International Journal of Electrical and Computer Engineering (IJECE) Vol.
No.
October 2025, pp.
ISSN: 2088-8708.
DOI: 10.
11591/ijece.
Development of a fuzzy logic-based greenhouse system for optimizing bio-fertigation Achouak Touhami1,2,3.
Amina Bourouis3,4.
Amel Mahammedi4.
Sana Mechraoui4.
Sana Touhami3,4 Department of Mathematics and Computer Science.
Ali Kafi University Center.
Tindouf.
Algeria Laboratory of Environmental and Energy Systems (LSEE).
Ali Kafi University Center.
Tindouf.
Algeria Laboratory of Innovations in Informatics and Engineering (INIE).
Tahri Mohamed University.
Bechar.
Algeria Department of Mathematics and Computer Science.
Tahri Mohamed University.
Bechar.
Algeria
Article Info
ABSTRACT
Article history:
Modern agriculture faces growing challenges in meeting food and resource demands, particularly with increasing pressure on water and fertilizer usage.
This study proposes a fuzzy logic-based algorithm to optimize biofertigation by managing key greenhouse parameters: temperature, humidity, soil pH, and soil moisture.
Implemented in MATLAB, the system automates the control of actuators .
an, heater, irrigation, fertilization and fertigation pump.
based on sensor data and fuzzy rules.
Results show a 27.
reduction in water use, 58.
82% decrease in fertilizer consumption, and a 5% increase in tomato yield.
Additionally, statistical error metrics mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), and mean absolute percentage error (MAPE) were reduced to zero, confirming the systemAos high precision and effectiveness in promoting sustainable agricultural practices.
Received Mar 2, 2025 Revised Jul 4, 2025 Accepted Jul 12, 2025 Keywords:
Bio-fertigation optimization Fuzzy logic Greenhouse automation Sensor-based control Water and fertilizer efficiency This is an open access article under the CC BY-SA license.
Corresponding Author:
Achouak Touhami Department of Mathematics and Computer Science.
Ali Kafi University Center 3700 Hai el Moustakbel.
Tindouf.
Algeria Email: achouak.
touhami@cuniv-tindouf.
INTRODUCTION
Rising global population, industrialization, and climate change are reducing arable land .
and increasing food production demands.
The Food and Agriculture Organization of the United Nations (FAO) estimates significant increases in cropland and water will be needed by 2050 .
, .
Challenges such as labor shortages and water scarcity make traditional greenhouse methods insufficient .
Greenhouse farmingAioriginating in the 19th centuryAinow offers a sustainable solution by enabling year-round cultivation through controlled environments .
, .
However, conventional greenhouses rely heavily on manual labor .
Smart greenhouse technologies can improve efficiency by automating the monitoring and control of growing conditions .
Tomatoes, a nutritious and widely consumed crop, require careful water and nutrient management .
, .
Overuse of water and fertilizers has caused environmental harm, emphasizing the need for sustainable regulation of inputs in controlled agriculture.
Organic fertilizers provide a balanced and gradual release of nutrients, enhancing soil fertility and microbial health while reducing reliance on synthetic fertilizers.
However, improper use can lead to nutrient imbalances .
Controlled-release formulations and compost further improve soil properties .
Drip irrigation conserves water and improves crop quality by delivering precise amounts at the right time .
FertigationAiapplying fertilizers through irrigationAiboosts nutrient efficiency, reduces labor, and can lower fertilizer use by 1525% without affecting yield, especially when tailored to crop growth stages through drip systems .
Journal homepage: http://ijece.
ISSN: 2088-8708
Diverse studies and research have been carried out in the area of monitoring internal climatic parameters and plant fertigation in greenhouses.
Deepak et al.
investigates the application of fertigationAia combined irrigation and fertilization techniqueAiin open-field agriculture with a focus on aquaponic systems.
Amudha et al.
focus on optimizing fertilizer application in agriculture using bioinspired algorithms, specifically the fruit fly optimization (FFO) algorithm and social spider algorithm (SSA).
These algorithms, inspired by biological species, aim to balance chemical fertilizer use with manure to enhance soil fertility, conserve resources, and minimize environmental impacts.
Dahlila et al.
propose an IoT-based smart fertigation management system designed for agricultural areas prone to power outages.
The system integrates sensors for real-time monitoring of irrigation and power outages, with features such as automated irrigation scheduling, pesticide management, and polybag cleaning.
Dwiratna et al.
introduce a modified hydroponic kit featuring a self-fertigation system, specifically designed for remote areas with unreliable electricity supply.
Bao et al.
investigate the impact of intelligent drip fertigation (IF) on watermelon production in a greenhouse environment over three growing seasons .
9Ae2.
IF integrates real-time soil moisture sensors with IoT-based automated irrigation and fertilization to optimize water and nutrient application.
Imbernyn-mulero et al.
evaluate an advanced autonomous fertigation system designed to optimize the use of variable-quality irrigation water.
