Robotic Process Automation for the Gaming Industry

Ciprian Paduraru, Adelina-Nicoleta Staicu and Alin Stefanescu

Department of Computer Science, University of Bucharest, Romania

Keywords:

RPA, Automation, Software Robot, Gaming Industry, Artiﬁcial Intelligence.

Abstract:

Robotic Process Automation has recently been used in many ﬁelds to automate business-oriented processes.

Industries such as ﬁnance, transportation, and retail report signiﬁcant return on investment (ROI) after replac-

ing redundant, repetitive, and error-prone work performed by human workers with RPA software agents. In

our research, we found that there is a great opportunity to use RPA to automate processes in the game devel-

opment and support industry. In this paper, we identify some of these opportunities and propose automation

domains, examples, and high-level blueprints that may be implemented and extended by both academia and

the game development industry. The requirements, missing gaps, development ideas, and prototyping work

were done in collaboration with local game development partners. Our empirical evaluation shows that the

identiﬁed automation capabilities can play an important role in automating various processes needed by the

game development industry in the future.

1 INTRODUCTION

Robotic Process Automation (RPA) is one of the

trending technologies used today to accelerate au-

tomation in the business world (Aguirre and Ro-

driguez, 2017), (Leno et al., 2021b). The use of artiﬁ-

cial intelligence (AI) and machine learning (ML) (Jha

et al., 2021) has made the adoption of these technolo-

gies in many domains and business applications very

attractive. In many cases, repetitive, redundant, and

error-prone tasks are performed by humans, such as

manually checking invoices, generating reports, col-

lecting data needed for internal systems, and routing

that data to other components, etc. All of these tasks

can be replaced by RPA by having software agents

perform these activities instead of humans. In this pa-

per, we refer to these software agents as RPA robots,

as is also common in the literature. Overall, the ap-

plication of RPA in business is to free humans from

unnecessary tasks by having RPA robots mimic iden-

tical human actions. In turn, the automation method

ensures greater productivity and accuracy of results,

and even boosts employee morale by allowing them

to perform more enjoyable work.

In addition to these beneﬁts, the adoption of RPA

is highly sustainable because it is a non-invasive sys-

tem, meaning it can be easily integrated into existing

systems and infrastructures. RPA robot workﬂows are

deﬁned in the form of a graph in an authoring tool

(e.g., UiPath Studio

). Authoring can be done by

both technical and non-technical stakeholders, which

is also an important feature as these business-oriented

processes can be deﬁned and maintained on a larger

scale. The operations deﬁned in the activity nodes

and links can call internal or external source code or

other building blocks. Thus, the system is extensi-

ble and can reuse existing work deﬁned by other sys-

tems. Speciﬁcally, a node in the workﬂow deﬁned

for an RPA robot can respond to user messages with

GPT-3-based services (Brown et al., 2020) or provide

resource scheduling with automated invocation of ex-

ternal services for a deep learning-based method with

available open source code such as (Aljunid and Dh,

2020).

As we have discussed with our industry partners,

the gaming industry (one of the most valuable sec-

tors of the entertainment industry) lacks automation

on several levels.

The main contribution of this paper is to research

areas in the gaming industry’s speciﬁc processes that

can be automated through RPA technology by using

intelligent robots to replace redundant human labor.

The software robots used in our methods can commu-

nicate via edge and cloud operations and use AI/ML

tools to achieve various proposed goals. To the au-

thors’ understanding, this is the ﬁrst work that ad-

dresses the issue of combining RPA with game devel-

https://www.uipath.com/product/studio

Paduraru, C., Staicu, A. and Stefanescu, A.

Robotic Process Automation for the Gaming Industry.

DOI: 10.5220/0012075700003538

In Proceedings of the 18th International Conference on Software Technologies (ICSOFT 2023), pages 37-45

ISBN: 978-989-758-665-1; ISSN: 2184-2833

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

opment and support. In short, we divide our research

and proposed automation improvement into several

sub-processes:

• Development Process: Automate the workﬂows

performed by stakeholders in creating and deploy-

ing the product itself.

• User satisfaction and retention background pro-

cesses: Help automate and respond to speciﬁc

user needs, retain users through communication

and appropriate promotions.

• Quality Assurance: Increase source code quality,

deploy, ﬁnd and mitigate bugs while incorporating

more human labor.

