A Fuzzy-Genetic Multi-Objective Optimization Method Applied to

Deployment of Routers in Agricultural Crop Areas

P. H. G. Coelho

, J. F. M. Amaral

, T. M. Carvalho

1,2

, R. A. Gomes

, I. S. Cardoso

and T. N. Souza

State Univ. of Rio de Janeiro, FEN/UERJ, R. S. Francisco Xavier, 524/Sala 5001E, Maracanã, RJ,20550-900, Brazil

Pontifical Catholic University of Rio de Janeiro, ELE, R. Marquês de São Vincente, 225, Gávea, RJ, 22453-900, Brazil

thaynans.souza@gmail.com

Keywords: Fuzzy-Genetic Systems, Multi-Objective Optimization, Precision Agriculture, AI Applications.

Abstract: High technology is increasingly applied to improving crop fields and coined an area as Precision Agriculture.

The main focus of this work is to increase production by performing data acquisition from an agricultural

crop area, monitoring sensor devices to measure temperature, humidity, etc. This allows the administrator of

the field to make good decisions related to the land management. This paper proposes a hybrid fuzzy-genetic

multi-objective intelligent method to place routers in an agricultural crop area so that to cover the sensor

monitoring devices spread over it. The method combines a genetic algorithm with a fuzzy aggregation

technique to evaluate multiples objectives, in order to determine an adequate location of the routers

considering the designer´s preferences. Case studies are presented and show the proposal results.

1 INTRODUCTION

Family farming is at the origin of important debates,

both academic and political, and can refer to a

diversity of subjects in the collective imagination.

Family farming represents a production model

structured around the family and which is integrated

into different types of markets, whether acquiring

inputs and new technologies, or selling the food and

raw materials they produce. It is not, therefore, a

social group that predominated in the past and

remained relatively distant from the market, averse to

technology and poorly integrated into society. The

family farmer in Brazil typically owns a portion of

land ranging from 5 to 110 hectares (one hectare is

equivalent to 10 thousand m²). At least half of the

labor used in the activity of the property must be

family-owned, and income must come predominantly

of rural activity. Management must be strictly family-

based. According to IBGE (Brazilian Institute of

Geography and Statistics), responsible for the census,

family farming occupies 23% of the total cultivated

area in the country.

The use of technology is already part of several

segments in our society, and agriculture would be no

different. To seek greater revenue and productivity in

the field, the Brazilian farmers need to be increasingly

connected with what happens inside and outside their

property. Precision agriculture (PA) is largely

responsible for bringing agriculture into everyday life

in the field. Until a few years ago, they were restricted

to certain places and properties, because they were

impacted due to the high investment and financial

availability of the owners. How can we imagine this

in the current scenario, in which the demand for food

and its quality grow as the world population

increases? It is impossible not to use technology. The

term Agriculture 4.0 refers to the industry’s

upcoming significant trends, such as the internet of

things (IoT), using big data and machine learning to

make businesses more efficient amid the challenges

of population growth and climate change to improve

output, efficiency, and support sustainable agriculture

by using accurate information to make strategic

decisions (Tjhin et al., 2022).

By 2050, it is estimated that there will be 9 billion

people on the planet, which will require more

production, smaller costs and preservation of natural

resources. It is planned that atypical occurrences and

climate change represent serious risks to agricultural

production. As a result, an increase of 70% or more is

expected in food production. The agricultural

industry is changing as a result of the adoption of

emerging technologies. Using cutting-edge

technology like IoT, AI and other sensors, smart

792

Coelho, P., Amaral, J., Carvalho, T., Gomes, R., Cardoso, I. and Souza, T.

A Fuzzy-Genetic Multi-Objective Optimization Method Applied to Deployment of Routers in Agricultural Crop Areas.

DOI: 10.5220/0012696000003690

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

In Proceedings of the 26th International Conference on Enterprise Information Systems (ICEIS 2024) - Volume 1, pages 792-799

ISBN: 978-989-758-692-7; ISSN: 2184-4992

agriculture will transform traditional farming

methods production and international agricultural

policies. Agriculture 4.0 looks at its possible benefits

and drawbacks of the implementation methodologies,

compatibility, reliability, and investigates the several

digital tools that are being utilized to change the

agriculture industry and how to mitigate the

challenges (Gyamfi et al., 2024).

Summing up, precision agriculture combines

artificial intelligence, the internet of things and GPS,

making the farmer's life easier to achieve greater

productivity and efficiency in the field, saving inputs,

optimizing the soil and practical operation. Precision

agriculture uses an agricultural management system

that considers the particularities of each point on the

property. Its application was intensified with the

improvement of GPS, which made it possible to

install receivers in seeders, harvesters and sprayers,

associating productivity data with geographic

coordinates using satellites. On the other hand, digital

agriculture or agriculture 4.0 is a set of technologies

that help the producer to monitor rural activities more

closely, such as software and devices responsible for

collecting and processing data about the farm.

Research in the area of computational intelligence

applied to farming is intense currently and there are a

large number of papers reporting an increase in crop

productivity. Recent works include (Ketheneni et al.,

2023) which uses an ensemble model, with four

classifiers, to recommending the fertilizer and a

sequential convolution neural network to

recommending the pesticide for the appropriate crop.

In the recent past, 3D machine vision techniques

have been widely employed in agriculture and food

systems, leveraging advanced deep learning

technologies. Following this path (Xiang et al., 2023)

published a survey of 3D vision techniques in food

and agriculture applications. (Karunathilake et al.,

2023) review recent innovations, challenges and

future prospects of precision agriculture and smart

farming.

For a sustainable production environment

(Mallinger et al., 2023) discusses the impact of AI in

farming concerning the fusion of AI and autonomous

farming machinery (e.g., drones and field robots) in

the daily work experience of farmers.

The purpose of this paper is to position a set of

routers so to cover the sensor monitoring devices

spread in an agricultural crop area, sending data such

as temperature, soil humidity, and the like, giving the

family farmer the chance of optimal decisions

regarding his/her crop. It is intuitive that random

positioning of router nodes is not a good choice as can

result in poor communication performance with the

sensor monitoring devices. Besides, the actual

deployment may have restrictions and geographic

characteristics of the area in question, making it

essential to seek for different topologies to distribute

them. Such a problem is of the NP hard type, so that is

a good indication for looking for the optimal solution

following an approach using evolutionary techniques

that includes genetic algorithms with fuzzy

aggregation. The placement of routers in a network is

not a trivial problem. Usually, such a problem can be

solved using traditional evolutionary techniques such

as weighted-sum approach genetic algorithms or

Pareto-based techniques. Weighted-sum evaluation for

genetic algorithms leads to difficult assignment of

appropriate weights, while Pareto techniques require

the designer to select the most suitable solution among

the set of presented solutions.

Research using intelligent computational systems

have been taken continuous attention in universities

and research centers around the globe. (Waqas et al.,

2022) presents an optimal sensor node placement

problem in Structural Health Monitoring (SHM)

application based on an optimized Wireless sensor

network (WSN). The sensor node placement problem

is formulated in multi-objective form by considering

the energy consumption, sensitivity area and network

lifetime. A hybrid optimization algorithm was used

by a combination of Chaotic Particle Swarm

Optimization (CPSO) with Gravitational Search

Algorithm (GSA) to provide optimal sensor node

placement in WSN based SHM system. The optimal

solution is achieved in the Pareto environment case,

which makes the algorithm multi-objective.

In (Akram et al., 2023), a meta-heuristic multi-

objective firefly algorithm (MOFA) is presented to

solve the layout optimization problem. Their main

goal is to cover a number of objectives related to

optimal layouts of homogeneous WSNs, which

includes coverage, connectivity, lifetime, energy

consumption and the number of sensor nodes.