Bonelli et al.
evaluated the performance of timer-based (TB) versus smart sensor-based (SB) irrigation strategies.
Wang et al.
conducted a field experiment in Shouguang.
Shandong Province, to assess the effects of various irrigation scheduling treatmentsAifarmer drip irrigation (FI), intelligent irrigation (II.
, and intelligent irrigation (II.
Aion tomato growth, irrigation water usage, and nutrient efficiency across two growing seasons.
Idris et al.
developed an internet of things (IoT)-based fertigation system to automatically deliver a fertilizer mixture with a consistent electrical conductivity (EC) value to plants.
Vojnovic et al.
aimed to explore new methods for fertigation and grafting to optimize cucumber yield and quality in the greenhouse.
Traditional greenhouse control systems struggle with complexity and uncertainty, often leading to inefficient resource use and poor plant growth.
Fuzzy logic provides a more adaptive and intelligent approach by mimicking human reasoning and handling nonlinear, uncertain conditions through expert-defined rules.
This results in more efficient, scalable, and sustainable greenhouse management, ultimately improving crop This study presents a fuzzy logic-based algorithm for optimizing bio-fertigation in a tomato greenhouse in southern Algeria.
Unlike traditional systems that manage only one or two environmental parameters, our approach integrates four key variables .
emperature, humidity, soil pH, and soil moistur.
and controls five actuators simultaneously .
an, heater, irrigation pump, fertilization pump, and fertigation The novelty lies in the combination of real-time sensor data, expert-derived fuzzy rules, and the use of biofertilizers instead of chemical fertilizersAian environmentally sustainable practice.
This multi-variable control system provides an adaptive, intelligent, and energy-efficient solution for optimizing tomato production in arid regions like southern Algeria.
The article details the algorithm design, membership functions, fuzzy rule base, experimental validation, and results, concluding with future research directions.
METHOD
In this section, we present the proposed algorithms developed to optimize the greenhouse microclimate using fuzzy logic.
The system integrates various sensors, actuators, and a microcontroller to monitor and control multiple key environmental parameters, including internal air temperature, internal air humidity, soil pH, and soil moisture.
By employing fuzzy inference techniques, the algorithms dynamically adjust control actions to enhance the efficiency of bio-fertigation and promote optimal plant growth Proposed algorithm In this section, we present the proposed algorithm that employs fuzzy logic to monitor and control four critical parameters in the greenhouse: internal air temperature, internal air humidity, soil pH, and soil These parameters are essential for maintaining a stable microclimate and supporting healthy plant By applying fuzzy logic techniques, the system ensures adaptive decision-making that improves microclimate regulation and enhances the efficiency of bio-fertigation.
The fuzzy system uses membership function graphs to illustrate how input parameters relate, where each x-axis value corresponds to two y-axis values .
Fuzzy logic, widely used in many fields including networking, helps convert multiple inputs into a single output .
, .
This study adopts the Mamdani model .
Aione of the most widely used fuzzy inference methodsAifollowing a four-step fuzzy process .
in Figure 1.
Algorithm 1 provides the pseudocode for controlling the greenhouse's microclimate.
Int J Elec & Comp Eng.
Vol.
No.
October 2025: 4555-4568
Int J Elec & Comp Eng
ISSN: 2088-8708
Figure 1.
Our approachAos fuzzy process model Algorithm 1.
Monitoring of the internal microclimate Fuzzy inputs: Internal air temperature (T.
, internal air humidity (H.
, internal soil pH (P.
, internal soil moisture (S.
, temperature threshold (T.
, humidity threshold (H.
, pH threshold (P.
, soil moisture threshold (S.
Fuzzy outputs: optimal temperature, optimal humidity, optimal soil pH, optimal soil moisture.
Initialize input parameters.
Define fuzzy sets and membership functions for each parameter.
Capture the real-time values of the input parameters.
Fuzzify the input parameters: Calculate the degree of membership for each input variable.
Perform fuzzy inference: Apply fuzzy rules based on input parameters to generate fuzzy output.
Aggregate the results from all fuzzy rules.
Defuzzify the output: Convert the fuzzy output into a precise control value.
Adjust the greenhouse systems .
eating, ventilation, irrigation, fertigatio.
based on the defuzzified output.
Repeat the process in real-time to maintain the optimal microclimate.
Fuzzy logic uses linguistic variablesAiranging between true and falseAito represent the strength of relationships among metrics and to determine the resulting output .
, .
Table 1 defines how the fuzzy logic system interprets real-world sensor values .
ike temperature and humidit.
and decides what to do with output devices .
ike a fan or heate.
using fuzzy sets.
The membership function shapes were chosen for their simplicity and suitability to the input data.
Their boundaries were defined through expert input from farmers and agronomists and by analyzing real environmental data, ensuring accurate representation of linguistic terms and real-world conditions.
Table 1.
Linguistic variables and Membership functions Variable type Input Output Variable Internal air temperature (AC) