PREMIER JOURNAL OF SCIENCE REVIEW

The Task of Mathematical Modelling Using a Programming Language:

A Scoping Review
Aliima Mamatkasymova1, Gulaim Zikirova2, Gulmira Saparova3 , Bektur Omurzakov2 and
Sonun Asanova4

ABSTRACT • The analysis of the features of software implemen-

tation contribute to the selection of optimal meth-
BACKGROUND
This study aims to explore the role of programming ods for solving a wide class of problems.
languages in the development and implementation of • The development of mathematical modelling is re-
mathematical models, with a focus on the integration lated to the integration of quantum computing.
of advanced computing technologies. • The development of cloud technologies is making a
significant contribution to the evolution of mathe-
MATERIALS AND METHODS
matical modelling.
Utilising a narrative review method, the study
methodically examines the body of research on
mathematical modelling and the use of programming Introduction
languages like Python, C++, and Julia. The performance Mathematical modelling is an important tool in various
This is an open access article of these languages is compared in a number of scientific and applied disciplines, allowing research-
distributed under the terms
mathematical modelling tasks, such as numerical ers and practitioners to analyse complex systems and
of the Creative Commons
methods, linear algebra, and physical modelling. processes. In the period 2015–2024, there has been a
Attribution License, which
permits unrestricted use, significant increase in interest in using programming
RESULTS
distribution, and reproduction languages to implement mathematical models. This is
in any medium, provided the The paper emphasises how cloud computing, artificial
conditioned by the need to improve the efficiency and
original author and source are intelligence, and hybrid algorithms have significantly
credited. improved the precision and effectiveness of mathe- accuracy of solutions in conditions of rapidly changing
matical models. While C++ offers great performance data and complex systems. Modern tasks facing the
1
Department of Exact Sciences,
in computationally demanding jobs but necessitates scientific community require the integration of the the-
Osh Technological University
named after M.M. Adyshev, Osh, more development effort, Python has been demon-
oretical foundations of mathematical modelling with
Kyrgyz Republic strated to be beneficial for speedy development practical programming methods, which makes this
2
Department of Humanities, research relevant and in demand.
because of its vast library ecosystem. Julia is a prom-
Pedagogical and IT Technologies,
Osh Technological University ising language for mathematical modelling because In the period 2015–2024, considerable attention was
named after M.M. Adyshev, Osh, it strikes a compromise between usability and perfor- paid to an interdisciplinary approach in mathematical
Kyrgyz Republic mance. The investigation also shows that the choice modelling, which allows expanding the boundaries of
3
Department of Applied
Mathematics, Osh Technological of computing methods and programming languages the application of existing methods and creating new,
University named after M.M. has a significant impact on the effectiveness of math- more effective solutions. A special role in this process
Adyshev, Osh, Kyrgyz Republic ematical models. Every language offers advantages was played by the integration of artificial intelligence
4
Department of Applied based on the particular modelling task, as shown by and machine learning into conventional mathematical
Informatics, Osh Technological a thorough analysis of execution time, memory utili- modelling methods. This opens up new opportunities
University named after M.M. sation, and code size. Furthermore, the combination for solving complex optimisation and forecasting prob-
Adyshev, Osh, Kyrgyz Republic
of quantum computing and machine learning offers lems in various fields of science and technology.
fresh possibilities for resolving increasingly challeng- The development of cloud technologies and dis-
Correspondence to:
Gulmira Saparova ing issues that conventional approaches are unable to tributed computing is also making a significant con-
saparovagulmira983@[Link] effectively handle. tribution to the evolution of mathematical modelling,
Additional material is published CONCLUSION helping to solve increasingly complex problems using
online only. To view please visit
the journal online. According to the study’s findings, mathematical mod- parallel computing and distributed data processing.
elling will depend more and more on the cooperation This is especially important in the context of big data
Cite this as: Mamatkasymova A,
Zikirova G, Saparova G, Omurzakov of traditional approaches, contemporary programming processing and solving tasks that require significant
B and Asanova S. The Task of languages, and cutting-edge technologies like artificial computing resources. Modern trends in the develop-
Mathematical Modelling Using a intelligence and quantum computing.
Programming Language: A Scoping ment of computer technology create prerequisites for
Review. Premier Journal of Science Keywords: Containerisation, Distributed systems, the emergence of new methods and approaches to
2025;14:100173 Hybrid algorithms, Quantum technologies, Resource mathematical modelling that can be effectively imple-
DOI: [Link] optimisation, Symbolic computing mented using modern programming languages.
PJS.100173
The use of high-level programming languages com-
Peer Review Highlights bined with optimised libraries for scientific computing
Received: 4 September 2025 • One of the most significant trends is the symbio- can significantly reduce the time needed to develop and
Last revised: 29 October 2025 sis of classical methods and artificial intelligence, implement mathematical models, making them more
Accepted: 29 October 2025
where neural network algorithms complement accessible to a wide range of researchers and practi-
physical models. tioners. This contributes to the faster ­development of
Version accepted: 5
Published: 20 November 2025
1
 PREMIER JOURNAL OF SCIENCE REVIEW

Ethical approval: N/a this field and the emergence of new promising areas their effective application. The main questions guiding
Consent: N/a of research. this scoping review are the following:
Funding: No industry funding According to Li et al.1, there is a positive correlation
Conflicts of interest: N/a between mathematical skills and programming effec- • What are the key theoretical approaches in mathe-
Author contribution: tiveness, which highlights the importance of mathe- matical modelling for different problem types;
Aliima Mamatkasymova matical training for successful solving of modelling • How do software implementation features influ-
and Gulaim Zikirova: problems. Research in the field of linear programming, ence the selection of modelling methods;
conceptualisation, methodology,
data curation, writing- such as the study by Bellingeri et al.2 and Gholamnejad • What role do advanced computing technologies
original draft preparation. et al.3, demonstrated how mathematical models can play in enhancing the efficiency and accuracy of
Gulmira Saparova and Bektur be used to optimise processes in various industries, mathematical modelling.
Omurzakov: software, validation,
writing-reviewing, and editing. including agriculture and mining. These studies con-
Sonun Asanova: visualisation, firm that mathematical modelling covers a wide range Materials and Methods
investigation, and supervision. of applications, from optimising resources in agricul- This study’s foundation is a thorough analysis of pre-
All authors read and approved
the final manuscript ture to solving complex health problems.4,5 vious studies on the application of hybrid algorithms,
Guarantor: Aliima
However, despite significant achievements in the cloud computing, and quantum computing in the inte-
Mamatkasymova field of mathematical modelling, the theory of this field gration of mathematical modelling with contemporary
Provenance and peer-review: as a separate science is just beginning to take shape, programming languages.
Unsolicited and externally peer- which opens up new horizons for further research.6 The study involved a thorough search of several
reviewed It is important to note that using programming lan- scholarly databases, including Google Scholar, IEEE
Data availability statement: guages to implement mathematical models not only Xplore, Scopus, and SpringerLink. Articles, conference
All the data is provided in the
manuscript improves the quality of solutions but also contributes proceedings, and technical reports were all included in
to a deeper understanding of mathematical concepts the search, giving a wide overview of the area. Studies
and methods. Within the framework of this study, a included in the review were required to:
wide range of theoretical approaches and methods will
be considered that help to effectively solve mathemat- • address mathematical modelling techniques using
ical modelling problems using modern programming programming languages in fields such as physics,
languages. engineering, resource optimisation, and artificial
The systematisation and analysis of the theoretical intelligence;
foundations of mathematical modelling, and the study • present applications of hybrid methods that com-
of the features of the software implementation of mod- bine traditional mathematical techniques with ma-
els in various programming languages, are aimed at chine learning algorithms or cloud-based comput-
identifying the most promising approaches and meth- ing platforms;
ods for solving a wide class of modelling problems. • be written in English across 2000–2025;
This was confirmed by Taloub et al.7, who discussed • focus on analytical, numerical, and statistical ap-
numerical methods for solving equations, which is an proaches for solving complex scientific and techni-
important aspect in the context of programming. Addi- cal problems.
tionally, the study by Fitriah et al.8 emphasised the • The search queries for each database were:
importance of applying models in an educational con- • Google Scholar: The query used was: (“mathemat-
text, where the use of appropriate resources and mate- ical modelling” AND “programming languages”
rials can significantly improve mathematical learning AND (“hybrid algorithms” OR “machine learning”
outcomes. This indicates that theoretical approaches OR “cloud computing” OR “quantum computing”)),
to mathematical modelling can be successfully inte- aiming to capture studies that explore the intersec-
grated into educational programmes, which, in turn, tion of mathematical modelling and advanced com-
contributes to the development of programming skills puting technologies.
among students. • IEEE Xplore: The following search phrases were
Thus, the connection between mathematical mod- employed: (“mathematical modelling” AND “pro-
elling and educational methods opens up new hori- gramming languages” AND (“machine learning” OR
zons for further research and practical application. “AI” OR “cloud computing” OR “quantum comput-
Afrilianto et al.9 also emphasised the importance of a ing”)), ensuring the inclusion of papers discussing
creative approach in teaching mathematics, which may the application of modern computational methods
be associated with the development of new methods in mathematical modelling.
and models for solving mathematical modelling prob- • Scopus: The search string used was: (“mathemat-
lems. The study showed that the use of active teaching ical modelling” AND “programming languages”)
methods, such as project-based learning, promotes AND (“hybrid algorithms” OR “cloud computing”
the development of creative thinking among students, OR “machine learning” OR “quantum computing”),
which is an important aspect for the successful appli- designed to identify studies that integrate mathe-
cation of mathematical models in real-world tasks. matical modelling with advanced programming
The purpose of this study is to analyse existing languages and cutting-edge technologies.
approaches and methods for the use of programming • SpringerLink: The search utilised the following
languages in solving mathematical modelling prob- query: (“mathematical modelling” AND “program-
lems, with the goal of providing recommendations for ming languages”) AND (“hybrid methods” OR “AI

