Ubiquitous-cloud-inspired deterministic and stochastic service provider models with mixed-integer-programming

IAES International Journal of Artificial Intelligence (IJ-AI)
Vol. 13, No. 2, June 2024, pp. 1304~1311
ISSN: 2252-8938, DOI: 10.11591/ijai.v13.i2.pp1304-1311  1304
Journal homepage: http://ijai.iaescore.com
Ubiquitous-cloud-inspired deterministic and stochastic service
provider models with mixed-integer-programming
Sumarlin1
, Muhammad Zarlis2
, Suherman1
, Syahril Efendi1
1
Faculty of Computer Science and Information Technology, University Sumatera Utara, Medan, Indonesia
2
Department of Information Systems Management, School of Information Systems, Bina Nusantara University, Jakarta, Indonesia
Article Info ABSTRACT
Article history:
Received May 11, 2023
Revised Oct 19, 2023
Accepted Dec 31, 2023
The ubiquitous computing system is a paradigm shift from personal
computing to physical integration. This study focuses on the deterministic
and stochastic service provider model to provide sub-services to computing
nodes to minimize rejection values. This deterministic service provider
model aims to reduce the cost of sending data from one place to another by
considering the processing capacity at each node and the demand for each
sub-service. At the same time, stochastic service provider aims to optimize
service provision in a stochastic environment where parameters such as
demand and capacity may change randomly. The novelties of this research
are the deterministic and stochastic service provider models and algorithms
with mixed integer programming (MIP). The test results show that the
solution found meets all the constraints and the smallest objective function
value. Stochastic modeling minimizes denial of service problems during
wireless sensor network (WSN) distribution. The model resented is the
ability of wireless sensors to establish connections between distributed
computing nodes. Stochastic modeling minimizes denial of service problems
during WSN distribution.
Keywords:
Mixed interger programming
Model
Optimization
Ubiquitous clouds
This is an open access article under the CC BY-SA license.
Corresponding Author:
Syahril Efendi
Faculty of Computer Science and Information Technology, Universitas Sumatera Utara
St. Dr. Mansur No. 9 Kampus Padang Bulan, Medan 20155, Indonesia
Email: syahril1@usu.ac.id
1. INTRODUCTION
The term "ubiquitous internet of things (IoT)" describes how data is connected and exchanged
anytime, anywhere [1]. The cloud is becoming a prominent research area. It is ubiquitous because it serves
client needs regardless of time and place [2], such as the use of wireless sensor networks (WSN) and
ubiquitous technology for intelligent agriculture with energy efficiency [3]. However, in reality, the signal
from WSN is lagging, which creates uncomfortable for users. Therefore, this manuscript proposes an
optimization model to overcome the problem. The model could decide the optimal base transmission station
(BTS) position to minimize the disturbances.
The widespread adoption of cloud computing creates a new paradigm for work-from-home culture
enterprises in the unprecedented crisis of the COVID-19 outbreak [4]. The blockchain ubiquitous
manufacturing (BCUM) platform is presented to securely and reliably integrate manufacturing systems into
the cloud paradigm that can provide a homogeneous and decentralized environment [5]. The deterministic
model is a mathematical model that allows for reasonably accurate measurement of symptoms. Examples of
deterministic model research are deterministic model studies to predict flooding [6], analysis understanding
the dynamics of tuberculosis with a focus on first-line treatments such as vaccination and drug resistance [7],
deterministic models for the dynamics of Q Fever transmission in dairy cattle using sensitivity analysis and

Int J Artif Intell ISSN: 2252-8938 
Ubiquitous-cloud-inspired deterministic and stochastic service provider models with … (Sumarlin)
1305
optimal control Yosua [8], a deterministic model for country inventory policies for vaccine procurement [9],
and a deterministic model for composable reactive systems [10].
In a stochastic model, symptoms can be identified mathematically with variable levels of accuracy.