Router Nodes Placement Using Artificial Immune

Systems is used in (Coelho et al., 2017), and (Coelho

et al., 2015) for industrial applications.

In this paper, we use genetic algorithms to

determine the location of routers in a mesh network

through evolutionary techniques associated with a

new Takagi-Sugeno-Kang (TSK) fuzzy aggregation

method inspired by Mamdani system aggregation

presented in (Coelho et al., 2019) and (Coelho et al.,

2023).

This paper is organized into four sections. The

second section deals with the new hybrid fuzzy-

genetic proposed model followed by discussion of the

case studies in section three. Finally, section four

closes the article with conclusions.

A Fuzzy-Genetic Multi-Objective Optimization Method Applied to Deployment of Routers in Agricultural Crop Areas

793

2 HYBRID FUZZY- GENETIC

PROPOSED MODEL

In this article, we utilized a fuzzy-based system for

optimizing router positioning. Fuzzy systems provide

a more intuitive and flexible approach to handling

imprecise data, which can be easily modeled by the

user. Through the implementation of a hybrid fuzzy-

genetic strategy, we facilitate the development and

adjustment of the objective function, while preserving

the optimization strategy of the Genetic Algorithm.

The method to optimize the placement of the

routers in the crop area must consider the multiple

objective aspects of this issue. For instance, besides

the necessity to full coverage of the sensor monitoring

devices in the area, it may impose restrictions on

positioning the routers due to high cost involved or,

in the limit, preventing prohibitive installation costs.

A hybrid fuzzy-genetic approach was used based on

Genetic Algorithm (GA) and Fuzzy Inference System

(FIS).

Genetic algorithms are inspired by biological

evolution and some aspects of genetics. Optimization

problems are the main application of GAs,

particularly in problems with complex or large search

areas. In a searching problem, the idea is to establish

an analogy between the evolution of the species and

the problem. A set of possible solutions (population

of individuals) are evaluated, and each individual is

associated with a fitness value (quality of the

individual). The population is subjected to a process

of simulated evolution, through genetic operators

(selection and reproduction) for many generations. At

the end of the evolutionary simulated process, the best

individual (higher fitness value) is associated with the

solution.

The fitness evaluation function, also called

objective function, is conceived based on problem

specification and is very important for obtaining a

good result. The traditional GA involves a single

criterion, but many real-world optimization problems

involve multiples objectives. To deal with multiple

objectives, one possible approach is to convert vector

quantities (objectives) into a scalar (only one fitness

value considering all objectives) by using techniques

like the weighted-sum approach or Mamdani fuzzy

aggregation.

The utilization of a Fuzzy Inference System (FIS)

to aggregate the objectives allows the evaluation of

all objectives, integrating designer´s preferences in

relation to each objective and for each particular

problem. The proposed hybrid method offers an

advantage over Pareto optimization techniques

because it does not require that the designer chooses

the best solution at the end of the optimization

process. Specifications and preferences are

established a-priori and entered before evolution in a

simpler way through the fuzzy system, to guide the

evolution process in the direction of pre-established

preferences.

It is worth mentioning that each GA individual

represents a possible solution to the search problem.

During the evaluation process each GA individual is

submitted to a particular objective function that

represents one aspect of the problem and the results

are used as inputs to the fuzzy inference system. The

fuzzy aggregation method is applied to each

individual in the population in order to produce a

single scalar overall fitness value. Figure 1 illustrates

the evaluation model using a Takagi-Sugeno-Kang

(TSK) Fuzzy Aggregation Method.

Figure 1: TSK Fuzzy System applied to fitness evaluation.

The values for GA parameters (selection,

crossover, mutation, population size and maximum

number of generations) must be defined by the

designer. The rules for the fuzzy aggregator system

are conceived according to the designer´s preferences

to solve the problem by considering each objective.