• Secure transactions and fraud prevention: Proto-

type RPA workﬂows that leverage other technolo-

gies to highlight potential leaks or abnormal user

behavior.

• Scalability and dynamic infrastructure for ser-

vices: Identify how RPA robots can be used in

migrating, creating, or shutting down services to

smooth and optimize regional spikes in the num-

ber of users connecting online to a game while re-

ducing unnecessary costs of keeping servers run-

ning.

At the technical level, our work also describes the

prototype of the proposed areas of automation for one

of the most common publicly available game engines,

Unreal Engine 5

, with RPA and game process au-

tomation.

For this paper, we also collaborated with local in-

dustry partners to better understand the repetitive pro-

cesses performed by people in developing games, de-

ploying infrastructure, and providing ongoing post-

launch support for titles, and then link this accumu-

lated knowledge to the ﬁeld of RPA and other indus-

tries that are already beneﬁting from it.

The remainder of the paper is organised as fol-

lows. Section 2 describes the literature that inspired

us in identifying use cases and implementing proto-

types. The description of the identiﬁed use cases and

areas of automation used by our prototype for im-

plementation are outlined in Section 3. Furthermore,

a preliminary evaluation of the proposed methods is

presented in Section 4. Finally, in the last section, our

conclusions and a plan for future work are presented.

https://www.unrealengine.com/en-US/

unreal-engine-5

2 RELATED WORK

As this is, to the authors’ understanding, the ﬁrst work

to discuss the use of RPA to automate various game

development and support services, this section pro-

vides an overview of the current state of the art in ap-

plications that combine the RPA and gaming domains,

as well as applications of RPA in other domains that

have inspired our automation use cases and prototype

development.

The work in (Andrade, 2022) focuses on using

automation for continuous testing of game services

in cloud platforms. We reuse and extend their work

in our automation examples for game testing. In

(Yokoyama et al., 2020), the authors propose using

RPA to run various games with the goal of teach-

ing students programming languages such as Python.

They show that automating the playing of mini-games

can increase student motivation. Another interesting

application for our goals is the work in (Qasrawi et al.,

2020), which automates the processes of statistical

data collection and evaluation of feedback from indi-

viduals in the context of serious games and education.

One of the most common use cases of RPA com-

bined with AI is the automatic processing of business

documents without human supervision (Ling et al.,

2020). Software robots are able to automatically ex-

tract data from PDF documents, credit cards, and ID

photos with little or no human guidance. In addition,

they can copy data into a machine-understandable for-

mat that can be visualized by humans to easily ap-

prove or update the content, rather than performing

the error-prone task of manually copying each ﬁeld

(e.g., copying data from a PDF ﬁle into a spreadsheet

format). This automated workﬂow can also be ap-

plied to the gaming industry, especially today where

subscriptions to services are a major trend and users’

personal documents can give access to special offers

and discounts (e.g., as a student or from an academic

institution).

RPA has also been used in human resources (HR)

(Jeeva Padmini et al., 2021) to perform web crawling

of resources such as LinkedIn

and then select poten-

tial candidates for a job. Again, a combination with

AI methods, particularly NLP and sentiment analy-

sis, is integrated into the software robot workﬂows.

The technology and methods behind this can also be

applied to the gaming industry, as users’ needs and

views of the products they see and what they want

can be determined through external web forums that

contain discussions and ratings on the topics of inter-

est.

Applications of software robots in cybersecurity

https://linkedin.com

ICSOFT 2023 - 18th International Conference on Software Technologies

are discussed in (Tyagi et al., 2021). Malware de-

tection and fraud prevention is a common problem in

online gaming, so combining RPA, AI-based methods

and blockchain can help the industry through simpli-

ﬁed and reusable automation workﬂows.

An evaluation of chatbots for business process au-

tomation is presented in (Dan et al., 2022). The suc-

cess of integrating chatbots into robotic workﬂows is

also a valuable idea for the gaming industry, as the

user is often blocked or needs general help to under-

stand the game mechanics and systems or to continue

their progress in the environment. Thus, an automated

chatbot driven by tailored game-speciﬁc knowledge

can reduce the effort of staff hired to respond to user

messages while increasing response time to achieve

greater user satisfaction.

3 AREAS OF AUTOMATION AND

CONCEPTS

In this section, we present the ideas and proposed au-

tomation domains and concepts that we have explored

and implemented as prototypes for connecting RPA,

game development and support pipelines. Note that

the methods used integrate or hierarchically lever-

age other known methods for separation of concerns,

such as AI, ML, and blockchain technologies. Our

proposed implementation methods are non-invasive,

meaning that RPA robots can be optionally deployed

in the proposed use cases and solutions in addition

to the developer’s existing infrastructure. This can

ensure that adoption of the methods and the proto-

type can be a smooth process, as many developers in

the gaming industry reuse components from the previ-

ous source code and infrastructure that were available

in previously developed projects. We further break

down this section by area of interest and present spe-

ciﬁc use cases and areas for automation, along with

the concepts used to implement them as supported by

our internal prototypes.

3.1 Quality Assurance

The quality assurance process of a game product is

mainly performed by humans. This has serious impli-

cations, as literature research has shown (Politowski

et al., 2021). First, this can become a costly process

on the part of the developer. Second, constant updates

to released products are required, and because human

processes can be error-prone in the testing process,

new releases can often have undiscovered problems

that can lead to important data leaks or user frustra-

tion. However, there are several AI-based methods

studied in the literature that can be used in this regard

to support automated product testing (Paduraru et al.,

2022). In addition, developers typically create their

own set of scripts for setting environments, actions,

and expected results that look like typical functional

tests. Using RPA software agents, products can be

tested in the background without a human having to

worry about different pipelines involved. As a con-

crete example of such a pipeline, ﬁrst a speciﬁc ver-

sion of the source code and data from the repository

needs to be downloaded, building it on the available

machines, assigning a deployment target (e.g., game

console, smart TV, smartphone, PC, etc.) from the

local server/cloud infrastructure resource pool, and ﬁ-

nally automatically reporting the results to the devel-

oper’s own issue tracking infrastructure, Figure 1.

Another problem addressed in other industries and

solved with RPA is the concept of know your cus-

tomer (KYC) (Vijai et al., 2020), which can also be

applied to gaming products. Fraud detection with ML

can be used by robots to analyze and report abnor-

mal user behavior in the background. Then a human

can further evaluate the suggestion and allow or re-

ject it. Overall, however, this automated detection can

minimize the human effort required for analysis, as

today’s industry mostly performs this process man-

ually through sampling or textual reports from other

users. Securing transactions also falls into this area,

and robots can leverage blockchain solutions such as

the work presented in (Paduraru et al., 2023). These

automated processes are illustrated in the workﬂow

shown in Figure 2.

3.2 Development Process

Many stakeholders intensively involved in the devel-

opment of a game product must perform many redun-

dant or repetitive operations that can be tedious, frus-

trating, and often error-prone (Aleem et al., 2016).

In many cases, code submitted by software engi-

neers requires additional post-script builds to convert

the code and/or assets into formats that can be used

by target platforms (e.g., shader source code used in

the rendering components of game consoles or smart-

phones). The work of artists and animators is even

more tendentious: Generally, they work with third-

party tools (e.g., Maya

, Blender

, Adobe Photoshop

) to create or modify existing graphical assets for the

game product. However, after creating and submit-

ting the assets in their raw format, they must re-import

them or convert them to a custom format usable for

https://www.autodesk.com/products/maya

https://www.blender.org

https://www.adobe.com/products/photoshop.html

Robotic Process Automation for the Gaming Industry

RPA robot for Testing

Developer's infrastructure

Issues detected

database

Choose a machine from

the pool to build the

code and test the game

Source code

and data repository

Get latest version (or

a specific one) from

repository and build

the game

Test the build using

different methods, i.e.,

scripts, AI agents, etc.

Report issues:

performance,

stress/loading tests,

bugs, etc.,

Set of machines for building

(local, servers, cloud resources)

......

Set of target test

machines

Select

Retrieve

Select Upload

Figure 1: The main workﬂow of the RPA robot when testing a game product in the development/update phase, from start

to ﬁnish. Note that the robot’s work is non-invasive and decoupled from the developer’s infrastructure by proxy messages.

Multiple instances of this workﬂow would normally run in the cloud or on local servers.

Users playing the

game

Developers side

human users

Collect samples

from real-users

Analyze behavioral samples using

classification models and data

Read trained

models and data

Real-time

monitoring

and reporting

Suggest

potential

abnormal

behaviors

Collect set of

normal vs

abnormal

samples from

trusted human

users.

Continuous training of

behavior classification

models

Store trained

models and data

Figure 2: Two different workﬂows that can be automated with RPA software robots. Essentially, the ﬁrst application (work-

ﬂow colored in blue) shows how samples of human behavior can be collected in real time, then evaluated through trained

classiﬁcation methods, and ﬁnally reports generated in real time to help humans assigned to developers evaluate fraudulent

or unusual transactional behavior. The second workﬂow (colored green) shows how training of production-ready classiﬁers is

performed. Continuous training is required because game logic, parameters, and assets typically change constantly after the

initial release.

the game and deployment target, and then resubmit

the changes to the repository. At the end of these op-

erations, all stakeholders typically need to schedule

a series of evaluation processes to ensure that basic

functionality is maintained.

We have found that these workﬂows can be auto-

mated by RPA robots, as shown in Figure 3, allowing

stakeholders to focus on their assigned main task in-

stead of manually performing the repetitive and error-

prone processes observed.

3.3 Scalability and Infrastructure

Common games developed today are mainly played

online, with a large increase in the number of users

depending on external events in certain regions of the

world (e.g., World Cup events (Kim et al., 2015)) or

in certain seasons (e.g., weekends, holidays, etc.). Re-

lated to this aspect, one of the main challenges in the

gaming industry is the need to dynamically adapt the

infrastructure to sustain these situations and provide

the required quality of service (QoS) to the connected

users at all times. According to our research, this

work is mostly done manually by humans who ana-

lyze the number of users in relation to QoS and then

set up new servers or shut down existing ones. We

found that this is another potential use case for lever-

aging the automation capabilities of RPA software

robots. A pool of available resources and the reuse of

AI and ML methods to perform predictions (Mart

ınez

et al., 2022) for predicting the required resources us-

ing time series-based methods and then scheduling

ICSOFT 2023 - 18th International Conference on Software Technologies

RPA robots

automation

Developers' side work

Software engineers Animators ActorArtists and designers

Source code

and data repository

Submit changes

Changes

need additional

data build ?

Yes

Build/Convert

assets to

game format

Convert

data ,

re-submit

the new

assets

Run end-to-end test processes

(using RPA robots is desired)

Figure 3: The diagram shows the division of effort between

the developer side and the automation that can be performed

by the RPA robots in the development phase with respect to

code submission and the required continuous testing pro-

cesses. First and foremost, RPA software robots can sig-

niﬁcantly reduce the cost of human users to deal with re-

importing assets, using third-party tools, and resubmitting

processes. They can also ensure that proposed changes are

effectively evaluated at a predeﬁned rate.

the required resources in advance can efﬁciently scale

these processes automatically and with less human ef-

fort. At a minimum, RPA can be used in this context

to: (a) suggest scaling up or down servers in different

regions of the world, Figure 4, and (b) display contex-

tual heatmaps of real-time usage and forecasts using

continuously trained models, Figure 5. Then, human

users on the developer side can correlate the sugges-

tions made by the robots with the real-time analyzes

and heatmaps, and eventually easily approve, reject,

or adjust the proposed changes.

3.4 User Satisfaction and Retention

User satisfaction is a very important issue for busi-

ness revenue in the game industry. The game product

Deployed infrastructure

Cluster 1

Cluster 2

Cluster 3

connected to

migrate

provision

resources

add back

resources

Dynamic real-time

adjustments proposal

from RPA robots

Figure 4: Examples of dynamic scaling suggestions given

by an RPA robot. The agent can automatically propose or

self-provision a new server to meet the promised QoS for

a new group of users in an isolated area (green cluster),

and simultaneously migrate a small group of users from

one server to another (red cluster) if resources can be saved

while still meeting the QoS.

itself must be enjoyable, rewarding, understandable,

and provide live support to engage users and remove

blockages (Fu et al., 2017). Typically, companies hire

employees with titles known as community managers,

coordinators, or community moderators to manage

live support, review statistics, and provide some sort

of individual reward to users (Karabinus and Ather-

ton, 2018). Based on our ﬁndings, some of this work

can be replaced by RPA robots. Overall, we conclude

that aspects such as loyalty beneﬁts, promotions, re-

moval of live in-game blockers, and discussions with

users can be fully or partially automated without hu-

man effort.