In the stochastic model, also called the probabilistic model, the probability of each event is calculated. There
have been several studies on the stochastic model, including one on a plug-in hybrid electric vehicle
stochastic model predictive control (MPC) technique based on reinforcement learning vehicle energy
management [11] and another on a stochastic model for virtual power plants' involvement in pool markets,
futures markets, and contracts with withdrawal penalties [12], a new non-geometrical stochastic model
(NGSM) for non-stationary wideband vehicle communication channels is presented [13], a discrete-time
stochastic epidemic model with binomial distributions to investigate the disease's transmission [14],
stochastic modeling of the cutting force in turning processes [15], and a stochastic model for forecasting the
novel coronavirus disease [16], and stochastic modeling reveals mechanisms of metabolic heterogeneity [17].
Mixed integer programming (MIP) is a type of mathematical optimization that involves optimizing
the objective function by considering variables that can take integer (integer) values or actual (continuous)
number values. In MIP, some variables may take integer values while others may take constant values [18].
There are several studies on MIP, namely Anderson et al. [19] proposing mixed solid integer programming
formulations for modeling and analysis of trained neural networks, Schrotenboer et al. [20] developing MIP
models for maintenance planning in wind farms offshore under conditions of uncertainty, and Belotti et al. [21]
discussed the use of non-convex mixed integer nonlinear programming (MINLP) tightening and branching
techniques. The simulation results show that the practical strategy increases the breaker's performance. The
book by Wolsey and Nemhauser [22] describes combinatorial and integer optimization and the techniques
used to solve these problems. This book provides a comprehensive guide on integer optimization models and
methods. The book by Bertsimas and Tsitsiklis [23] describes linear optimization, which provides a complete
introduction to linear optimization and optimization techniques. Tawarmalani and Sahinidis [24] discusses
branch-and-cut for mixed integer linear programming (MILP) based on polyhedra and providing
computational evidence of the effectiveness of this approach in global optimization problems,
Marcucci and Tedrake [25] develops a warm start method in mixed-integer programs for predictive control of
hybrid system models, Yu and Shen [26] develops a MIP approach distributionally robust with decision-
dependent uncertainty settings, Agrawal and Bansal [27] focuses on security in the use of cloud computing in
the context of education, and Krishnaraj [28] implementing a human activity monitoring system through the
use of IoT sensors and the Blynk cloud platform. IoT refers to a network of physical objects connected to the
internet and interacting with each other. Blynk is a cloud platform that enables developers and users to build
IoT-based applications quickly.
The primary reference of this research is Tomasgard et al. [29] that aims to develop mathematical
models and algorithms that can be used to analyze and predict the performance of distributed
telecommunication networks. In a distributed telecommunications network, data and network resources are
distributed across multiple geographic locations, such as distributed data centers or base stations on a cellular
network. This study aims to understand the complex interactions between various elements in a distributed
telecommunications network and how traffic density, network capacity, and connectivity level can affect
network performance. In this study, the researchers used a mathematical modeling approach to visualize and
understand the dynamics of distributed telecommunication networks. They may use techniques such as
queuing theory, graph theory, or probability models to describe the behavior of networks in different
situations. The results of this research can assist in planning and developing a more efficient distributed
telecommunication network. For example, the developed model can be used to predict the traffic load on the
web at a specific time or to identify overload points that can cause a decrease in service quality. This research
can also provide insights into improving the efficient use of network resources, such as optimal network
capacity or better traffic management.
The ubiquitous computing system is a paradigm shift from personal computing to physical
integration. Physical integration includes material considerations in serving clients accessing cloud services.
Access to cloud infrastructure at various points is branched. Network branching is caused by client access
points that are close together, thus causing shared access to network hardware such as switches and routers.
Access sharing can be managed using a static scheduling system that requires minimal software to apply to
the limited memory of switches and routers. Static scheduling algorithms require detailed information about
tasks, such as length, the quantity of jobs, and execution deadlines, as well as information about the resources
to be provisioned, such as available processing power, memory capacity, and energy consumption, before
(advance) execution. Due to the dynamic nature, inconsistencies in the cloud computing environment, and
the volume of requests and resources, static algorithms must be revised for this type of system, as they cannot
correctly adjust the workload distribution between resources.