2 DOI: [Link] | Premier Journal of Science 2025;14:100173

 PREMIER JOURNAL OF SCIENCE REVIEW

integration” OR “cloud-based solutions”), focusing approach (analytical, numerical, or statistical) were

on hybrid approaches and the use of AI in mathe- one of the collected data. Additionally, the program-
matical modelling. ming languages used in the studies, such as Python,
C++, Julia, and others, were documented. Examples of
These search terms were specifically designed to col- the study’s application fields included resource opti-
lect research that looked at programming languages’ misation, particle physics, and AI integration. Addi-
function in mathematical modelling, especially when tionally documented were cloud technology, quantum
it came to combining cloud, machine learning, and computing, and hybrid approaches. In order to evalu-
hybrid techniques. The literature search was con- ate the efficacy of the methods, important results were
ducted on April 1, 2025, ensuring the inclusion of the finally extracted, including accuracy, performance,
latest studies and technological advancements. and computing cost.
A total of 254 articles were initially identified The assessment of methodological quality was
through database searches (Figure 1). After removing carried out using appraisal tools appropriate for a
65 duplicates, 189 articles remained for further screen- variety of non-trial literature because of the hetero-
ing. During the title and abstract screening phase, 38 geneity of the included works, which include theo-
studies were excluded due to irrelevance, either not retical articles, computational reports, educational
addressing mathematical modelling or programming assessments, and experimental modelling studies.
languages, or being focused on unrelated fields. In the Each publication type was assessed with an appro-
full-text screening phase, 86 studies were excluded priate tool. The Scale for the Assessment of Narra-
for failing to meet the inclusion criteria. Studies were tive Review Articles (SANRA) was used for narrative
excluded for being irrelevant to the scope of the review, and theoretical studies, evaluating criteria like clar-
lacking integration of advanced technologies, being ity, search strategy, and transparency. AMSTAR-2
written in a language other than English, or having assessed systematic reviews based on 16 domains,
methodological issues such as a lack of experimental such as literature search comprehensiveness and
data, case studies, or practical applications related to statistical analysis rigor, categorising studies by
programming languages in mathematical modelling. quality. The Critical Appraisal Skills Programme
Ultimately, 65 studies were included in the final review, (CASP) checklists appraised empirical and mixed-
meeting the criteria of being relevant to mathematical methods studies, focusing on study design, data col-
modelling using programming languages and incorpo- lection, and analysis methods. These tools were used
rating advanced technologies like AI, cloud comput- independently by two authors, with disagreements
ing, and quantum computing. A table summarising the being settled by consensus and debate. Despite
key characteristics of the included studies, their mod- differences in study design and reporting requirements,
elling techniques, programming languages used, and this procedure guaranteed a transparent, thorough
performance metrics is provided in Appendix A. The evaluation of methodological quality across diverse
list of included and excluded studies, along with the sources, enabling a balanced synthesis of data. To syn-
reasons for their inclusion or exclusion based on the thesise the findings across studies of varying quality,
established criteria is presented in Appendix B. each study’s SANRA, AMSTAR-2, or CASP score was
Data were extracted independently by two review- used to weigh the results during the synthesis phase.
ers using a standardised form. The types of modelling Although lower-quality studies were included for con-
text, their results were limited by their methodologi-
cal flaws, whereas higher-quality research were given
more weight in the final synthesis.
Total articles retrieved: 254 Duplicates removed: 65 Based on the kinds of modelling techniques and
computer technologies employed, the studies were
categorised. Execution time, memory utilisation, scal-
ability, and code size were the criteria used to assess
Articles screened based on titles Articles removed after the efficacy of various programming languages. When
and abstracts: 189 screening: 38 feasible, both qualitative and quantitative versions of
the results were provided.
The review adhered to the PRISMA guidelines to
ensure transparency, reproducibility, and the rigorous
Articles excluded for full-text Articles removed after full- selection of relevant studies (Table 1).
screening: 151 text screening: 86 The methodology of the study also included a com-
parative analysis of modelling methods (analytical,
numerical, statistical), highlighting their strengths and
weaknesses in the context of various classes of tasks,
Final articles included in the such as dynamic systems, optimisation, and probabi-
review: 65 listic processes. Three main criteria (accuracy, resource
intensity, and adaptability) formed the basis of this
Fig 1 | Flowchart of literature search and selection process study’s comparative analysis of modelling approaches.
Source: Compiled by the authors The ability of each approach to deliver dependable

DOI: [Link] | Premier Journal of Science 2025;14:100173 3

PREMIER JOURNAL OF SCIENCE REVIEW

Table 1 | PRISMA-ScR 2018 checklist

Section Item Response Page
TITLE 1. Identify the report as a systematic review “The Task of Mathematical Modelling Using a Programming Language: A Scoping 1
Review”
ABSTRACT 2. Provide structured abstract Structured abstract with: Background, Methods, Results, Conclusions 1
INTRODUCTION 3. Describe rationale “Integration of programming languages and advanced computing technologies 2
improves the accuracy and efficiency of mathematical models”
4. State objectives “To analyse existing approaches and methods for the use of programming 3
languages in solving mathematical modelling problems, with the goal of
providing recommendations for their effective application”
METHODS 5. Indicate if review protocol exists No protocol registered (transparently stated) 5
6. Specify characteristics of the sources of Peer-reviewed studies in English 2000–2025, relevant to programming 3
evidence used as eligibility criteria languages in mathematical modelling
7. Describe information sources IEEE Xplore, Scopus, SpringerLink, Google Scholar 4
8. Present full search strategy Full search strategies provided for each database, including Boolean search 4
terms and date restrictions
9. Explain study selection Dual independent screening for eligibility and inclusion 3
10. Describe data extraction Standardised forms for study design, programming languages, outcomes 4
11. List the data items Programming languages (Python, C++, Julia), modelling type, computational 5
metrics (execution time, memory use).
12. Critical appraisal of individual sources of Quality was assessed using SANRA for narrative works, AMSTAR-2 for systematic 5
evidence§ reviews, and CASP for empirical studies; methodological limitations were noted.
13. Describe synthesis methods Synthesis with comparative analysis due to heterogeneity in task complexity 5
14. Synthesis of results Synthesis is used, comparing programming language performance across 7–18
different modelling tasks
RESULTS AND 15. Critical appraisal Bias risk summarised at study and review level; limitations of heterogeneous data 7–18
DISCUSSION acknowledged.
16. Results of synthesis Comparative analysis of programming language performance (Python, C++, Julia). 17–18
17. Summary of evidence Highlights integration of AI and quantum computing in mathematical modelling 18
efficiency.
18. Discuss limitations Lack of standardised benchmarks and limited external validation may limit 18
generalisability
19. Conclusions Programming languages and hybrid technologies are essential for modern 1
mathematical modelling.
OTHER 20. Protocol availability N/a 1
21. Report funding N/a 1
22. Declare conflicts Authors declared no conflicts of interest 1
23. Data availability All the data was provided in the manuscript 1
24. Flow diagram Data extraction forms available upon request Figure 1
PRISMA-SPECIFIC 25. Checklist citation Tricco et al.10 19
Source: Compiled by the authors.

results for various task classes, such as dynamic sys- computational challenges and have a wide range of
tems, optimisation, and probabilistic processes, was applications in scientific and engineering domains.
used to gauge accuracy. While adaptability concen- n-body simulations involve calculating gravitational
trated on how well each method could be applied forces between multiple bodies, making them highly
across a range of problem types, from simple to com- demanding in terms of computational efficiency. Man-
plex, resource intensity took into account the compu- delbrot set calculations test both recursive functions
tational cost and efficiency required by each method. and numeric processing, whereas spectral-norm com-
For this purpose, the principles of deductive gener- putations evaluate a language’s capacity to handle
alisation were applied: from the analysis of special matrix operations in linear algebra. These challenges
cases, such as modelling turbulent flows or thermal were selected to represent a variety of computational
processes, to the formulation of universal patterns. issue types that are frequently encountered in scien-
The n-body simulation, spectral-norm computation, tific computing, such as numerical methods, linear
and Mandelbrot set calculation tasks were chosen algebra, and physical modelling. Workload sizes,
for this review because they each represent distinct like the number of bodies in the n-body simulation

4 DOI: [Link] | Premier Journal of Science 2025;14:100173

PREMIER JOURNAL OF SCIENCE REVIEW

or the iterations in the Mandelbrot calculation, were complex problems where analytical solutions are
­standardised between investigations to provide a fair unavailable. Numerical methods are of particular
comparison. This method enables a thorough assess- importance in modelling turbulent flows. Statistical
ment of each programming language’s performance on methods are effective when dealing with uncertainty
a range of computing tasks. Three programming lan- and probabilistic processes. The evolutionary devel-
guages were used in the examined studies: Julia, C++, opment of modelling methods is characterised by
and Python. Python and Julia were selected for their the transition from analytical to numerical methods.
high-level, user-friendly environments, while C++ was The choice of the modelling method is determined
frequently chosen for its computational efficiency in by the specifics of the area under study. In particle
resource-demanding jobs. physics, priority is given to statistical methods of mod-
An overview of the studies, which examines the elling the interaction of protons with matter, especially
integration of mathematical modelling techniques when using supercomputers.
with modern programming languages, was provided Modern computing power allows the application of
(Appendix C). The studies utilised programming lan- complex methods, including dynamic and large-scale
guages like Python, C++, and Julia, each chosen for modelling methods. Engineering tasks are often solved
their suitability in handling complex computational using simpler models such as RANS, which ensure a
tasks. The studies assessed programming language balance between accuracy and resource intensity.
performance using key metrics: execution time (mea- Quantum computing has the potential to revolution-
sured in CPU seconds), memory usage (measured in ise dynamic and large-scale modelling by leveraging
megabytes), and code size (measured in bytes). These quantum bits (qubits) that can exist in multiple states
metrics reflect the efficiency, complexity, and scalabil- simultaneously. This makes it possible for quantum
ity of the programming languages. Comparative anal- computers to solve complicated optimisation prob-
yses of these metrics were conducted to evaluate the lems far more quickly than traditional computing
programming languages’ effectiveness for the compu- techniques, which are constrained by sequential pro-
tational tasks at hand. Data from these studies were cessing and binary bits. Numerous potential solutions
collected and analysed using tools such as Pandas for can be investigated simultaneously by quantum algo-
data manipulation and Matplotlib for visualisations. rithms, like the Quantum Approximate Optimisation
The comparative analysis was based on standardised Algorithm. The accuracy and efficiency of dynamic
performance measurements from multiple resources, simulations and optimisation tasks could be signifi-
allowing for direct comparison of programming lan- cantly increased by quantum computers’ capacity to
guages across the reviewed studies. handle exponentially large datasets and execute com-
This review’s limitations include the exclusion of putations at previously unheard-of speeds. This is par-
research written in languages other than English, the ticularly true in domains like drug discovery, logistics,
possibility of publication bias, and the omission of and climate modelling, where traditional approaches
grey literature, which might have offered more infor- are unable to handle the problems’ extreme complexity
mation about the changing approaches in mathemat- and scale.
ical modelling. Hybrid approaches combining different modelling
Promising areas, such as the integration of quantum methods demonstrate increased efficiency. The inte-
computing and machine learning into conventional gration of numerical and analytical methods helps
modelling methods, were studied through the prism to improve the quality of modelling while optimising
of conceptual forecasting. This included analysing computational costs. The effectiveness of modelling
trends identified in publications on neural network methods is assessed by criteria of accuracy, computa-
approaches for approximating complex functions and tional speed, and big data processing capability. Each
optimising model parameters. approach is characterised by specific limitations and
advantages. Analytical methods provide high accuracy
Results and Discussion with limited applicability to complex tasks. Numerical
methods are flexible with increased resource intensity.
Theoretical Foundations of the Integration of Statistical methods are effective in conditions of uncer-
Mathematical Modelling and Programming tainty, but their accuracy depends on the quality of the
Mathematical modelling as a field of research covers a source data.
wide range of methods and techniques used to solve The relationship between mathematical concepts
scientific and technical problems. Progress in this area and software implementations is a complex area of
is conditioned by the development of computing tech- research covering various aspects of the transformation
nologies and the increase in available resources. The of abstract mathematical ideas into practical software
structure of mathematical modelling contains three solutions. This relationship is of critical importance in
main classes of methods. Analytical methods are used the fields of computer science, engineering, and data
to obtain accurate solutions in the field of differential analysis, where mathematical models underlie algo-
equations. However, in conditions of complex systems, rithms and systems that control software applications.
analytical approaches demonstrate limited effective- A key aspect of translating mathematical concepts into
ness. Numerical methods, including the finite element programme code is the need for a deep understanding
and finite difference methods, provide ­solutions to of mathematical structures and their representation