To stop the evolution, a certain stopping criterion also

must be specified. The most common ones are related

to the maximum number of generations and a certain

fitness value to be reached.

The fuzzy aggregation method uses an ordinary

FIS in which each input corresponds to a particular

objective and the membership functions are triangular

or trapezoidal in shape for simplicity. The Multiple-

objective GA with the TSK Fuzzy aggregator system

used in this work is presented in Figure 2.

3 CASE STUDIES

In this study, we explored three scenarios as case

studies, aiming to strategically position low-power,

battery-operated routers in a mesh network for data

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

794

acquisition in an agricultural environment, with

different size and number of sensors. Throughout the

investigations, we leveraged the MATLAB Fuzzy

Toolbox. For all experiments, the main objective is to

cover all sensors using the routers, while avoiding the

positioning in forbidden areas, which might represent

places such as rivers, houses, and so on.

Figure 2: Optimization flowchart with TSK Fuzzy

Aggregator.

Figure 3: Membership functions for Cost variable.

To investigate the effectiveness of the TSK Fuzzy

Aggregator, we compare the results against those

obtained using the Mamdani Fuzzy Aggregator

(Coelho et al., 2023) across all of the used

scenarios in this article. To facilitate a meaningful

comparison of results, we configured both the TSK

(Takagi-Sugeno-Kang) fuzzy aggregator and the

Mamdani fuzzy aggregator to exhibit similar

behaviors. This involved maintaining consistency in

the information related to the input, specifically the

linguistic variables and their associated fuzzy sets.

The fuzzy sets for the number of points covered

by routers (SenCovered) are depicted in Figure 3,

while Figure 4 illustrates the fuzzy sets for the cost

variable (Cost).

Figure 4: Membership functions for SenCovered variable.

Figure 5: Membership functions for evaluation variable.

Despite the different specifications for the TSK

and Mamdani output, both methods employ a

common evaluation scale ranging from 0 to 10.

Figure 5 displays the fuzzy sets associated with the

Mamdani fuzzy aggregator, while the TSK fuzzy

aggregator output is described in terms of linear

equations for the linguistic variables:

• 𝑜𝑣𝑒𝑟𝑎𝑙𝑙𝑏𝑎𝑑  0.006 𝑆𝑒𝑛𝐶𝑜𝑣𝑒𝑑  0.4 𝐶𝑜𝑠𝑡  4

• 𝑎𝑡𝑡𝑒𝑛𝑑𝑏𝑎𝑑  4.5

• 𝑐𝑜𝑠𝑡𝑏𝑎𝑑  0.006 𝑆𝑒𝑛𝐶𝑜𝑣𝑒𝑟𝑒𝑑  0.4 𝐶𝑜𝑠𝑡  7

•

𝑔𝑜𝑜𝑑  0.04 𝑆𝑒𝑛𝐶𝑜𝑣𝑒𝑟𝑒𝑑  0.8 𝐶𝑜𝑠𝑡  2

Throughout all case studies, we maintained the

rules of the FIS, considering the problem as

essentially the same across scenarios. Figure 6

displays the rules for the TSK system, while Figure 7

illustrates the rules for the Mamdani fuzzy

aggregator. It is noteworthy that the rules used in the

Mamdani system can be simplified to 5 rules.

Figure 6: Rules for TSK Fuzzy Aggregator.

Figure 7: Rules for Mamdani Fuzzy Aggregator.

A Fuzzy-Genetic Multi-Objective Optimization Method Applied to Deployment of Routers in Agricultural Crop Areas

795

Figure 8: Surface for TSK Fuzzy Aggregator.

The TSK system was designed to progressively

reach toward an optimal solution, as illustrated in

Figure 8. This approach may be considered an

advantageous feature of the TSK system, offering

simplicity and precision in its rule-based outputs.

For each case study, we ran the optimization

strategy ten times, and compared the results in terms

of convergence (basically looking at the output

score). As the objective and evaluation for both

strategies were tailored to have similar behavior, we

compare how good the fuzzy aggregation might be

into finding the optimal solution. Also, we aim to

observe the consistency of the results provided by the

optimization strategies.

3.1 First Case Study

In the initial case study, the scenario unfolds in an

agricultural expanse covering 2500 m² (50 m x 50 m),

necessitating a strategic spatial layout of routers.

Monitoring devices (sensors), with a reach of 13 m,

are strategically positioned to achieve two primary

objectives: ensuring each monitoring point is covered

by at least one router and minimizing installation

costs by avoiding high-cost areas. The parameters

employed in this experiment using genetic algorithm

are outlined in Table 1.

Table 1: GA Parameters for the Case Study 1.

Parameter Value

Generations 300

Population 50

Mutation 0.01

Crossove

0.6

The results of the first case study are illustrated in

Figure 9, showcasing the optimization outcomes

employing TSK and Mamdani as fuzzy aggregation

strategies. Notably, TSK and Mamdani exhibited

similar behaviors in terms of consistency. Both TSK

and Mamdani strategies converged to solutions with

similar fitness values, which showed that there might

be no difference in terms of robustness to achieve an

optimal solution. However, it is important to note that

the first case study can be considerably easier to

obtain a solution that fulfill the requirements.

Figure 9: Fitness curve for Case Study 1.

Figure 10: Routers’ deployment using TSK Fuzzy

Aggregator in Case Study 1.

Figure 10 provides a visual representation of the

monitored points and router positions for the best

individual identified by the GA. In this depiction,

smaller blue circles denote monitoring points, large

blue circles represent the coverage area of each

router, routers are marked by a red "x," the obstacles

are illustrated as red rectangles, and the crop limit

area is highlighted in green.

A visual inspection affirms the optimal solution

achieved through the optimization strategy. All

routers were strategically positioned to enable sensor

reach, avoiding forbidden areas, albeit with some

proximity to them.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

796

3.2 Second Case Study

For the second case study, we created one more

obstacle in the router position, which simulates the

obstacle of a house and a river crossing the

environment. In this experiment, we kept the

experimental setup, except for the number of

individuals, as presented in Table 2. The rationale

behind this change is based on the complexity of the

problem.

Table 2: GA Parameters for the Case Study 2.

Parameter Value

Generations 300

Population 150

Mutation 0.01

Crossove

0.6

The results of the optimization strategies can be

seen in Figure 11. We observed that the Mamdani

fuzzy aggregator obtained more unstable results and,

in average, noticeable worse results than the TSK

strategy. In a more complex scenario (if compared to

the previous case study), the TSK fuzzy aggregator

obtained results closer to the optimal, which may

indicate a more appropriate approach to achieve the

desired router placement configuration. We also

observed that the Mamdani-type fuzzy aggregator did

not converge gradually, but rather took some jumps

between the fitness values. For both observations in

this case study, the simplified building of the

Mamdani fuzzy system may led to worse

optimization results, which requires a much larger

number of estimates than for the TSK fuzzy

aggregator to achieve the best solutions in the

optimization process.

Figure 11: Fitness curve for Case Study 2.

Figure 12: Router’s deployment using TSK Fuzzy

Aggregator in Case Study 2.

To illustrate the best solutions found in the second

case study, Figure 12 shows one of the results using

the TSK fuzzy aggregator. As observed in the router

deployment, one of the difficulties is that some

routers probably need to be placed near one obstacle

to accomplish completely one of the objectives.

Nevertheless, it was possible to achieve a solution

that fulfilled all requirements, showing the

effectiveness of finding a set of routers’ deployment

for this case study.

3.3 Third Case Study

For the third case study, we increased the complexity

of the scenario by expanding: the number of sensors

(20), number of obstacles (3) and covered area

10000m² (100 m x 100 m). Due to that constraints, we

also increased the number of routers to be placed in

the environment from 5 to 9.