On a technical level, RPA robots can perform

background data mining (Ketkar and Gawade, 2021),

make decisions and provide recommendations in the

background using common ML-based recommender

systems such as (Ribeiro et al., 2021). These recom-

mendations can come from different perspectives de-

pending on the context. For example, an important is-

sue is an RPA robot that analyses the user’s activities

with the environment in the background and detects

when there are problems in understanding the game

Robotic Process Automation for the Gaming Industry

Analyze current

infrastructure and real-

time usage

Geographically

deployed

servers and

connected

users data

Build a heatmap of

features to encapsulate

users' QoS and

resources used

Record features

Data store of real-time

analytics and models

Continuous training of

forecasting models

Predict servers

provisioning as needed or

suggest turning-off

existing ones

Use trained models

Figure 5: RPA software robots can analyze resource utilization in real time and then create a heatmap of characteristics that

contain the information needed to understand the quality of services and how well they are being delivered compared to what

was promised. In parallel, this heat map can be used to record, store, and re-train the predictive models while providing

suggestions for scaling up or down the existing infrastructure with new servers or other types of resources.

mechanics and procedures or when the game progress

is blocked. The RPA robot can then adapt to the situa-

tion and provide live feedback in the form of a tutorial

(Paduraru. et al., 2022) or a chatbot (Rosmalen et al.,

2012), (Ouyang et al., 2022), to help the user and

keep them engaged. To keep users satisﬁed and feel-

ing like they are making progress while being appro-

priately rewarded, robots can beneﬁt from data col-

lection performed in the background to award game

points, items, or other game-speciﬁc objects that they

enjoy interacting with. These processes are outlined

in Figure 6.

One of the problems reported by users is also that

the game is often constantly updated or takes too long

to download due to the large amount of memory re-

quired. RPA robots can solve this problem by running

on users’ devices where the game is installed either on

smartphones, game consoles or PCs, and then updat-

ing in the background by downloading things in ad-

vance so that the user does not have to wait for these

processes to complete. Robots can be programmed

to perform all of these processes to the satisfaction of

the user and keep them updated at all times, Figure 7.

4 EVALUATION AND

TECHNICAL

IMPLEMENTATION

An in-depth evaluation of the implemented prototypes

and automation concepts would require quantifying

the adoption of RPA in the gaming industry from var-

ious perspectives such as cost savings, return on in-

vestment, morale of employees whose work has been

replaced by other tasks, etc. However, this can only

be done after years of implementation, which is very

difﬁcult for an academic paper, at least when the pre-

sented topic is new. This problem of lack of theo-

retical basis and evaluation is also described in (Syed

et al., 2020).

Instead, we ﬁrst conduct an empirical (and pos-

sibly subjective) evaluation based on the collected

feedback by using our prototypes in automating pre-

viously small or released projects from our industry

partners. We then motivate the adoption of RPA by

looking at the success of automation with RPA in the

other sectors mentioned along Section 2, where the

adoption is mature enough and can provide real ben-

eﬁt metrics. Note that our prototypes have been im-

plemented as plugins in one of the most popular game

engines, namely Unreal Engine 5, so adoption and in-

tegration is smooth or at least familiar for game devel-

opers. The interfaces of the implemented prototypes

to the game products are able to call or retrieve call-

backs from the RPA software robots.

Motivation for Using RPA Instead of Previously

Used Automation Methods. After discussions with

the gaming industry partners, the conclusion is that at

the moment the automation methods used are based

on programmed pieces of code, i.e., scripts, written

by software engineers to facilitate the automation of

pipelines. We sought to explore the advantages and

disadvantages of custom scripts performing automa-

tion versus using RPA technology. While using our

prototypes on small sized released or open sourced

projects, we made a few observations by connecting

the feedback received with the potential that the RPA

is giving in automation related to our proposed goals.

The main disadvantage observed is the time re-

quired for stakeholders to become familiar with RPA

technology and the lack of conﬁdence in integrating

new technologies alongside existing projects. How-

ever, the beneﬁts observed appear to outweigh these

ICSOFT 2023 - 18th International Conference on Software Technologies

Users playing the game

Online data mining

Use / collect data

from human users

Store features

data

Online retrain

and infer

clusterization

methods

store

Run

Recommendation

systems

Suggest

promotion, in-

game tips,

rewards, etc.

Open live

chatbots with

players

Figure 6: Workﬂow of an RPA software robot that performs continuous data evaluation of features collected from players.