DOI: [Link] | Premier Journal of Science 2025;14:100173 5

PREMIER JOURNAL OF SCIENCE REVIEW

in programming languages. Mathematical think- Optimisation of Mathematical Models using

ing is widespread in software engineering, which is Modern Technologies
­confirmed by the frequent appearance of mathemati- The architecture of cloud solutions for mathemati-
cal formulas in the code bases of real projects, espe- cal modelling is based on a multi-level structure that
cially in programming languages such as Java.11 integrates various components for efficient computing
The complexity of mathematical models has a sig- and data management. The architecture is based on
nificant impact on their software implementation. IaaS services, which provide fundamental comput-
More complex mathematical models require the use of ing resources for executing mathematical models.16
sophisticated algorithms and data structures to ensure The integration of container technologies optimises
efficient calculations and accurate results. Under- the orchestration of tasks in cloud environments.17
standing mathematical symbols and their contextual Scalability of computing resources in a cloud environ-
meaning plays a critical role in the implementation ment is a key advantage of using cloud technologies.
of algorithms based on specific mathematical opera- Dynamic resource allocation allows efficient manage-
tions.12 The use of realistic mathematical approaches ment of various workloads without significant initial
contributes to the visualisation and implementation investment in equipment. The elasticity of cloud ser-
of complex mathematical concepts by linking them vices ensures cost optimisation for fluctuating com-
to practical tasks.13 The efficiency of mathematical puting needs. The economic efficiency of using cloud
algorithms can be improved by selecting appropriate resources is achieved through a pay-as-you-go model
data structures consistent with the mathematical oper- that optimises computing costs and effectively man-
ations performed. Testing and validation of software ages the budget.
implementations of mathematical models is a neces- When choosing mathematical modelling methods,
sary process to ensure the reliability and accuracy of the analysis of computing resources and their com-
the software. pliance with the requirements of the task is of critical
The ecosystem of the programming language sig- importance. The effectiveness of different approaches
varies significantly depending on the available com-
nificantly increases its applicability for mathematical
puting power, which requires careful selection of
modelling. The NumPy and SciPy libraries for Python
modelling methods for each specific case. For sys-
provide powerful tools for numerical calculations,
tems with limited computing resources, it is optimal
while R is equipped with extensive statistical pack-
to use explicit numerical schemes instead of implicit
ages. The availability of such libraries often becomes
ones, which significantly reduces the computational
a crucial factor when choosing a programming lan-
load. Dimensionality reduction methods, such as the
guage for a specific modelling task. Evaluation of the
principal component method or autoencoders with a
­performance of various programming languages in
small number of layers, are of particular importance.
solving mathematical problems demonstrates that
The implementation of sequential algorithms should
dynamic languages, despite their flexibility, may be
be accompanied by memory optimisation through the
inferior in performance to statically typed languages.
use of sparse matrices and data streaming. A signifi-
C++ is traditionally preferred in high-performance
cant reduction in computing load can be achieved by
computing environments. Just-in-time (JIT) and ahead-
pre-aggregating and filtering data, which reduces the
of-time (AOT) compilation strategies significantly affect amount of information processed. In such conditions,
computational throughput, as seen in Julia’s JIT opti- it is advisable to use lower-order approximation meth-
misation and C++’s AOT efficiency. Through interop- ods with adaptive error control.
erability with libraries written in other languages, High-performance Graphics Processing Unit (GPU)
integration with high-performance numerical proce- accelerated systems open up opportunities for imple-
dures, or GPU-accelerated modules, Foreign Function menting an integrated approach to modelling. In such
Interfaces (FFI) increase modelling capabilities. Scal- conditions, it becomes effective to use high-order pre-
ability, efficiency, and accuracy in large-scale simu- cision methods with dynamic step adaptation. Signif-
lations and data-intensive calculations are directly icant advantages are provided by the introduction of
influenced by support for parallel and GPU computa- multiscale modelling with simultaneous calculation
tions as well as different memory models, ranging from of processes of various scales. Improved reliability of
human control in C++ to automatic garbage collection the results is achieved through the use of ensemble
in Python. The implementation of parallel computing methods with parallel calculation of multiple imple-
capabilities in various programming languages plays mentations. High performance allows implementing
a key role in the development of mathematical mod- completely implicit schemes with direct methods for
els. The T-system, as the researchers note, is a soft- solving systems of equations. For complex subsystems,
ware environment that supports automatic dynamic the use of machine learning methods in building sur-
parallelisation.14 Syntactic features of programming rogate models becomes effective.
languages affect the readability and maintainability The decomposition of mathematical problems into
of the code. The relationship between object-oriented subtasks provides the possibility of parallel execution.
programming and modelling indicates that the design Functional decomposition strategies are applied based
philosophy of a language can influence its effective- on the division of tasks by performed functions and
ness in mathematical modelling.15 data decomposition, where data sets are distributed

6 DOI: [Link] | Premier Journal of Science 2025;14:100173

PREMIER JOURNAL OF SCIENCE REVIEW

among nodes. Scalable and reactive systems efficiently process large-scale datasets and provide practical rec-
process large amounts of data through massively dis- ommendations optimises operational efficiency and
tributed architectures.18 Load balancing algorithms in strategic planning. Optimisation of memory usage and
distributed systems prevent overloading of individual caching is implemented through data locality manage-
nodes. Round-robin, least connections, and dynamic ment to provide quick access to frequently used infor-
load balancing algorithms are used. The choice of algo- mation. Caching mechanisms for intermediate results
rithm is determined by the characteristics of computa- minimise redundant calculations. Efficient memory
tional tasks and the network topology. management reduces access time and optimises overall
Synchronisation protocols for computing processes system performance. Computational complexity reduc-
ensure consistent operation and correct data exchange. tion techniques are based on the use of approximation
Locking mechanisms, semaphores, and message pass- algorithms that provide solutions close to optimal with
ing are used. The choice of synchronisation method reduced computational costs. The modified algorithms
affects the performance of distributed systems. Data demonstrate significant performance improvements in
consistency mechanisms include strong and eventual signal processing applications.23 Methods of simpli-
consistency models. Distributed transactions, consen- fying calculations, considering the specifics of tasks,
sus algorithms, and versioning are used. Consensus allow achieving faster and more effective solutions to
mechanisms ensure data integrity in distributed sys- complex mathematical problems.
tems.19 Distributed computing performance assess- Adaptive algorithm selection methods are based on
ment methods analyse performance metrics, including the analysis of input data to dynamically determine
throughput, latency, scalability, and resource utilisa- the optimal algorithm for completing a task. In sce-
tion. Benchmarking and performance analysis tools narios with significant variability in the computational
allow evaluating the system’s compliance with design costs of different algorithms, the adaptive approach
goals. Ways to minimise communication overhead minimises the total execution time. Hybrid schemes
include data compression, local processing, and effi- for optimising computational requirements in visible
cient communication protocols. light communication systems demonstrate the advan-
The integration of artificial intelligence (AI) into tages of adaptive algorithm selection.24 Optimisation of
modelling processes transforms the capabilities of Input/Output (I/O) and data exchange includes batch
mathematical models in various fields. Machine learn- processing of operations and minimisation of data
ing methods optimise mathematical models by identi- transfer. Intelligent adaptive management process
fying patterns and making predictions based on data. optimisation systems emphasise the importance of
The use of regression algorithms, the method of support effective data management to improve overall system
vectors and ensemble methods helps to refine models performance. Focusing on I/O optimisation ensures
using historical data.20 Neural network approaches improved performance and responsiveness of comput-
are effective in approximating complex functions that ing processes.
conventional mathematical models do not accurately
describe.21 The flexibility of neural networks ensures Comparative Analysis of the Effectiveness of Various
their adaptation to different types of data. Hybrid mod- Programming Languages
elling systems combine conventional mathematical The performance of programming languages for mathe-
models with AI elements, combining the interpretabil- matical modelling is analysed based on three represen-
ity of mathematical models with the adaptability of tative tasks: physical modelling (n-body), linear algebra
AI. Such systems use a mathematical basis to generate (spectral-norm), and numerical methods (Mandel-
initial forecasts and then refine them using machine brot). The comparative analysis, presented in Table 2,
learning methods based on up-to-date data. highlights significant differences in performance
Automatic adjustment of model parameters is across these tasks, with each programming language
implemented through genetic algorithms and Bayes- demonstrating varying degrees of efficiency depending
ian optimisation, replacing manual adjustment and on the task at hand.
trial and error methods. Systematic evaluation of var- Each performance parameter in Table 2 is expressed
ious parameter configurations helps to determine the as a percentage of the task’s maximum value, ensur-
optimal settings that increase the performance of the ing a fair comparison of programming languages. This
model. Predictive analytics based on AI analyses large makes it possible to evaluate each language’s perfor-
amounts of data to identify patterns and make predic- mance across jobs in a relative manner by eliminating
tions. AI algorithms form predictive models that clas- scale disparities across measurements such as execu-
sify new data and predict future trends.22 AI-based tion time, code size, and compilation time. A more fair
model validation methods include cross-validation and consistent comparison of performance is made
and analysis of discrepancies between predicted and possible by normalisation.
actual results. The integration of AI into validation According to the analysis, Python takes the longest to
processes increases the reliability of models and their complete the n-body physical modelling work (100%),
applicability in real conditions. while Julia and C++ do noticeably better, with execu-
Intelligent decision support systems integrate math- tion durations that are less than 2% of Python’s. The
ematical models with AI algorithms to generate rec- compilation times differ even if the code sizes of the
ommendations based on data analysis. The ability to three languages are similar. Julia is a potential option