For the GA parameters, we also increased the

number of individuals of population, to balance the

increase in the search space. The parameters used in

this experiment can be seen in Table 3.

Table 3: GA Parameters for the Case Study 3.

Parameter Value

Generations 300

Population 300

Mutation 0.01

Crossove

0.6

Figure 13 displays the fitness curve for the third

case study. Some of the aspects of the graphs are also

present in previous experiments, specifically the

predominance of TSK method over the Mamdani.

However, we noticed that the Mamdani fuzzy

A Fuzzy-Genetic Multi-Objective Optimization Method Applied to Deployment of Routers in Agricultural Crop Areas

797

aggregator did not obtain reasonable results, with

small increase of fitness score over the epochs.

As depicted in Figure 14, it was possible to obtain

a solution that completely achieves all requirements

for the router deployment. For a larger area and

random positioning of sensors, it is plausible to

observe that all routers were allocated to be in range

to at least one sensor.

Figure 13: Fitness curve for Case Study 3.

Figure 14: Routers’ deployment using TSK Fuzzy

Aggregator in Case Study 3.

4 CONCLUSIONS

This paper focused on the placement of routers in a

small crop area for data acquisition of sensor

monitoring devices to optimize the production of

family agriculture. Case studies were presented with

scenarios with restrictions closer to real applications.

These scenarios consider not only the need to cover

all sensors but also the avoidance of areas where

routers’ installation cost is high. A hybrid TSK fuzzy-

genetic multiple-objective optimization method was

applied to place the routers in a crop area taking into

account the number of covered sensors and the cost

objectives in the routing problem.

The multiple-objective technique based on fuzzy

aggregation allows the evaluation of the objectives

simultaneously, including designer´s preferences

before the evolution takes place. This a-priori method

offers an interesting advantage over Pareto method

because it does not require that the designer chooses

the solution at the end of the process. It´s worth noting

that both the fuzzy aggregation specifications and

designer´s preferences are integrated before evolution

in a simpler manner, so the evolution is guided to the

desired preferences.

The TSK fuzzy aggregation presented here

showed, in two case studies, better performance to

this kind of application in comparison with Mamdani

fuzzy aggregation (Coelho et al., 2023). It is up to the

designer to choose between these two fuzzy

inferences which one should be applied to a particular

problem. However, it should be stressed that the

Mamdani fuzzy aggregation may require a higher

effort to achieve a similar behavior, mostly adding

new fuzzy sets for inputs, outputs and new rules to

increase the granularity of evaluation.

We plan for future works on precision agriculture

run case studies with some other a-priori methods, for

instance, such as a weighted-sum, and also a-

posteriori based on Pareto traditional method for

comparison’s sake. We also plan to include new

objectives for fuzzy aggregation and conceive a

model that can also select the adequate minimum

number of routers for complete field coverage. In

addition, comparisons with other techniques such as

African Vulture Optimization Algorithm (AVOA)

(Abdollahzadeh et al., 2021), Bat Algorithm (BA)

(Yang et al., 2013), Whale Optimization Algorithm

(WOA) (Mirjalili et al., 2016) and Particle Swarm

Optimization (PSO) (Marinakis et al., 2008) are

foreseen in future works related to precision

agriculture.

ACKNOWLEDGEMENTS

This study was financed in part by the Coordenação

de Aperfeiçoamento de Pessoal de Nível Superior –

Brasil (CAPES) – Finance Code 001, and FAPERJ.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

798

REFERENCES

Tjhin, V. U. and Riantini, R. E., 2022. Smart farming:

Implementation of industry 4.0 in the agricultural

sector, in Proceedings of the 6th International

Conference on E-Commerce, E-Business and E-

Government, pp. 416–421.

Gyamfi, E. K., ElSayed, Z., Kropczynski, J., Yakubu, M.