These features can be customized by developers, but generally relate to connections between the context of the game, game-

play, previous rewards, and objects of interest to the user. These features are processed, aggregated, and stored in a data store

for later use, which is then used to update existing clustering models commonly used for recommender systems. At the end

of the workﬂow, suggestions can be made automatically for other rewards, product promotions, or even using chatbots with

users to unlock their progress.

Set of devices used to

play the game on

Find/update set of user's devices

Prioritize devices by

predicting next time

of being used

Update game content or

releases in background

Continuously train

forecasting models for game

use and devices

Infer trained

models

Figure 7: An RPA robot that takes care of updating users’ devices in the background so that they can immediately play on the

predicted devices. Algorithms for predicting the devices on which the user will play the game (e.g., smartphones, consoles)

are continuously trained based on user data that takes into account, for example, time of day or seasonal data.

problems:

• The RPA workﬂows (or a portion of them, i.e., a

subset of the activity nodes) may be shared among

projects within the same organisation (Neifer

et al., 2022). There is even a marketplace (e.g.,

UiPath Marketplace

for components that can be

reused between organisations or communities. As

we work with game industry partners on our

project to gather requirements and ﬁnd missing

gaps for automation processes in today’s work-

ﬂows, we believe that providing a set of reusable

components would accelerate the adoption of au-

tomation in video games and reduce costs, as the

lack of expertise in many nice/unrelated technolo-

gies is a major problem in the industry.

• The graphical interface for creating RPA work-

ﬂows can be used by non-technical stakehold-

ers too, thus providing better scale in developing

and maintaining automation services (Chakraborti

et al., 2022).

• Workﬂows and activity nodes can be created to

https://marketplace.uipath.com

execute external, existing source code. In addi-

tion, the RPA robots can be triggered by the code

as needed. This makes the method non-invasive to

existing projects and can be replaced or added at

any time during the project. Our empirical testing

in Unreal Engine 5 has also shown that there is

no signiﬁcant performance penalty when calling

or registering callbacks for operations performed

by the robots, as the languages used can vary as

needed, e.g., C#/Visual Basic, Python, C++, etc.

• Process and Task Mining (Leno et al., 2021a) can

help in creating a ﬁrst draft of workﬂow deﬁni-

tions by simply observing patterns in existing re-

dundant human work through AI-based methods.

According to studies, this can signiﬁcantly reduce

the cost of implementing the technology in the

ﬁrst phase. Communication mining can also be

used to understand communication channels, cap-

ture messages and manually performed work, and

then automate these processes to improve opera-

tions and their efﬁciency.

Compared to the above beneﬁts, the problem with

using custom, project-speciﬁc automation scripts is

Robotic Process Automation for the Gaming Industry

often that it is difﬁcult to reuse these scripts within the

organisation or in a follow-on project. The ﬁrst prob-

lem is that the source code is written only by software

engineers and the automation services cannot scale in

terms of development and maintenance. In addition,

the source code is closely related to the project code

itself, making it difﬁcult to transfer it to a new project.

In addition, engineers from the different teams were

found to have different skills and a lack of knowledge

about related technologies. All of this made it difﬁcult

to reuse the previous automation scripts. In compari-

son, when using RPA workﬂows, there is a separation

of concerns, with decoupled and implementation, as

our preliminary prototypes show.

5 CONCLUSION

This paper presents a set of areas that can be au-

tomated in the game development pipeline through

the use of RPA software robots, along with proposed

concepts and implementation details. The presented

methods are non-invasive and can be integrated into

existing projects and infrastructures, as workﬂows

written in graphical editors can call both internal and

external libraries and third-party tools. We believe

that the adoption of RPA can help the gaming industry

in the future by providing automation processes that

can be shared between projects, can be easily deﬁned

by non-technical personnel, and have a high level of

maintenance. Our requirements, ideas and prototypes

have also been discussed and validated with our local

gaming industry partners. We are also exploring fur-

ther collaborations to incorporate the prototypes into

real projects and conduct an in-depth evaluation of

their beneﬁts. Last but not least, we plan to open-

source our prototypes, so that the interested develop-

ers may use or extend them.

ACKNOWLEDGEMENTS

This research was supported by the European Re-

gional Development Fund, Competitiveness Oper-

ational Program 2014-2020 through project IDBC

(code SMIS 2014+: 121512). We would also like to

thank our game development industry partners from

Amber, Ubisoft, and Electronic Arts for their feed-

back.