On the other hand, dynamic scheduling, which requires more memory, is difficult to implement. To
improve static scheduling performance on devices with limited memory, this research integrates deterministic

 ISSN: 2252-8938
Int J Artif Intell Vol. 13, No. 2, June 2024: 1304-1311
1306
and stochastic models into static scheduling so that scheduling produces better performance. The problem in
creating a model is determining how to distribute subservience by counting nodes. This is done to satisfy
demand using the finest resources that service providers have available. A linear MIP model of this issue is
possible [30]. Sub-services are to be made available to compute nodes to decrease reject values.
2. METHOD
The research framework is a plan or structure used to organize and compile the research steps to be
carried out. This framework assists researchers in designing and conducting research with clear objectives
and a systematic methodology. The following research framework is shown in Figure 1.
Figure 1. Research work framework
The following is an explanation of the research stages according to Figure 1:
- Collection of study materials: this stage involves gathering information, references, literature, and other
resources relevant to the research topic. This study material is used to understand the issues studied and
develop a theoretical framework.
- Research data collection: at this stage, the researcher collects the data needed to answer research
questions or test the hypotheses that have been formulated. Data collection methods such as observation,
interviews, questionnaires, or document analysis can vary.
- Formulate the model function: after the data is collected, the next step is formulating the model function.
This means developing approaches or concepts that will be used to analyze or explain the data that has
been collected. Model functions can be mathematical formulas, theories, or conceptual frameworks that
describe the relationships between variables or constructs in research.
- Formulate constraints: this stage involves identifying and formulating obstacles that may arise in the
research process. These constraints can be in the form of limited data, resources, time, or methodological
problems. Defining conditions helps in devising an effective research strategy and foreseeing potential
obstacles that may be encountered.
- Model building: after formulating the function of the model and identifying the constraints, the next step
is building the model. This involves developing a structure or framework that systematically applies
model functions to research data. Modelling also consists in selecting a technique or method of analysis
appropriate to the research objective.
- Model testing and simulation: the model is tested and simulated using the relevant data at this stage.
Testing and simulation aim to validate whether the model can produce accurate results by research
objectives. The model can be revised or adjusted if the results do not match.
- Model validation: the model validation stage involves more comprehensive and in-depth testing to ensure
that the model created can produce consistent and reliable results. Validation involves statistical analysis,
comparison with previous research, and assessing the model's suitability with existing data.
- Proposed models: the final stage is to develop the proposed model based on research results and model
validation. The proposed model is the final result of the research and reflects the research contribution to
the field under study. This model can include policy recommendations, new frameworks, or solutions
based on findings.

1307
3. RESULTS AND DISCUSSION
The results of this study and novelties are the deterministic and stochastic service provider models
for providing sub-services to computing nodes to minimize rejection values. By using this concept, the need
for sub-service 𝑗 in a location or another place connected to the need for the subsurface itself will be reduced.
As a result, sub-service j received the bare minimum amount of requests for its processing capacity. The
following is the deterministic service provider model and algorithm, as well as testing the algorithm built
based on (1):
min ∑ 𝜋(𝑗)(𝑑(𝑗) − ∑ 𝑥(𝑖, 𝑗))
𝑖∈𝐼
𝑗∈𝐽 (1)
with constraints on (2)-(4):
∑ 𝑟(𝑖, 𝑗)𝑧(𝑖, 𝑗) + ∑ 𝑥(𝑖, 𝑗) ≤ 𝑠(𝑖), 𝑖 ∈ 𝐼
𝑗∈𝐽
𝑗∈𝐽 (2)
∑ 𝑥(𝑖, 𝑗) ≤ 𝑑(𝑗), 𝑗 ∈ 𝐽
𝑖∈𝐼 (3)
𝑀(𝑖, 𝑗)𝑧(𝑖, 𝑗) − 𝑥(𝑖, 𝑗) ≥ 0, (𝑖, 𝑗) ∈ 𝐼𝑥𝐽 (4)
where
𝐼 is computational notation set
𝐽 is subservice set
𝑑(𝑗) is demand for service
𝜋(𝑗) is get the value for sub-service j from another source
𝑟(𝑖, 𝑗) is source for sub-service j on node I defined
𝑠(𝑖) is total node I capacity (total capacity of node i)
(𝑖, 𝑗) is number of requests for processing power made by assembly node I and subsurface j
𝑧(𝑖, 𝑗) is programming connection between node I and sub-service j
𝑀(𝑖, 𝑗) is upper limit values for variablesd (j)−∑ 𝑥(𝑖, 𝑗)
𝑗∈𝐽 in the objective function is part of the
subservice request j. The upper bound and lower bound for the variable 𝑥(𝑖, 𝑗) are represented by 𝑀(𝑖, 𝑗)in
the constraint function. 𝑀(𝑖, 𝑗) is min{ 𝑠(𝑖)−𝑟(𝑖, 𝑗), 𝑑(𝑗)}.
This model is a linear optimization model to minimize the cost of sending data from one place to
another by considering the processing capacity at each node and the demand for each sub-service. The
variables used in this model are 𝑥(𝑖, 𝑗), the Handling of capacity requests number originating from junction
node I subsurface j, and 𝑧(𝑖, 𝑗), which in programming is the link between node I and sub-service j.
Constraints in this model include: i) capacity limits on each node, ii) requests for each sub-service, and iii)
upper limits for variables x(i, j).
The capacity constraints at each node are defined by s(i) and limit the total capacity available at
node i. The requests for each sub-service are defined by d(j) and π(j), limiting the quantity of capacity
requests received made by subsurface j from elsewhere. Another constraint on this model is M(i, j)z(i, j) - x(i,
j) ≥ 0, which ensures that the processing requests for capacity made by subsurface j and meeting node i do
not exceed the available capacity at node i. The maximum for the variables x(i, j) the lowest bound, are
defined by M(i, j), which is the minimum between s(i) - r(i, j) and d(j). This model is designed to determine
the optimal allocation of processing capacity at each node to meet sub-service demands efficiently. The
Algorithm can see:
1) Initialize variables and parameters
Set I as a set of computational notation
Set J as the subservices collection
Set d(j) as a demand for assistance for j
Set π(j) as the value of the subservice j request coming from another source
Set r(i, j) as the specified resource for node I and subservice j
Set s(i) as the power of node I overall
Set x(i, j) as the quantity of handling requests for capacity produced by assembly node I and subservice j
Set z(i, j) as the link programming between subservice j and node I
Set M(i, j) as the maximum value for the variable
2) Calculate the lessening M (i, j) using the formula: M(i, j) = min{ s(i) - r(i, j), d(j)}
3) Define the objective function: min ∑ π(j) (d(j) − ∑ x(i, j))
4) Define the constraint function:
∑ r(i, j) z(i, j) + ∑ x(i, j) ≤ s(i), i ∈ I, j ∈ J
∑ x(i, j) ≤ d(j), i ∈ I, j ∈ J
M(i, j)z(i, j) − x(i, j) ≥ 0, (i, j) ∈ IxJ

 ISSN: 2252-8938
1308
5) Use a linear programming solver to solve optimization problems with defined objective and constraint
functions.
6) If the optimal solution is found, return the objective value as the solution. If not, try again with different
variable values or increase the upper bound M(i, j) and repeat step 5.
The test results from the algorithm above are given to show that the optimal objective value is 0, and
all decision variables have a value of 0. This indicates that the solution meets all the constraints and the
smallest accurate function value. In general, the results of this test are good. The following is the stochastic
and algorithmic service provider model as well as testing of the algorithms built on (5) and (6):
min 𝑄(𝑍) ∑ 𝑟(𝑖, 𝑗)𝑧(𝑖, 𝑗) ≤ 𝑠(𝑖), 𝑖 ∈ 𝐼
𝑗∈𝐽 (5)
𝑧(𝑖, 𝑗) ∈ {0,1}, {𝑖, 𝑗} ∈ 𝐼𝑥𝐽 (6)
The maximum number of sub-services allowed by the gusset capacity is determined by the
constraint function in (7).