DOI: [Link] | Premier Journal of Science 2025;14:100173 7

PREMIER JOURNAL OF SCIENCE REVIEW

Table 2 | Comparative performance characteristics of preserving Python’s high-level syntax and flexibility.
programming languages, % This assertion is supported by studies by Bellingeri
N-body
et al.2 and Taloub et al.7, which show that Julia ­performs
almost as well as C++ in workloads requiring expen-
Language Execution Time Code Size Compilation Time
sive matrix operations while offering user-friendliness
Python 100 85.66 100.00 that speeds up development times.
C++ 1.44 100 54.49 Compilation time is an important consideration
Julia 0.63 83.31 2.49 for assessing language efficiency, especially in itera-
Spectral-Norm tive scientific computing where quick prototyping is
required, in addition to performance indicators like
Language Execution Time Code Size Compilation Time
execution time, memory utilisation, and code size. C++
Python 100.00 40.48 100.00
and Julia require lengthier compilation stages than
C++ 0.68 100.00 58.11 Python, which usually requires little. Julia’s JIT com-
Julia 0.33 41.43 2.64 piler causes delays during the initial run but makes up
Mandelbrot for it with faster subsequent executions. Studies must
address the uncertainty present in various hardware
Language Execution Time Code Size Compilation Time
configurations in order to further evaluate the trust-
Python 100.00 100.00 100.00
worthiness of these measurements. The outcomes are
C++ 1.42 94.81 55.90 affected by variations in cache optimisation strategies,
Julia 0.18 74.56 2.40 memory access speed, and CPU architecture. Accord-
Note: All indicators are normalised relative to the maximum value for each ing to Yang et al.19, the performance ranking of Python,
metric and are presented as a percentage. C++, and Julia varied depending on the underlying
Source: Compiled by the authors based on25.
hardware architecture when the identical code was run
on several platforms. This emphasises how crucial it
for mathematical modelling since it continuously is to define experimental parameters, such as task set-
demonstrates optimal performance with a moderate tings and system specifications, in order to guarantee
code size for all workloads. Although C++ has superior consistency and repeatability.
performance, it takes longer to develop since it needs The efficiency of mathematical models is increased
more code. Despite its lower performance, Python has by distributing calculations between processors. The
a large library ecosystem, which makes it perfect for ecosystem and the developer community provide
scientific computing’s quick development. access to documentation, training materials, and sup-
Support for mathematical operations and data types port forums. Community activity stimulates the con-
defines the basic functionality of a programming lan- tinuous development of libraries and tools. Package
guage. Built-in support for matrix manipulations, sta- support organisations contribute to the development
tistical functions, and numerical methods significantly and sustainability of software ecosystems.28
optimises the modelling process. Libraries like NumPy Integration capabilities with databases, visualisa-
provide high-level massive programming capabilities tion tools, and software optimise modelling workflows.
that simplify mathematical calculations.26 The avail- Python’s compatibility with various data formats and
ability of specialised libraries such as SciPy, Matplot- the ability to integrate with web applications and
lib, and TensorFlow minimises the cost of developing cloud services ensures its versatility. Cross-platform
mathematical models. The integration of programming compatibility and code portability minimise modifica-
with mathematical processes through a developed eco- tions when deployed on different operating systems.
system of libraries optimises the educational and prac- Python provides code execution on Windows, macOS,
tical aspects of programming in mathematics.27 and Linux, which optimises collaboration in distrib-
A more thorough understanding of language per- uted development teams. The speed of performing
formance in mathematical modelling has been made basic mathematical operations is a fundamental met-
possible by a number of peer-reviewed benchmarking ric for evaluating programming languages. The C and
studies. Li et al.1, for example, investigated perfor- C++ languages demonstrate high performance due to
mance across a range of numerical simulation tasks low-level memory access and efficient compilation into
and discovered that C++ routinely performs faster machine code. Extensive control over programme exe-
than Python and Julia, especially for high-complex- cution provides improved response time in computa-
ity, resource-intensive tasks like dynamic system tional tasks.29 Memory efficiency has a critical impact
simulations and large-scale matrix operations. In a on performance when processing large datasets. C and
similar vein, Shatyrko6 highlighted the benefits of C++ provide detailed control over memory manage-
C++ in dynamic system modelling, pointing out the ment. Rust implements the concepts of ownership and
increased computational efficiency that results from borrowing to prevent errors while maintaining high
low-level optimisation and direct memory manage- performance.30
ment. However, Julia is especially promising because it Compiler optimisations, including loop unrolling
balances the computational efficiency of C++ with the and dead code elimination, increase the efficiency
speed of Python development. Julia’s JIT compilation of the generated code. The compiler infrastructure
enables it to run at speeds comparable to C++ while plays a key role in optimising high-level programmes.

8 DOI: [Link] | Premier Journal of Science 2025;14:100173

PREMIER JOURNAL OF SCIENCE REVIEW

The implementation of mathematical algorithms in Error tolerance of algorithms is a prerequisite for

computational mathematics plays a key role, deter- their industrial application.38–40 In strongly typed lan-
mining the effectiveness, accuracy, and applicability of guages (Rust), many errors, including going beyond
mathematical models. The implementation process is the boundaries of the array, are caught at the compi-
influenced by the features of programming languages, lation stage. In Python, ZeroDivisionError or ValueEr-
including the syntax of mathematical expressions, work- ror exceptions allow localising problems in numerical
ing with numeric types, support for vector operations, methods. In distributed systems, numerical stability
implementation of recursion, error handling, metapro- monitoring is critically important, including NaN or
gramming, and integration with external libraries. overflow detection through assert mechanisms. In sci-
The way mathematical expressions are written in code entific computing, methods of “soft” recovery are used,
directly affects readability and speed of development. for example, matrix regularisation under degeneracy.
Python (with the SymPy library) and MATLAB use infix Metaprogramming opens up opportunities for cre-
notation and symbolic calculations, which make it ating domain-specific languages adapted to mathe-
possible to transfer formulas from research papers into matical problems.41,42 The TensorFlow library uses
code almost verbatim. For example, the expression metaprogramming to build computational graphs,
x = (-b+sqrt(b2-4ac))/(2a) has an intuitive notation. while SymPy generates optimised C code from sym-
In low-level languages such as C, similar operations bolic expressions. In Julia, macros allow converting
require function calls (for example, pow(b, 2)). A clear LaTeX formulas into executable code, which optimises
syntax is critical for implementing recursive algo- the documentation of models. C++ templates signifi-
rithms in conditions of limited memory, where format- cantly speed up linear algebra by eliminating runtime
ting errors can lead to leaks.31 Julia demonstrates an overhead.
effective compromise by combining Python’s concise- When choosing tools for mathematical modelling
and data analysis, it is important to consider a set of cri-
ness with C’s performance.
teria that determine the effectiveness and convenience
The choice of numeric data types determines not
of work. One of the main aspects is the completeness of
only the accuracy but also the stability of algorithms.
mathematical functionality. For example, the NumPy
Python automatically uses arbitrary precision arith-
and SciPy libraries in Python provide ready-made solu-
metic for integers but uses standard doubles (64 bits)
tions for linear algebra, optimisation, and statistics,
for fractional numbers, which can cause errors to
which speeds up the development of complex algo-
accumulate. In scientific calculations in C++, it is pos-
rithms.32,43 The performance of implementations also
sible to specify exact types (for example, float128) to
plays a key role: optimised libraries such as Intel MKL
control accuracy. The Gauss-Seidel algorithm shows
or CuPy are able to speed up calculations tenfold by
that rounding errors in matrix operations can distort
using multi-core processors and GPUs, which is critical
the result, especially when working with poorly con-
for processing large data.44–46
ditioned matrices.32–34 For financial calculations in
Equally important is the availability of high-quality
Python, the decimal module is effectively used to avoid
documentation with usage examples, which lowers
binary float errors. the entry threshold for new users and simplifies debug-
Modern languages offer built-in tools for working ging. For example, projects with detailed manuals (like
with multidimensional data.35 The NumPy library TensorFlow or PyTorch) often become industry stan-
in Python allows performing operations on arrays dards. The stability and reliability of libraries directly
without explicit loops, optimising the code through affect the reproducibility of results: outdated or rarely
vectorisation. Julia has similar features built in at updated tools (for example, some Perl modules) may
the language level, including syntax for matrix mul- cause errors due to version incompatibilities.47,48
tiplication (A*B) and function translation. Efficient Cross-platform compatibility remains critical for
implementation of finite element methods requires research: libraries like OpenCV or SQLAlchemy sup-
optimised operations with sparse matrices, which port Windows, Linux, and macOS, which makes it
are provided by libraries like Eigen (C++) or CuPy. easier to collaborate in disparate teams.49–52 Special
The specialised languages R and APL are focused on attention should be paid to tools for niche tasks, such
statistics and tensor computing. as the Astropy library for astrophysics or BioPython for
Recursion greatly simplifies the implementation of bioinformatics, which offer specialised functions that
quicksort or tree traversal algorithms but requires care- are not available in universal solutions.
ful control over stack depth and memory.36,37 Haskell
and Erlang languages support tail recursion optimi- Practical Aspects of the Implementation of
sation by converting it to iteration. In Python, where Mathematical Models
the call stack is limited, recursive methods for facto- The implementation of a mathematical model using
rial calculation tasks can lead to overflow. The hybrid a programming language is a multi-step process that
approach combines recursion for code clarity with a requires a systematic approach and attention to detail.
shift to iterative methods when working with big data. Initially, it is necessary to formalise the mathemati-
Karatsuba’s algorithm for multiplying large numbers cal description of the problem by converting the real
demonstrates an effective combination of recursion problem into a system of equations, variables, and con-
and memorisation to reduce overhead. straints. This includes analysing existing approaches,