A., and Elsayed, N., 2024. Agricultural 4.0 Leveraging

on Technological Solutions: Study for Smart Farming

Sector. arXiv preprint arXiv:2401.00814.

Ketheneni, K., Yenuga, P., Garnepudi, P., Paleti, L., Burla,

N. R., Srinivas, O., and Yamarthi, N. R., 2024. Crop,

Fertilizer and Pesticide Recommendation using

Ensemble Method and Sequential Convolutional

Neural Network. International Journal of Intelligent

Systems and Applications in Engineering, 12(2), 473-

485.

Xiang, L., and Wang, D., 2023. A review of three-

dimensional vision techniques in food and agriculture

applications, Smart Agricultural Technology 5 (2023).

Karunathilake, E. M. B. M., Le, A. T., Heo, S., Chung, Y.

S., & Mansoor, S.,2023. The path to smart farming:

Innovations and opportunities in precision

agriculture. Agriculture, 13(8), 1593.

Mallinger, K., & Baeza-Yates, R. ,2024. Responsible AI in

Farming: A Multi-Criteria Framework for Sustainable

Technology Design. Applied Sciences, 14(1), 437.

Waqas, M., and Qureshi, S. S.,2022. Optimized wireless

Sensor Nodes Placement with multi-Objective Hybrid

Optimization Algorithm, International Journal of

Computational Intelligence in Control, Vol.14 No. 2.

Akram, A., Zafar, K., Mian, A. N., Baig, A. R., Almakki,

R., AlSuwaidan, L., & Khan, S.,2023. On Layout

Optimization of Wireless Sensor Network Using Meta-

Heuristic Approach. Computer Systems Science &

Engineering, 46(3).

Coelho, P. H. G., do Amaral, J. L. M., do Amaral, J. F. M.,

de Arruda Barreira, L F., and Barros, A. V., 2017.

Router nodes placement using artificial immune

systems for wireless sensor industrial networks, in

Lecture Notes in Business Information Processing, vol.

291, Springer International Publishing, pp. 155-172.

Coelho, P. H. G., do Amaral, J. L. M., do Amaral, J. F. M.,

de Arruda Barreira, L F., and Barros, A. V., 2015.

Applying artificial immune systems for deploying

router nodes in wireless networks in the presence of

obstacles, in Lecture Notes in Business Information

Processing, vol. 227, Springer International Publishing,

pp. 167-183.

Coelho, P., do Amaral, J. M., Guimarães, K., and Bentes,

M., 2019. Layout of routers in mesh networks with

evolutionary techniques, in Proceedings of the 21st

International Conference on Enterprise Information

Systems – vol. 1, pp. 438-445.

Coelho, P.H.G., Amaral, J.F.M., Guimarães, K.P., Rocha,

E.N., and Souza, T.S., 2023. Smart Placement of

Routers in Agricultural Crop Areas in Proceedings of

the 25st International Conference on Enterprise

Information Systems – vol. 1, pp. 99-106.

Abdollahzadeh, B., Gharehchopogh, F.S. and Mirjalili, S.,

2021. African vultures optimization algorithm: A new

nature-inspired metaheuristic algorithm for global

optimization problems. Computers & Industrial

Engineering, 158, p.107408.

Yang, X.S. and He, X., 2013. Bat algorithm: literature

review and applications. International Journal of Bio-

inspired computation, 5(3), pp.141-149.

Mirjalili, S. and Lewis, A., 2016. The whale optimization

algorithm. Advances in engineering software, 95,

pp.51-67.

Marinakis, Y. and Marinaki, M., 2008. A particle swarm

optimization algorithm with path relinking for the

location routing problem. Journal of Mathematical

Modelling and Algorithms, 7, pp.59-78.

A Fuzzy-Genetic Multi-Objective Optimization Method Applied to Deployment of Routers in Agricultural Crop Areas

799