𝑄(𝑍) = 𝐸[𝑞(𝑧, 𝜋, 𝛿)] (7)
The expectations for the stochastic parameters 𝜋
̅ and 𝛿̅ on (8)-(11):
𝑞(𝑧, 𝜋, 𝛿) = min ∑ 𝜋(𝑗)[𝛿(𝑗) − ∑ 𝑥(𝑖, 𝑗)]
𝑖∈𝐼
𝑗∈𝐽 (8)
∑ 𝑥(𝑖, 𝑗) ≤ 𝑠(𝑖) − ∑ 𝑟(𝑖, 𝑗)𝑧(𝑖, 𝑗) = 𝑠 ∗ (𝑖), 𝑖 ∈ 𝐼
𝑖∈𝐼
𝑗∈𝐽 (9)
∑ 𝑥(𝑖, 𝑗) ≤ 𝛿(𝑗), 𝑗 ∈ 𝐽
𝑖∈𝐼 (10)
𝑥(𝑖, 𝑗) ≤ 𝑀(𝑖, 𝑗)𝑧(𝑖, 𝑗), (𝑖, 𝑗) ∈ 𝐼𝐽 𝑥(𝑖, 𝑗) ≥ 0, (𝑖, 𝑗) ∈ 𝐼 (11)
To optimize the model, the mathematical optimization algorithm is used to find the best solution for
the objective function 𝑄(𝑍) by considering all constraint functions. The results of testing the algorithm show
that this model can provide good results in optimizing service provision in a stochastic environment. The
Algorithm can see:
1) Initialize variables:
𝑧(𝑖, 𝑗): 0 or 1, indicates whether sub-service
j is connected to node i or not
𝑥(𝑖, 𝑗): integer, indicating the amount of processing capacity produced by sub-service j on node i
2) Calculate the expected value of the objective function:
𝑄(𝑍) = 𝐸[𝑞(𝑧, 𝜋, 𝛿)]
𝑞(𝑧, 𝜋, 𝛿) = min ∑ 𝜋(𝑗)[𝛿(𝑗) − ∑ 𝑥(𝑖, 𝑗)]
𝑖∈𝐼
𝑗∈𝐽
3) Calculate the expected value for each stochastic parameterk π dan δ
4) Determine the the maximum amount of possible sub-services provided:
∑ 𝑥(𝑖, 𝑗) ≤ 𝛿(𝑗), 𝑗 ∈ 𝐽
𝑖∈𝐼
5) Determine the constraints for the variable x:
∑ 𝑥(𝑖, 𝑗) ≤ 𝑠(𝑖) − ∑ 𝑟(𝑖, 𝑗)𝑧(𝑖, 𝑗) = 𝑠 ∗ (𝑖), 𝑖 ∈ 𝐼
𝑖∈𝐼
𝑗∈𝐽
∑ x(i, j) ≤ δ(j), j ∈ J, i∈I
x(i, j) ≤ M(i, j)z(i, j), (i, j) ∈ IxJ
x(i, j) ≥ 0, (i, j) ∈ IxJ
6) Solve the model to find the optimal solution:
- Minimize the 𝑄(𝑍) values with the constraints in steps 3 and 4.
- Determine the values of the variables z(i, j) and x(i, j) that produce the minimum 𝑄(𝑍)value.
7) Optimum solution output: print out the 𝑄(𝑍), 𝑧(𝑖, 𝑗), and 𝑥(𝑖, 𝑗)values that yield the minimum 𝑄(𝑍)
value.
The test results from the algorithm above are given to show that the optimal objective value is 0, and
all decision variables have a value of 0. However, in applying the algorithm to network traffic density, it can
be seen in Figure 2 which is the initial condition that has been simulated with increased traffic demand and
unlimited capacity. Figure 2 is an example of a planned network obtained by fixed rate model (FRM) for the
same example with two different examples of requirements on end-user traffic, d=600 Kb/s and d=3 Mb/s for
all MCs, and with capacities limited from the link that connects the mesh access point (MAP) to the wired

1309
network. So that by using the optimization of the algorithm, the traffic density results are formed as shown in
Figure 3. Figure 3 where the solution provided by the algorithm model is such that an increase in traffic
demand implies an increase in the dimensions used to convey traffic, both in wired networks and in wireless.
This indicates that the solution meets all the constraints and the smallest accurate function value. In general,
the results of this test are good.
(a) (b)
Figure 2. Network traffic density level: (a) d=600 Kb/s and (b) d=3 Mb/s
(a) (b)
Figure 3. Optimal network traffic density level: (a) d=600 Kb/s and (b) d=3 Mb/s
4. CONCLUSION
Researchers have successfully built a model and algorithm for deterministic and stochastic service
providers with MIP. The test results show a good value, with the optimal objective value being 0 and all
decision variables having a value of 0. This indicates that the solution meets all constraints and has the
smallest accurate function value. The distribution transparency model is essential for building a distributed
network investment and service provider model—operational resource allocation on service provider
processing nodes. The model presented is the ability of wireless sensors to establish connections between
distributed computing nodes. Stochastic modeling minimizes denial of service problems during wireless
sensor network distribution.
REFERENCES
[1] A. Jiang, H. Yuan, D. Li, and J. Tian, “Key technologies of ubiquitous power Internet of Things-aided smart grid,” Journal of
Renewable and Sustainable Energy, vol. 11, no. 6, pp. 1–20, 2019, doi: 10.1063/1.5121856.
[2] D. K. A, “Review on Task Scheduling in Ubiquitous Clouds,” Journal of ISMAC, vol. 1, no. 1, pp. 71–79, 2019, doi:
10.36548/jismac.2019.1.006.
Time Time
Speed
Speed

 ISSN: 2252-8938
1310
[3] R. Dhaya, R. Kanthavel, and A. Ahilan, “Developing an energy-efficient ubiquitous agriculture mobile sensor network-based
threshold built-in MAC routing protocol (TBMP),” Soft Computing, vol. 25, no. 18, pp. 12333–12342, 2021, doi:
10.1007/s00500-021-05927-7.
[4] S. Mandal, D. A. Khan, and S. Jain, “Cloud-Based Zero Trust Access Control Policy: An Approach to Support Work-From-Home Driven
by COVID-19 Pandemic,” New Generation Computing, vol. 39, no. 3–4, pp. 599–622, 2021, doi: 10.1007/s00354-021-00130-6.
[5] A. V. Barenji, Z. Li, W. M. Wang, G. Q. Huang, and D. A. Guerra-Zubiaga, “Blockchain-based ubiquitous manufacturing: a
secure and reliable cyber-physical system,” International Journal of Production Research, vol. 58, no. 7, pp. 2200–2221, 2020,
doi: 10.1080/00207543.2019.1680899.
[6] X. Jiang, H. V. Gupta, Z. Liang, and B. Li, “Toward Improved Probabilistic Predictions for Flood Forecasts Generated Using
Deterministic Models,” Water Resources Research, vol. 55, no. 11, pp. 9519–9543, 2019, doi: 10.1029/2019WR025477.
[7] D. Otoo, S. Osman, S. A. Poku, and E. K. Donkoh, “Dynamics of Tuberculosis (TB) with Drug Resistance to First-Line
Treatment and Leaky Vaccination: A Deterministic Modelling Perspective,” Computational and Mathematical Methods in
Medicine, vol. 2021, pp. 1–14, 2021, doi: 10.1155/2021/5593864.
[8] M. Mubarak, J. Berneburg, and C. Nowzari, “Stochastic vs. Deterministic Modeling for the Spread of COVID-19 in Small
Networks,” in 2021 American Control Conference (ACC), 2021, pp. 1346–1351, doi: 10.23919/ACC50511.2021.9482985.