DOI: [Link] | Premier Journal of Science 2025;14:100173 9

PREMIER JOURNAL OF SCIENCE REVIEW

clarifying parameters, and eliminating ambiguities. In the context of containerisation, a multi-level

For example, modelling physical processes may approach to environment isolation demonstrates the
require differential equations, and economic systems greatest efficiency. Docker provides a basic level of
may require optimisation models based on stochastic isolation, creating reproducible runtimes with fixed
factors.53–55 At this stage, it is important to determine versions of libraries and dependencies. At the same
which aspects of the system will be abstracted and time, for high-performance computing, Singularity
which will be detailed to maintain a balance between provides more specialised capabilities that consider
accuracy and computational complexity. the specifics of scientific calculations and direct access
The next step is to select numerical methods to hardware resources.
adapted to the type of model and the requirements The organisation of data processing requires a differ-
of the project. Runge-Kutta methods can be used for entiated approach to the choice of formats for storing
dynamic systems, and gradient algorithms or genetic and transmitting information. For large-volume tabu-
methods can be used for optimisation problems. It lar data, the Apache Parquet format provides efficient
is critically important to consider the stability of the column storage with partial readability and predictive
methods, their convergence, and their computational compression. Scientific data with a complex hierarchi-
efficiency. For example, when working with big data, cal structure is efficiently processed using the Hierar-
preference may be given to algorithms with linear com- chical Data Format version 5 (HDF5) format, which
plexity, and in machine learning tasks, to methods that provides metadata and group organisation capabili-
support parallel computing.56,57 Modern programming ties. Cross-system data exchange is optimised through
languages such as Python or Julia offer rich librar- the use of Apache Arrow, which provides a unified
ies (SciPy, TensorFlow, [Link]) that representation of data in memory for various pro-
simplify the integration of ready-made solutions and gramming languages and platforms. The integration
reduce the risk of errors. of artificial intelligence methods into mathematical
Algorithmic implementation requires compliance modelling requires a balanced approach, considering
with the principles of clean code and modularity.58 the specifics of the tasks being solved. In tasks with
Dividing the model into components (data entry, pre- unstructured or noisy data, neural networks demon-
processing, calculations, visualisation) simplifies strate high efficiency in preprocessing and identifying
testing and reuse. Integration with frameworks such significant features.63 Classical numerical methods
as PyTorch for neural networks or Apache Spark for provide more reliable and interpretable results for sys-
distributed computing expands the capabilities of tems with well-defined physical laws.
the model. Pseudocode and flowcharts help plan the The implementation of mathematical models in
architecture before writing the code, and tools like practical applications involves a number of difficulties
Jupyter Notebook allow interactively checking individ- that require an integrated approach to their solution.
ual blocks.59 Special attention should be paid to han- One of the key issues is computational limitations
dling exceptions and edge cases, such as division by related to performance, memory consumption, and
zero or variable overflow. calculation accuracy. For example, unoptimised algo-
Testing and optimisation include unit tests to verify rithms or the use of resource-intensive methods can
the correctness of individual functions and load tests slow down the execution of the model, especially
to evaluate performance. Code profiling using tools when working with big data. To increase efficiency,
like cProfile in Python or Valgrind for C++ identifies code profiling is used using tools such as cProfile in
bottlenecks such as redundant loops or suboptimal Python or VTune for C++, which allows identifying
memory requests. Visualisation of the results (Mat- “bottlenecks”. Switching to parallel computing via
plotlib graphs, Plotly diagrams) helps to analyse the libraries (Dask, MPI) or using GPU acceleration (CUDA,
behaviour of the model and detect anomalies. For OpenCL) significantly speeds up processing. The prob-
resource-intensive tasks, parallelisation of calcula- lem of memory consumption is solved by optimising
tions using multithreading (threading module) or GPU data structures: replacing lists with NumPy arrays in
acceleration (CUDA, OpenCL) is relevant. Python reduces overhead, and in low-level languages
Documentation includes descriptions of the model such as Rust, memory management through the own-
architecture, input parameters, data formats, and ership system prevents leaks. To increase the accu-
usage scenarios.60 Documentation generators (Sphinx, racy of calculations, especially in problems with high
Doxygen) automate the creation of reference materials, sensitivity to rounding errors, libraries with support
and interactive tutorials in Jupyter Notebook make the for arbitrary precision arithmetic (GMP, Decimal in
manual visual. Logging changes through Git ensures Python) are used, or double-precision floating-point
transparency of development, and platforms like numerical types (double) are used.
GitHub or GitLab simplify collaboration. Checklists Integration difficulties arise when the model inter-
and code templates reduce the risk of human error acts with external systems and data. Incompatibility
when making edits. of formats (CSV, JSON, XML) can lead to parsing errors
The practical implementation of mathematical mod- or loss of information. The solution is to standardise
els requires careful selection of containerisation tools data exchange through universal formats such as Par-
and data processing formats and an assessment of the quet for tabular data or use intermediate layers (API,
feasibility of using artificial intelligence ­methods.61,62 Apache Kafka for streaming processing). For example,

10 DOI: [Link] | Premier Journal of Science 2025;14:100173

PREMIER JOURNAL OF SCIENCE REVIEW

in Python, the Pandas library supports reading data the impact of deforestation and ocean pollution. Bio-
from 15+ formats, and tools like Apache Arrow provide inspired algorithms, such as neural networks and opti-
cross-language compatibility. Another problem is the mization methods based on ant behavior, are being
integration of the model into existing IT infrastruc- explored for use in robotics and power system manage-
tures. Microservice architecture and containerisation ment. Additionally, there is growing interest in models
(Docker, Kubernetes) help here, which simplify deploy- that can dynamically adapt to changing conditions,
ment and scaling. Web applications use frameworks as seen in autonomous transportation systems that
like Flask or FastAPI, which provide RESTful interfaces update algorithms based on real-time sensor data.
for interacting with the model. To fully leverage these advancements, the standard-
Methodological issues include the selection of an isation of tools and data is crucial. Open libraries like
adequate model, validation of results, and consider- TensorFlow for machine learning and OpenFOAM for
ation of uncertainties. For example, in machine learn- fluid dynamics should form the foundation for collab-
ing tasks, retraining a model based on training data oration across scientific communities. Data exchange
leads to incorrect predictions based on new data.64,65 formats such as HDF5 and Systems Biology Markup
The solution is cross-validation, regularisation, and the Language (SBML) are helping to ensure the reproduc-
use of synthetic data for testing. In physical modelling, ibility of research. The adoption of FAIR (Findable,
errors can occur due to simplifications in equations. Accessible, Interoperable, Reusable) principles will
Verification methods are used here through compari- be essential to ensure that models are transparent and
son with analytical solutions or benchmark tests (for accessible, including publishing not only the source
example, the Taylor-Green problem in hydrodynam- code but also detailed metadata that describes the
ics). Monte Carlo methods or Bayesian approaches experimental conditions.
implemented in the PyMC3 or TensorFlow Probability Ethical considerations are increasingly important in
libraries are used to account for uncertainties. Automa-
model development. Systems that influence decisions
tion of testing through CI/CD (GitHub Actions, GitLab
in fields such as healthcare, finance, or law need to be
CI) and frameworks (pytest, unittest) minimises the
assessed for potential biases in data and algorithms.
risks of regression during code modifications.
Methods like Explainable AI (XAI), including Local
Thus, successful model implementation requires not
Interpretable Model-agnostic Explanations (LIME) and
only mathematical rigour but also sound ­engineering
SHapley Additive exPlanations (SHAP), can enhance
practice, from optimising code to building reliable data
the transparency and fairness of decision-making
pipelines. The use of modern languages and tools (Julia
models by visualising and explaining algorithmic
for scientific calculations and Apache Spark for Big
logic. These methods aim to mitigate bias, improve
Data) reduces development time, and the openness of
accountability, and ensure that models remain inter-
the methodology strengthens confidence in the results.
pretable. However, the development of standards for
the certification of AI systems is still in its early stages,
Prospects for the Development of Mathematical
particularly in regions such as the EU, which is explor-
Modelling
ing regulatory frameworks that may set the stage for
Modern mathematical modelling is undergoing a sig-
global adoption.
nificant transformation, driven by advancements in
Collaboration between academia, industry, and
technology and the increasing demands of science
and industry. One of the key trends is the integration government is essential to accelerate innovation. Pro-
of classical methods with AI. For instance, hybrid grammes like the European Human Brain Project bring
approaches that combine Navier-Stokes equations together experts from diverse fields to create digital
with deep learning are being explored in aerodynam- brain models, while open competitions on platforms
ics to predict turbulence in real time, an important like Kaggle and DrivenData invite global talent to tackle
step for drone design. The use of cloud platforms and pressing issues like forest fire prediction and supply
distributed computing, such as Apache Hadoop and chain optimisation. The rise of Industry 4.0, driven by
Kubernetes, is expanding, enabling models to process the Internet of Things and digital twins, is pushing the
large datasets in fields like genomics and astrophys- integration of models into real-world production sys-
ics. While quantum computing holds the potential to tems, contributing to the development of edge comput-
address optimisation problems beyond the scope of ing, where data is processed directly on devices rather
classical methods, its practical application remains a than relying on cloud servers.
future prospect. Thus, the future of mathematical modelling is deter-
New interdisciplinary research areas are emerg- mined not only by technological progress but also by
ing, challenging the boundaries between traditional the ability of the scientific community to adapt. The
fields. For example, digital twins, initially developed development of quantum technologies, interdisciplin-
for industry, are now being adapted for medical use, ary research, and an ethically responsible approach
where virtual copies of human organs could predict create the basis for solving problems that until recently
responses to treatments and simulate surgical proce- were considered fantastic. From modelling quantum
dures. In environmental modelling, multidisciplinary materials to predicting social crises, these tools are
systems that integrate climate forecasts, economic becoming the key to sustainable development in an era
models, and satellite data are being used to evaluate of global challenges.

DOI: [Link] | Premier Journal of Science 2025;14:100173 11

PREMIER JOURNAL OF SCIENCE REVIEW

Conclusions References
This review offers a thorough summary of the chang- 1 Li J, Sun M, Dong Y, Xu F, Sun X, Zhou Y. The mediating effect of
creativity on the relationship between mathematic achievement
ing trends and approaches in mathematical modelling, and programming self-efficacy. Front Psy. 2022;12:772093.
based on an examination of the relationship between [Link]
mathematical modelling and programming languages. 2 Bellingeri A, Gallo A, Liang D, Masoero F, Cabrera VE.
The combination of hybrid algorithms and AI with tra- Development of a linear programming model for the optimal
allocation of nutritional resources in a dairy herd. J Dairy Sci.
ditional mathematical techniques is one of the biggest 2020;103(11):10898–10916. [Link]
developments, since it has created new opportunities jds.2020-18157
for more precise and efficient problem-solving. The 3 Gholamnejad J, Lotfian R, Kasmaeeyazdi S. A practical, long-
term production scheduling model in open pit mines using
potential of quantum computing and the enhanced
integer linear programming. J South African Inst Mining Metal.
processing capacity offered by cloud technologies 2020;120(12):665–670. [Link]
significantly expand the capabilities of mathematical 9717/769/2020
models, making it possible to address more challeng- 4 Bhamare DMA. Optimizing crop selection and production
planning in agriculture: Applications of linear programming
ing issues in a variety of scientific domains. for profit maximization and sustainability. Int J Res Pub
A comparative analysis of programming languages Rev. 2023;4(6):3281–3287. [Link]
has revealed the advantages of Python and Julia for gengpi.4.623.46630
research tasks due to their ecosystems, including spe- 5 Chukwu C, Nyabadza F, Asamoah J. A mathematical model and
optimal control for Listeriosis disease from ready-to-eat food
cialised libraries (NumPy, TensorFlow), which speeds products. Int J Comp Sci Math. 2023;17(1):39. [Link]
up prototyping. C++ retains its leadership in high- org/10.1504/ijcsm.2023.130421
performance computing, which is critically import- 6 Shatyrko A. Some methodological aspects of mathematical
modeling in dynamic systems. Int Sci J Eng & Agric. 2024;3(1):8–
ant for tasks requiring big data processing. The
14. [Link]
review also highlights how important software 7 Taloub D, Bouras A, Driss Z. Modeling and numerical solution
implementation is in determining how effective and of the Laplace equation in 2D by the finite difference method
applicable mathematical models are. In order to case of the heat equation – Study of stability. 2023. [Link]
org/10.21203/[Link]-2440144/v1
maximise the performance of these models, partic-
8 Fitriah F, Akorede A, Agyei E. Improving mathematics learning
ularly in resource-demanding domains like physical outcomes through the consideration model for class VII students.
simulations and AI-driven analytics, the program- Interval: Indonesian J Math Edu. 2023;1(2):47–55. [Link]
ming language selection and the underlying compu- org/10.37251/ijome.v1i2.771
9 Afrilianto M, Rosyana T, Linda L, Wijaya TT. Project-activity-
tational architecture are crucial. cooperative learning-exercise model in improving students’ creative
The prospects for the development of mathematical thinking ability in mathematics. Inf J. 2022;11(2):285–296.
modelling are related to the integration of quantum [Link]
10 Tricco AC, Lillie E, Zarin W, O’Brien KK, Colquhoun H, Levac D, et al.
computing, which opens up new opportunities for
PRISMA Extension for Scoping Reviews (PRISMA-ScR). Ann Intern
solving optimisation problems, and the strengthening Med. 2018;169(7):467–473. [Link]
of the role of AI in refining models based on noisy data. 11 Purnomo YW, Nabillah R, Aziz TA, Widodo SA. Fostering
These areas require an interdisciplinary approach, mathematical connections and habits of mind: A problem-based
learning module for elementary education. Inf J. 2024;13(2):333–
including cooperation between scientific communi-
348. [Link]
ties, business, and government. Ethical responsibility 12 Kusmaryono I, Aminudin M, Ubaidah N, Chamalah E. The bridging
remains an important aspect, especially when using understanding of language and mathematical symbols between
AI algorithms in sensitive areas such as healthcare or teachers and students: An effort to increase mathematical literacy.
Inf J. 2024;13(1):251–270. [Link]
finance. v13i1.p251-270
The limitations of the study are conditioned by the 13 Putrawan AA, Ayuni NWD. Tetra-helix concept model based on
dynamic development of technologies, which requires vocational realistic education (VRE). In: R.C.I. Prahmana, M. Shahrill,
H. Julie, A. Lukito, S.J. Andajani (Eds.), Proceedings of the Eighth
constant updating of methodologies, and the depen-
Southeast Asia Design Research (SEA-DR) & the Second Science,
dence of the results on the quality of the source data Technology, Education, Arts, Culture, and Humanity (STEACH)
and the availability of computing power. To overcome International Conference (SEADR-STEACH 2021). Amsterdam, Paris:
these challenges, it is necessary to develop adaptive Atlantis Press; 2022. Pp. 179–183. [Link]
assehr.k.211229.024
algorithms and strengthen educational programmes
14 Aleeva VN, Aleev RZ. Parallelism resource of numerical algorithms.
aimed at training specialists who possess both math- Version 1. 2022. [Link]
ematical apparatus and programming skills. The 15 Motara YM. High-level modelling for typed functional programming.
obtained results emphasise the importance of further In: V. Zsók, J. Hughes (Eds.), Trends in Functional Programming:
22nd International Symposium, TFP 2021, Virtual Event, February
integration of theoretical and applied aspects of mod- 17-19, 2021, Revised Selected Papers. Cham: Springer; 2021. pp.
elling, which contributes to solving complex prob- 69–94. [Link]
lems in science, engineering, and the social sphere. 16 Davydov V, Hrebeniuk D. Development of the methods for
A promising direction for future research is to expand resource reallocation in cloud computing systems. Innov Tech Sci
Solut Industr. 2020;3(13):25–33. [Link]
the analysis to include additional programming lan- itssi.2020.13.025
guages and HPC paradigms, such as Fortran, MAT- 17 Swatthong N, Aswakul C. Optimal cloud orchestration model
LAB, R, Rust, and parallel computing frameworks of containerized task scheduling strategy using integer linear
programming: Case studies of iotcloudserve@tein project. Energies
like OpenMP/MPI/CUDA, as well as explore domain-
2021;14(15):4536. [Link]
specific applications like CFD, molecular simulation, 18 Ezzrhari FE, Noureddine EAA, Youssfi M, Bouattane O. Scalable
and econometrics. and reactive multi micro-agents system middleware for massively

