

elling. Enterprise Modelling and Information Systems

Architectures (EMISAJ), 13:69–76.

Rosmansyah, Y. et al. (2016). Gamiﬁcation framework for

designing online training and collaborative working

system in statistics indonesia. In 2016 International

Conference on Information Technology Systems and

Innovation (ICITSI), pages 1–6. IEEE.

Roungas, B. and Dalpiaz, F. (2015). A model-driven frame-

work for educational game design. In International

Conference on Games and Learning Alliance, pages

1–11. Springer.

Runeson, P., Host, M., Rainer, A., and Regnell, B. (2012).

Case study research in software engineering: Guide-

lines and examples. John Wiley & Sons.

Santorum, M. (2011). A serious game based method for

business process management. In 2011 Fifth Interna-

tional Conference on Research Challenges in Infor-

mation Science, pages 1–12. IEEE.

Tang, S. and Hanneghan, M. (2010). A model-driven frame-

work to support development of serious games for

game-based learning. In 2010 Developments in E-

systems Engineering, pages 95–100. IEEE.

Thillainathan, N. (2013). A model driven development

framework for serious games. Available at SSRN

2475410.

Thillainathan, N. and Leimeister, J. M. (2016). Educators as

game developersmodel-driven visual programming of

serious games. In Knowledge, Information and Cre-

ativity Support Systems, pages 335–349. Springer.

Van Broeckhoven, F. and De Troyer, O. (2013). Attac-l:

A modeling language for educational virtual scenar-

ios in the context of preventing cyber bullying. In

2013 IEEE 2nd International Conference on Serious

Games and Applications for Health (SeGAH), pages

1–8. IEEE.

Van Hoecke, S., Samyn, K., Deglorie, G., Janssens, O.,

Lambert, P., and Van de Walle, R. (2015). Enabling

control of 3d visuals, scenarios and non-linear game-

play in serious game development through model-

driven authoring. In International Conference on Se-

rious Games, Interaction, and Simulation, pages 103–

110. Springer.

Whittle, J. and Schumann, J. (2000). Generating state-

chart designs from scenarios. In Proceedings of the

22nd international conference on Software engineer-

ing, pages 314–323.

Zribi, S., Jorquera, T., and Lorr

e, J.-P. (2016). To-

wards a ﬂexible gamiﬁcation model for an interop-

erable e-learning business process simulation plat-

form. In Enterprise Interoperability VII, pages 283–

291. Springer.

MODELSWARD 2021 - 9th International Conference on Model-Driven Engineering and Software Development

146