REFERENCES

Aguirre, S. and Rodriguez, A. (2017). Automation of

a business process using robotic process automation

(RPA): A case study. In Proc. of Applied Computer

Sciences in Engineering - 4th Workshop on Engineer-

ing Applications, volume 742 of CCIS, pages 65–71.

Springer.

Aleem, S., Capretz, L., and Ahmed, F. (2016). Game de-

velopment software engineering process life cycle: A

systematic review. Journal of Software Engineering

Research and Development, 4:1–30.

Aljunid, M. F. and Dh, M. (2020). An efﬁcient deep learn-

ing approach for collaborative ﬁltering recommender

system. Procedia Computer Science, 171:829–836.

Third International Conference on Computing and

Network Communications.

Andrade, D. L. (2022). Designing cloud-based gameplay

automation: Exploratory software testing, game state-

analysis, and test-driven development (tdd) applied to

robotic process automation (rpa). International Jour-

nal of Intelligent Computing Research (IJICR), 13(1).

Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J. D.,

Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G.,

Askell, A., Agarwal, S., Herbert-Voss, A., Krueger,

G., Henighan, T., Child, R., Ramesh, A., Ziegler, D.,

Wu, J., Winter, C., Hesse, C., Chen, M., Sigler, E.,

Litwin, M., Gray, S., Chess, B., Clark, J., Berner,

C., McCandlish, S., Radford, A., Sutskever, I., and

Amodei, D. (2020). Language models are few-shot

learners. In Larochelle, H., Ranzato, M., Hadsell, R.,

Balcan, M., and Lin, H., editors, Advances in Neu-

ral Information Processing Systems, volume 33, pages

1877–1901. Curran Associates, Inc.

Chakraborti, T., Rizk, Y., Isahagian, V., Aksar, B., and

Fuggitti, F. (2022). From natural language to work-

ﬂows: Towards emergent intelligence in robotic pro-

cess automation. In Business Process Management:

Blockchain, Robotic Process Automation, and Central

and Eastern Europe Forum, pages 123–137. Springer

International Publishing.

Dan, G., Claudiu, D., Alexandra, F., Adela, P.-P., Ovidiu,

S., Honoriu, V., and Liviu, M. (2022). Multi-channel

chatbot and robotic process automation. In 2022 IEEE

International Conference on Automation, Quality and

Testing, Robotics (AQTR), pages 1–6.

Fu, X., Chen, X., Shi, Y.-T., Bose, I., and Cai, S. (2017).

User segmentation for retention management in online

social games. Decision Support Systems, 101:51–68.

Jeeva Padmini, K. V., Perera, G. I. U. S., Dilum Bandara,

H. M. N., and Silva, R. K. O. H. (2021). A decision

support tool to select candidate business processes in

robotic process automation (rpa): An empirical study.

In Smys, S., Balas, V. E., Kamel, K. A., and Lafata, P.,

editors, Inventive Computation and Information Tech-

nologies, pages 567–582. Springer Singapore.

Jha, N., Prashar, D., and Nagpal, A. (2021). Com-

bining Artiﬁcial Intelligence with Robotic Process

Automation—An Intelligent Automation Approach,

pages 245–264. Springer International Publishing.

ICSOFT 2023 - 18th International Conference on Software Technologies

Karabinus, A. and Atherton, R. (2018). Games, ux, and the

gaps: Technical communication practices in an ama-

teur game design community. In Proceedings of the

36th ACM International Conference on the Design of

Communication, SIGDOC ’18. Association for Com-

puting Machinery.

Ketkar, Y. and Gawade, S. (2021). Effectiveness of robotic

process automation for data mining using uipath. In

2021 International Conference on Artiﬁcial Intelli-

gence and Smart Systems (ICAIS), pages 864–867.

Kim, J. W., Kim, D., Keegan, B., Kim, J. H., Kim, S.,

and Oh, A. (2015). Social media dynamics of global

co-presence during the 2014 ﬁfa world cup. In Pro-

ceedings of the 33rd Annual ACM Conference on Hu-

man Factors in Computing Systems, CHI ’15, page

2623–2632. Association for Computing Machinery.

Leno, V., Polyvyanyy, A., Dumas, M., La Rosa, M., and

Maggi, F. M. (2021a). Robotic process mining: vi-

sion and challenges. Business & Information Systems

Engineering, 63(3):301–314.