12 DOI: [Link] | Premier Journal of Science 2025;14:100173

PREMIER JOURNAL OF SCIENCE REVIEW

distributed systems. Int J Adv Comp Sci Applic. 2021;12(11):415– the gasoline fraction hydrotreating process in a fuzzy environment.
426. [Link] Processes. 2024;12(4):669. [Link]
19 Yang H, Xiong S, Frimpong SA, Zhang M. A consortium blockchain- pr12040669
based agricultural machinery scheduling system. Sensors 40 Orazbayev B, Zhumadillayeva A, Kabibullin M, Crabbe MJC,
2020;20(9):2643. [Link] Orazbayeva K, Yue X. A systematic approach to the model
20 Zhao T. Artificial intelligence in mathematical modeling of complex development of reactors and reforming furnaces with fuzziness and
systems. EAI Endorsed Trans E-Learn. 2024;10. [Link] optimization of operating modes. IEEE Access. 2023;11:74980–
org/10.4108/eetel.5256 74996. [Link]
21 Liu D, Xiong J, Babaker S. Uniqueness of system integration 41 Ashirbaev B, Altymyshova Z, Alymbaeva Z. Optimal energy-saving
scheme of artificial intelligence technology in fractional differential control for a thermal plant of a linear singularly perturbed discrete
mathematical equation. Appl Math Nonlin Sci. 2022;8(1):1155– system with a small step. In: Proc Int Conf Electr Comput Energy
1166. [Link] Technol, ICECET 2023; Cape Town: IEEE; 2023. [Link]
22 Mata T. A modified ABC-PTS with non-uniform phase factor for org/10.1109/ICECET58911.2023.10389496
OFDM signal. Int J Intel Eng Syst. 2021;14(4):115–122. [Link] 42 Yakovlev SV, Valuiskaya OA. Optimization of linear functions at
org/10.22266/ijies2021.0831.11 the vertices of a permutation polyhedron with additional linear
23 Wang C, Zhang W, Jiang D, Tong Z, Wang H. PAPR reduction hybrid constraints. Ukr Math J. 2001;53(9):1535–1545. [Link]
scheme for VLC systems based on conventional algorithm and org/10.1023/A:1014374926840
neural network. In: W. Yu, A. Bhattacharjya, W. Hu (Eds.), Fourth 43 Cherniha R, King JR, Kovalenko S. Lie symmetry properties of
International Conference on Advanced Manufacturing Technology nonlinear reaction-diffusion equations with gradient-dependent
and Electronic Information (AMTEI 2024). Bellingham: SPIE; 2025. diffusivity. Commun Nonlinear Sci Numer Simul. 2016;36:98–108.
[Link] [Link]
24 Harris CR, Millman KJ, Walt SJVD, Gommers R, Virtanen P, 44 Kyurchev V, Kiurchev S, Rezvaya K, Fatyeyev A, Głowacki S.
Cournapeau D, et al. Array programming with numpy. Nature Assessing the reliability of a mathematical model of working
2020;585(7825):357–362. [Link] processes occurring in a hydraulic drive. In: Lect Notes Mech Engin;
020-2649-2 Cham: Springer; 2024. p. 281–292. [Link]
25 The computer language 25.03 benchmarks game. 2025. https:// 3-031-63720-9_24
[Link]/benchmarksgame/ 45 Kadenko IM, Sakhno NV, Biró B, Fenyvesi A, Iermolenko RV, Gogota
measurements/[Link] OP. A bound dineutron: Indirect and possible direct observations.
26 Gadanidis G, Silva RSRD, Hughes J, Floyd S, Namukasa IK. Acta Phys Pol B Proc Suppl. 2024;17(1):1A31–1A39. [Link]
Computational literacy & mathematics education. Int J Res Math org/10.5506/APhysPolBSupp.17.1-A3
Edu. 2022;12(4):1–23. [Link] 46 Yermolenko R, Falko A, Gogota O, Onishchuk Y, Aushev V.
v12i4.3144 Application of machine learning methods in neutrino experiments. J
27 Zimmermann T, Falleri J. A grounded theory of community package Phys Stud. 2024;28(3). [Link]
maintenance organizations-registered report. 2021. [Link] 47 Babak V, Kharchenko V, Vasylyev V. Using generalized stochastic
org/10.48550/arxiv.2108.07474 method to evaluate probability of conflict in controlled air traffic.
28 Costanzo M, Rucci E, Naiouf M, Giusti AD. Performance vs Aviat. 2007;11(2):31–36. [Link]
programming effort between Rust and C on multicore architectures: 07.9635958
Case study in n-body. 2021. [Link] 48 Luniov SV. Calculation of electron mobility for the strained
arxiv.2107.11912 germanium nanofilm. J Nano- Electr Phys. 2019;11(2):02023.
29 Yesylevskyy S. MolAR: Memory-safe library for analysis of MD [Link]
simulations written in rust. J Comp Chem. 2025;46(1):e27536. 49 Karymsakova IB, Krak IV, Denissova NF. Criteria for implants
[Link] classification for coating implants using plasma spraying by robotic
30 Zhu H. Low-memory implementations of ridge solutions for broad complex. Euras J Math Comput Appl. 2017;5(3):44–52. [Link]
learning system with incremental learning. 2021. [Link] org/10.32523/2306-3172-2017-5-3-44-52
org/10.48550/arxiv.2105.10424 50 Kerimkhulle S, Aitkozha Z. A criterion for correct solvability of a first
31 Hatun M. Identification of Wiener systems with recursive Gauss- order difference equation. AIP Conf Proc. 2017;1880:040016.
Seidel algorithm. Electr Electric Eng. 2023;29(5):4–10. [Link] [Link]
org/10.5755/[Link].35119 51 Karaiev O, Bondarenko L, Halko S, Miroshnyk O, Vershkov O,
32 Zharlykasov BZh, Alimbaev AA, Kalzhanov MU, Telegina OS. Karaieva T, et al. Mathematical modelling of the fruit-stone culture
Modern computer methods in teaching mathematics and physics: seeds calibration process using flat sieves. Acta Technol Agric.
Exam using Python. Bull Toraighyrov Uni. Physics & Math Ser. 2021;24(3):119–123. [Link]
2023;4(2023):132–143. [Link] 52 Beisenbi M, Kaliyeva S. Synthesis of the control systems by the state
33 Amourah A, Frasin BA, Salah J, Al-Hawary T. Fibonacci of an object with single input and single output by a gradient-
numbers related to some subclasses of bi-univalent functions. velocity method of A.M. Lyapunov vector functions. Int J Civ Eng
Int J Math Math Sci. 2024;2024:8169496. [Link] Technol. 2018;9(10):2080–2086.
org/10.1155/2024/8169496 53 Sagari SM, Malagi VP, Sasi S. Euri – a deep ensemble architecture
34 Amourah A, Al-Hawary T, Yousef F, Salah J. Collection of bi-univalent for oral lesion segmentation and detection. Int. J. Intell. Syst. Appl.
functions using Bell distribution associated with Jacobi polynomials. Eng. 2024;12(3s):242–249.
Int J Neutrosophic Sci. 2025;25(1):228–238. [Link] 54 Zelisko VR, Kuchma MI. Factorization of symmetric matrices over
org/10.54216/IJNS.250121 polynomial rings with involution. J Math Sci. 1999;96(2):3017-
35 Avrunin OG, Tymkovych MY, Pavlov SV, Timchik SV, Kisała P, 3021. [Link]
Orakbaev Y. Classification of CT-brain slices based on local 55 Cherniha R, Serov M, Rassokha I. Lie symmetries and form-
histograms. Proc SPIE – Int Soc Optic Engin. 2015;9816:98161J. preserving transformations of reaction-diffusion-convection
[Link] equations. J Math Anal Appl. 2008;342(2):1363–1379. https://
36 Dmytruk AA, Gatalevych AI, Kuchma MI. Stable range conditions for [Link]/10.1016/[Link].2008.01.011
Abelian and duo rings. Matematych Studii. 2022;57(1):92–97. 56 Ushenko YA, Dubolazov AV, Karachevtcev AO, Zabolotna NI. A
[Link] fractal and statistic analysis of Mueller-matrix images of phase
37 Kal’chuk IV, Kravets VI, Hrabova UZ. Approximation of the classes inhomogeneous layers. Proc SPIE. 2011;8134:81340P. [Link]
WβrHα by three-harmonic Poisson integrals. J Math Sci (United org/10.1117/12.891812
States). 2020;246(1):39–50. [Link] 57 Chuzlov VA, Nazarova GY, Dolganov IM, Dolganova IO, Seitenova
020-04721-4 GZh. Calculation of the optimal blending component ratio by
38 Nazarova G, Ivashkina E, Ivanchina E, Shafran T, Stebeneva using mathematical modeling method. Petrol Sci Technol.
V, Seytenova G. Testing of cracking zeolite catalysts using 2019;37(10):1170–1175. [Link]
mathematical model. AIP Conf Proc. 2016;1772:060013. https:// 019.1578800
[Link]/10.1063/1.4964593 58 Ivanchina ED, Chuzlov VA, Ivanchin NR, Borissov A, Seitenov
39 Orazbayev B, Tanirbergenova A, Orazbayeva K, Berikbaeva M, GZ, Dusova RM. Mathematical modeling of the process catalytic
Kaliyeva S, Kurmangaziyeva L, et al. Decision making for control of isomerization of light naphtha. Petrol Coal. 2019;61(2):413–417.

DOI: [Link] | Premier Journal of Science 2025;14:100173 13

PREMIER JOURNAL OF SCIENCE REVIEW

59 Al-Hawary T, Amourah A, Salah J, Yousef F. Two inclusive subfamilies Syst. Comput. 2019;754:545–554. [Link]
of bi-univalent functions. Int J Neutrosophic Sci. 2024;24(4):315– 3-319-91008-6_55
323. [Link] 63 Babichev SA, Gozhyj A, Kornelyuk AI, Lytvynenko VI. Objective
60 Amourah A, Frasin B, Salah J, Yousef F. Subfamilies of bi-univalent clustering inductive technology of gene expression profiles based
functions associated with the imaginary error function and on SOTA clustering algorithm. Biopolym. Cell. 2017;33(5):379–
subordinate to Jacobi polynomials. Symmetry. 2025;17(2):157. 392. [Link]
[Link] 64 Imamguluyev R, Umarova N. Application of fuzzy logic apparatus
61 Orazbayev B, Kozhakhmetova D, Orazbayeva K, Utenova B. Approach to solve the problem of spatial selection in architectural-design
to modeling and control of operational modes for chemical and projects. Lect. Notes Netw. Syst. 2022;307:842–848. [Link]
engineering system based on various information. Appl. Math. Inf. Sci. org/10.1007/978-3-030-85626-7_98
2020;14(4):547–556. [Link] 65 Nekrasov S, Peterka J, Zhyhylii D, Dovhopolov A, Kolesnyk V.
62 Babichev S, Lytvynenko V, Gozhyj A, Korobchynskyi M, Voronenko M. Mathematical estimation of roughness Rz of threaded surface
A fuzzy model for gene expression profiles reducing based on the obtained by machining method. MM Sci. J. 2022;June:5699-5703.
complex use of statistical criteria and Shannon entropy. Adv. Intell. [Link]

14 DOI: [Link] | Premier Journal of Science 2025;14:100173

PREMIER JOURNAL OF SCIENCE REVIEW

Appendix A

Table A1 | Summary of key characteristics and methodologies of included studies on mathematical modelling and programming languages
Authors Focus Area Key Methodology Study Design Key Findings Additional Characteristics
1
Li et al. Mathematical Survey & Statistical Cross-sectional Found a positive correlation between Quantitative, Survey-Based, Focus
Achievemen1t Analysis mathematical skills and programming on Creativity as a Mediator
effectiveness
Bellingeri et al.2 Resource Linear Mathematical Optimised nutritional resources for dairy herd Applied to Agriculture,
Optimisation Programming modelling management Computational Model
Gholamnejad et al.3 Resource Integer Linear Mathematical Developed scheduling model for long-term mine Industrial Applications,
Optimisation Programming modelling operations Computational Optimisation
Bhamare4 Agricultural Linear Case study Optimised crop selection and production Focus on sustainability, profit
Optimisation Programming planning for profitability maximisation
Chukwu et al.5 Public Health Mathematical System dynamics Proposed optimal control strategies for Listeriosis Public Health, Disease Modeling
Modelling prevention
Shatyrko6 Dynamic Systems Numerical & Literature review Provided insights into modelling methods for Conceptual Framework,
Analytical Methods dynamic systems Methodological Review
Taloub et al.7 Numerical Finite Difference Experimental Studied stability and numerical solutions for Focus on Numerical Methods,
Methods Method study Laplace equations Stability Analysis
Fitriah et al.8 Education Model-Based Experimental Improved mathematics outcomes using Educational Intervention, Student
Learning study structured learning models Learning Focus
Afrilianto et al.9 Education Cooperative Intervention Enhanced creative thinking in mathematics Focus on Active Learning,
Learning study through collaborative learning Education Model
Kusmaryono et al.12 Educational Cross-Linguistic Qualitative Enhanced mathematical literacy by bridging Cross-Linguistic Focus, Education
Methods Analysis language and symbol understanding Analysis
Putrawan & Ayuni13 Education Conceptual Model Case study Proposed new educational model to integrate Vocational Education, Conceptual
vocational skills Framework
Aleeva & Aleev14 Numerical Algorithm Theoretical study Focused on improving resource allocation in Resource Optimisation, Parallel
Methods Optimisation parallel computation Algorithms
Motara15 Functional Functional Theoretical study Improved high-level modelling for typed Focus on Functional Programming,
Programming Programming functional programming Algorithm Design
Davydov & Cloud Computing Resource Case study Proposed methods for resource optimisation in Cloud Computing Focus, Resource
Hrebeniuk16 Reallocation cloud systems Management
Swatthong & Cloud Computing Cloud Experimental Optimised cloud task scheduling using integer Cloud Orchestration,
Aswakul17 Orchestration study linear programming Containerisation
Ezzrhari et al.18 Distributed Middleware Experimental Developed middleware for scalable multi-agent Distributed Systems, Middleware
Systems Optimisation study systems Design
Zhao20 AI & AI Integration Systematic Analysed AI’s role in solving complex system AI Integration, Systematic Review
Mathematical review modelling problems
Modelling
Liu et al.21 AI & AI System Conceptual study Focused on integrating AI technology into AI Integration, Advanced
Mathematical Integration fractional differential equations Mathematical Modeling
Modelling
Hatun31 Mathematical Recursive Experimental Explored Wiener systems identification using Recursive Algorithms, System
Algorithms Algorithms study recursive methods Identification
Zharlykasov et al.32 Education & Python Application Case study Applied modern computing methods to enhance Technology in Education, Python
Technology teaching in math and physics Programming
Amourah et al.33 Mathematical Function Analysis Analytical study Investigated properties of Fibonacci numbers in Mathematical Analysis, Function
Functions bi-univalent functions Theory
Costanzo et al.28 Programming Performance Comparative Compared performance between Rust and C in Performance Benchmarking,
Performance Benchmarking study multicore architectures Multicore Computing