[9] A. Pınarbaşı and B. Vizvári, “A deterministic model for the inventory policy of countries for procurement of vaccines,” Central
European Journal of Operations Research, vol. 31, no. 4, pp. 1029–1041, 2023, doi: 10.1007/s10100-022-00831-3.
[10] M. Lohstroh et al., “Reactors: A Deterministic Model for Composable Reactive Systems,” in Cyber Physical Systems. Model-
Based Design, Cham: Springer, 2020, pp. 59–85, doi: 10.1007/978-3-030-41131-2_4.
[11] C. Zhang, W. Cui, Y. Du, T. Li, and N. Cui, “Energy management of hybrid electric vehicles based on model predictive control and deep
reinforcement learning,” in 2022 41st Chinese Control Conference (CCC), 2022, pp. 5441–5446, doi: 10.23919/CCC55666.2022.9902409.
[12] A. R. Jordehi, “A stochastic model for participation of virtual power plants in futures markets, pool markets and contracts with
withdrawal penalty,” Journal of Energy Storage, vol. 50, pp. 1–35, 2022.
[13] Z. Huang, X. Zhang, and X. Cheng, “Non‐geometrical stochastic model for non‐stationary wideband vehicular communication
channels,” IET Communications, vol. 14, no. 1, pp. 54–62, 2020, doi: 10.1049/iet-com.2019.0532.
[14] S. He, S. Tang, and L. Rong, “A discrete stochastic model of the COVID-19 outbreak: Forecast and control,” Mathematical
Biosciences and Engineering, vol. 17, no. 4, pp. 2792–2804, 2020, doi: 10.3934/MBE.2020153.
[15] G. Fodor, H. T. Sykora, and D. Bachrathy, “Stochastic modeling of the cutting force in turning processes,” International Journal
of Advanced Manufacturing Technology, vol. 111, no. 1–2, pp. 213–226, 2020, doi: 10.1007/s00170-020-05877-8.
[16] N. Sene, “Analysis of the stochastic model for predicting the novel coronavirus disease,” Advances in Difference Equations, vol.
2020, no. 1, pp. 1–19, 2020, doi: 10.1186/s13662-020-03025-w.
[17] M. K. Tonn, P. Thomas, M. Barahona, and D. A. Oyarzún, “Stochastic modelling reveals mechanisms of metabolic
heterogeneity,” Communications Biology, vol. 2, no. 1, pp. 1–9, 2019, doi: 10.1038/s42003-019-0347-0.
[18] T. Kleinert, M. Labbé, I. Ljubić, and M. Schmidt, “A Survey on Mixed-Integer Programming Techniques in Bilevel
Optimization,” EURO Journal on Computational Optimization, vol. 9, pp. 1–21, 2021, doi: 10.1016/j.ejco.2021.100007.
[19] R. Anderson, J. Huchette, W. Ma, C. Tjandraatmadja, and J. P. Vielma, “Strong mixed-integer programming formulations for
trained neural networks,” Mathematical Programming, vol. 183, no. 1–2, pp. 3–39, Sep. 2020, doi: 10.1007/s10107-020-01474-5.
[20] A. H. Schrotenboer, E. Ursavas, and I. F. A. Vis, “Mixed Integer Programming models for planning maintenance at offshore wind farms under
uncertainty,” Transportation Research Part C: Emerging Technologies, vol. 112, pp. 180–202, Mar. 2020, doi: 10.1016/j.trc.2019.12.014.
[21] P. Belotti, J. Lee, L. Liberti, F. Margot, and A. Wächter, “Branching and bounds tighteningtechniques for non-convex MINLP,”
Optimization Methods and Software, vol. 24, no. 4–5, pp. 597–634, 2009, doi: 10.1080/10556780903087124.
[22] L. A. Wolsey and G. L. Nemhauser, Integer and combinatorial optimization. USA: John Wiley & Sons, 1999.
[23] D. Bertsimas and J. N. Tsitsiklis, Introduction to Linear Optimization. Belmont, MA: Athena scientific, 1997.