Leno, V., Polyvyanyy, A., Dumas, M., Rosa, M. L., and

Maggi, F. M. (2021b). Robotic process mining: Vi-

sion and challenges. Bus. Inf. Syst. Eng., 63(3):301–

314.

Ling, X., Gao, M., and Wang, D. (2020). Intelligent doc-

ument processing based on rpa and machine learning.

In 2020 Chinese Automation Congress (CAC), pages

1349–1353.

Mart

ınez, F., Charte, F., Fr

ıas, M. P., and Mart

ınez-

Rodr

ıguez, A. M. (2022). Strategies for time series

forecasting with generalized regression neural net-

works. Neurocomputing, 491:509–521.

Neifer, T., Bossauer, P., Lawo, D., Volkening, R., and

Gadatsch, A. (2022). The role of marketplaces for

the transformation from robotic process automation

to intelligent process automation. In Proceedings of

the 19th International Conference on Smart Business

Technologies (ICSBT), pages 15–25. SCITEPRESS-

Science and Technology Publications.

Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright,

C. L., Mishkin, P., Zhang, C., Agarwal, S., Slama, K.,

Ray, A., Schulman, J., Hilton, J., Kelton, F., Miller,

L. E., Simens, M., Askell, A., Welinder, P., Christiano,

P. F., Leike, J., and Lowe, R. J. (2022). Training lan-

guage models to follow instructions with human feed-

back. ArXiv, abs/2203.02155.

Paduraru, C., Cristea, R., and Stefanescu, A. (2023). En-

hancing the security of gaming transactions using

blockchain technology. In 37th IEEE/ACM Interna-

tional Conference on Automated Software Engineer-

ing, ASE22. Association for Computing Machinery.

Paduraru., C., Paduraru., M., and Iordache., S. (2022). Us-

ing deep reinforcement learning to build intelligent

tutoring systems. In Proceedings of the 17th Inter-

national Conference on Software Technologies - IC-

SOFT,, pages 288–298. INSTICC, SciTePress.

Paduraru, C., Paduraru, M., and Stefanescu, A. (2022).

Rivergame - a game testing tool using artiﬁcial intelli-

gence. In 2022 IEEE Conference on Software Testing,

Veriﬁcation and Validation (ICST), pages 422–432.

Politowski, C., Petrillo, F., and Gu

eneuc, Y.-G. (2021).

A survey of video game testing. In 2021 IEEE/ACM

International Conference on Automation of Software

Test (AST), pages 90–99.

Qasrawi, R., Amro, M., and Jayousi, R. (2020). Automatic

analytics model for learning skills analysis using game

player data and robotic process automation in a seri-

ous game for education. In 2020 International Confer-

ence on Promising Electronic Technologies (ICPET),

pages 94–98.

Ribeiro, J., Lima, R., Eckhardt, T., and Paiva, S. (2021).

Robotic process automation and artiﬁcial intelligence

in industry 4.0–a literature review. Procedia Com-

puter Science, 181:51–58.

Rosmalen, P., Eikelboom, P., Bloemers, E., Winzum, K.,

and P.Spronck (2012). Towards a game-chatbot: Ex-

tending the interaction in serious games. In Proceed-

ings of the European Conference on Games-based

Learning.

Syed, R., Suriadi, S., Adams, M., Bandara, W., Leemans,

S. J., Ouyang, C., ter Hofstede, A. H., van de Weerd,

I., Wynn, M. T., and Reijers, H. A. (2020). Robotic

process automation: Contemporary themes and chal-

lenges. Computers in Industry, 115:103162.

Tyagi, A. K., Fernandez, T. F., Mishra, S., and Kumari,

S. (2021). Intelligent automation systems at the core

of industry 4.0. In International conference on intel-

ligent systems design and applications, pages 1–18.

Springer.

Vijai, C., Suriyalakshmi, S., and Elayaraja, M. (2020). The

future of robotic process automation (rpa) in the bank-

ing sector for better customer experience. Journal of

Commerce, 8(2):61–65.

Yokoyama, H., Matsumoto, T., Fukui, M., Mori, Y., and

Takami, T. (2020). Development of programming ma-

terials using gameplay rpa. In 2020 IEEE 9th Global

Conference on Consumer Electronics (GCCE), pages

248–250.

Robotic Process Automation for the Gaming Industry