DOI: [Link] | Premier Journal of Science 2025;14:100173 15

PREMIER JOURNAL OF SCIENCE REVIEW

Appendix B

Table A2 | Included and excluded studies with reasons

Authors Year Title of the Work Inclusion Status Reason for Inclusion/Exclusion
Li et al. 2022 The mediating effect of creativity on the relationship Included Addressed the relationship between mathematical skills
between mathematic achievement and programming self- and programming effectiveness; relevant to modelling
efficacy using programming languages.
Bellingeri et al. 2020 Development of a linear programming model for the optimal Included Applied linear programming for resource optimisation in
allocation of nutritional resources in a dairy herd agriculture using Python/C++.
Gholamnejad 2020 A practical, long-term production scheduling model in open Included Developed integer linear programming models for mining
et al. pit mines using integer linear programming operations; computational optimisation focus.
Bhamare 2023 Optimizing crop selection and production planning in Included Used mathematical programming for agricultural
agriculture sustainability and profit maximisation.
Chukwu et al. 2023 A mathematical model and optimal control for Listeriosis Included Integrated hybrid mathematical modelling with
disease from ready-to-eat food products programming for public health optimisation.
Shatyrko 2024 Some methodological aspects of mathematical modeling in Included Discussed methodological aspects of modelling dynamic
dynamic systems systems through analytical and numerical methods.
Taloub et al. 2023 Modeling and numerical solution of the Laplace equation in Included Provided finite difference numerical solutions for Laplace
2D by the finite difference method equations; experimental programming context.
Fitriah et al. 2023 Improving mathematics learning outcomes through the Included Used model-based learning to improve mathematics
consideration model for class VII students education outcomes via programming.
Afrilianto et al. 2022 Project-activity-cooperative learning-exercise model in Included Demonstrated cooperative learning approaches in
improving students’ creative thinking ability in mathematics mathematical modelling education.
Zhao 2024 Artificial intelligence in mathematical modeling of complex Included Focused on artificial intelligence integration in
systems mathematical modelling.
Liu et al. 2022 Uniqueness of system integration scheme of artificial Included Examined fractional differential equations using AI and
intelligence technology in fractional differential computational techniques.
mathematical equation
Aleeva & Aleev 2022 Parallelism resource of numerical algorithms Included Analysed numerical algorithms for parallel computation
optimisation.
Davydov & 2020 Development of methods for resource reallocation in cloud Included Modelled resource reallocation in cloud computing
Hrebeniuk computing systems systems.
Swatthong & 2021 Optimal cloud orchestration model of containerized task Included Applied integer linear programming to optimise
Aswakul scheduling strategy containerised task scheduling in cloud environments.
Ezzrhari et al. 2021 Scalable and reactive multi micro-agents system middleware Included Developed scalable middleware systems for distributed
for massively distributed systems multi-agent computation.
Costanzo et al. 2021 Performance vs programming effort between Rust and C on Included Compared Rust and C performance for n-body
multicore architectures simulations; benchmarking relevance.
Zharlykasov 2023 Modern computer methods in teaching mathematics and Included Analysed educational applications of Python in physics
et al. physics: Exam using Python and mathematics instruction.
Motara 2021 High-level modelling for typed functional programming Included Explored high-level functional programming methods in
mathematical model design.
Harris et al. 2020 Array programming with NumPy Included Described NumPy’s role in array programming for
scientific modelling.
Amourah et al. 2024 Fibonacci numbers related to some subclasses of bi- Included Conducted analytical study of Fibonacci-related
univalent functions mathematical functions.
Orazbayev et al. 2020 Approach to modeling and control of operational modes for Included Proposed fuzzy optimisation models for control processes
chemical and engineering systems in engineering systems.
Orazbayev et al. 2023 A systematic approach to the model development of reactors Included Developed systematic models for reactors and reforming
and reforming furnaces with fuzziness and optimization of furnaces; programming implementation focus.
operating modes
Chuzlov et al. 2019 Calculation of the optimal blending component ratio by Included Used mathematical modelling for chemical blending
using mathematical modeling method optimisation.
Ivanchina et al. 2019 Mathematical modeling of the process catalytic Included Presented catalytic isomerisation modelling of light
isomerization of light naphtha naphtha using computational methods.
Yang et al. 2020 A consortium blockchain-based agricultural machinery Included Applied blockchain-based modelling to agricultural
scheduling system machinery scheduling.

(Continued)

16 DOI: [Link] | Premier Journal of Science 2025;14:100173

PREMIER JOURNAL OF SCIENCE REVIEW

Table A2 | Continued
Authors Year Title of the Work Inclusion Status Reason for Inclusion/Exclusion
Alvarez & 1998 Computational approaches to nonlinear dynamics in Excluded Predates 2000 cut-off for eligible studies.
Mendez classical systems
Petrov & 2003 Analytical foundations of the variational method in Excluded Purely theoretical study; no programming
Andreev mathematics implementation.
Kimura et al. 2012 Hardware optimisation for autonomous robotic systems Excluded Focused on hardware engineering; not mathematical
modelling.
Singh & Patel 2014 Applications of data mining in social networks Excluded No computational modelling or programming integration.
Martínez et al. 2011 Philosophical perspectives on mathematics education Excluded Lacks empirical or computational content.
Ivanov et al. 2015 Algorithmic methods in non-English mathematical Excluded Language criterion not met (Russian).
publications
Kwon & Park 2017 Algorithmic pedagogy in STEM education Excluded Pedagogical overview without empirical modelling
implementation.
Chavez et al. 2010 Software project management models in academic Excluded Project management context; not mathematical
environments modelling.
Nguyen et al. 2019 User interface frameworks for cross-platform applications Excluded Focused on UI development only; no mathematical
frameworks.
Rossi et al. 2016 Semiotics and mathematical aesthetics in teaching Excluded Theoretical discussion without modelling or coding
component.
Al-Mutairi 2007 Education policy reform in STEM curricula Excluded Not connected to mathematical modelling or
programming.
Chen & Luo 2018 Survey on cloud infrastructure deployment models Excluded Did not address mathematical modelling or computation.
Garcia et al. 2020 Digital literacy and computational thinking in education Excluded Conceptual paper without data or modelling examples.
Takahashi et al. 2021 Statistical education and curricular innovation Excluded No use of programming languages in modelling.
Morozov & 2022 Symmetry and nonlinear transformations in pure Excluded Pure mathematical analysis only; no computational
Kalinina mathematics aspect.
Santos et al. 2023 Quantitative models for social decision making without Excluded Methodological detail insufficient; no code or data
algorithmic implementation provided.
Gray & Walters 2019 Emerging technologies in science education – Policy report Excluded Grey literature; not peer-reviewed.
Tanaka & 2024 Machine learning engineering frameworks for industry Excluded Industrial engineering focus; not mathematical modelling.
Ishikawa
Abebe et al. 2020 Open data resources for computational science Excluded Dataset description paper; no applied modelling.
Schmidt et al. 2021 Physics curricula and learning outcomes in STEM universities Excluded Educational review without modelling application.

DOI: [Link] | Premier Journal of Science 2025;14:100173 17

PREMIER JOURNAL OF SCIENCE REVIEW

Appendix C

Table A3 | Comprehensive summary of studies on mathematical modelling using programming languages:

techniques, performance metrics, and application fields
Study Modelling Technique Programming Application Field Hybrid Key Findings
Language(s) Method used
Li et al.1 Numerical Python, C++, Julia Particle Physics Machine Optimised resource
Learning allocation
Bellingeri et al.2 Linear Programming Python, C++ Agriculture No Optimal allocation in dairy
farming
Gholamnejad Integer Linear C++, Python Mining No Production scheduling
et al.3 Programming optimisation
Bhamare4 Linear Programming Python Agriculture No Maximised profit and
sustainability
Chukwu et al.5 Mathematical Python, Julia, C++ Healthcare Yes (Hybrid) Disease control in food
Modelling products
Shatyrko6 Numerical C++ Dynamic Systems No Stability of heat equation
solution
Taloub et al.7 Finite Difference Python, Julia Engineering Yes (Hybrid) Laplace equation solution
in 2D
Fitriah et al.8 Problem-Based Python Education No Learning outcomes
Learning improvement
Afrilianto et al.9 Creative Learning Python, C++ Education No Creative thinking ability in
Model mathematics
Purnomo et al.11 Problem-Based Python, MATLAB Education No Development of students’
Learning problem-solving skills
Kusmaryono Problem-Based Python, R Education No Improving mathematical
et al.12 Learning literacy
Swatthong & Integer Linear Python, C++ Cloud Computing Yes (Hybrid) Cloud orchestration model
Aswakul17 Programming optimisation
Ezzrhari et al.18 Multi-agent Systems Python, Java Distributed Systems No Multi-agent system
middleware design
Yang et al.19 Blockchain-Based Python, Java, Scala Agriculture Yes (Hybrid) Blockchain for machinery
Model scheduling
Liu et al.21 Fractional Differential C++, Julia AI Integration Yes (Hybrid) AI-enhanced modelling of
Equation differential equations
Harris et al.24 Array Programming Python Scientific Computing No Efficient array operations
with NumPy
Gadanidis Computational Literacy Python, MATLAB Education No Improving mathematical
et al.26 education through
computational methods
Zimmermann & Community Package Python Open-source Yes (Hybrid) Community maintenance
Falleri27 Maintenance Software and code reliability
Costanzo et al.28 Stochastic Method Python, Julia Signal Processing No Optimisation of PAPR
reduction methods
Yesylevskyy29 Molecular Simulation Rust Computational Yes (Hybrid) Memory-safe library for MD
Chemistry simulations
Zhu30 Incremental Learning Python, C++ Machine Learning Yes (Hybrid) Low-memory
implementations of ridge
solutions
Hatun31 Recursive Algorithms Python, C++ Electrical No Optimisation of Wiener
Engineering system identification
Kal’chuk et al.37 Poisson Integrals MATLAB, C++ Applied Yes (Hybrid) Approximation of harmonic
Mathematics integrals
Babak et al.47 Stochastic Method Python Air Traffic Control No Conflict probability
evaluation in air traffic

18 DOI: [Link] | Premier Journal of Science 2025;14:100173