[24] M. Tawarmalani and N. V. Sahinidis, “A polyhedral branch-and-cut approach to global optimization,” Mathematical
Programming, vol. 103, no. 2, pp. 225–249, 2005, doi: 10.1007/s10107-005-0581-8.
[25] T. Marcucci and R. Tedrake, “Warm start of mixed-integer programs for model predictive control of hybrid systems,” IEEE
Transactions on Automatic Control, vol. 66, no. 6, pp. 2433–2448, 2021, doi: 10.1109/TAC.2020.3007688.
[26] X. Yu and S. Shen, “Multistage distributionally robust mixed-integer programming with decision-dependent moment-based
ambiguity sets,” Mathematical Programming, vol. 196, no. 1–2, pp. 1025–1064, 2022, doi: 10.1007/s10107-020-01580-4.
[27] R. Agrawal and S. Bansal, “Cloud Computing: Security with Educational Usage,” Journal of ISMAC, vol. 4, no. 3, pp. 165–173,
2022, doi: 10.36548/jismac.2022.3.003.
[28] N. Krishnaraj, “Implementation of a Human Activity Monitoring System through IoT Sensor and Blynk Cloud Platform,” Journal
of Information Technology and Digital World, vol. 4, no. 2, pp. 105–113, 2022, doi: 10.36548/jitdw.2022.2.005.
[29] A. Tomasgard, S. Dye, S. W. Wallace, J. A. Audestad, L. Stougie, and M. H. van der Vlerk, “Modelling in Distributed
Telecommunications Networks,” Tinbergen Institute, pp. 1–25, 1997.
[30] V. Nair et al., “Solving Mixed Integer Programs Using Neural Networks,” Optimization and Control, vol. 1, pp. 1–57, 2020.
BIOGRAPHIES OF AUTHORS
Sumarlin is a Universitas Sumatera Utara computer science Ph.D. student. He is
also a lecturer at the Indonesian Institute of Technology and Business. He completed his
bachelor's degree in 2005 at STMIK Multimedia Prima Medan, and master's degree at the
University of North Sumatra Medan in 2011. He can be contacted at email:
sumarlin@students.usu.ac.id.

1311
Muhammad Zarlis is a professor at the Faculty of Computer Science and
Information Technology at Universitas Sumatera Utara, Medan, Indonesia. He completed his
Bachelor of Physics Education at the Faculty of Mathematics, Universitas Sumatera Utara, his
Masters in Computer Science at the Faculty of Computer Science, Universitas Indonesia,
Jakarta, and his doctorate program in Computer Science at Universiti Sains Malaysia (USM),
Penang. His current research interests are in the fields of intelligent systems, computer
security, and computation. He is the author/co-author of more than 187 research publications.
He can be contacted at email: muhammad.zarlis@binus.edu.
Suherman works as a lecturer at Universitas Sumatera Utara since 2003. He
completed B.Eng. in Electrical Engineering Universitas Sumatera Utara in 2000, M.Sc. in
Computing from RMIT Australia in 2009 and Ph.D. in Electrical Engineering, DMU-UK in
2013. He followed some non-degree courses: mobile communication at Ilmenau, Germany
(2010) and bioinformatics in EBI-Cambridge, UK (2011). In 2022, he follows online M.Sc.
program in Financial Engineering, WorldQuant University, New Orleans, USA. He can be
contacted at email: suherman@usu.ac.id.
Syahril Efendi professor from Universitas Sumatera Utara, Medan Indonesia. He
worked as a lecturer at the University of Sumatera Utara, fields: mathematics computation,
optimization, and computer science. His main research interest is data science, big data, and
machine learning. He can be contacted at email: syaril1@usu.ac.id or syahnyata1@gmail.com.

Ubiquitous-cloud-inspired deterministic and stochastic service provider models with mixed-integer-programming

More Related Content

Similar to Ubiquitous-cloud-inspired deterministic and stochastic service provider models with mixed-integer-programming (20)

More from IAESIJAI (20)

Recently uploaded (20)

Ubiquitous-cloud-inspired deterministic and stochastic service provider models with mixed-integer-programming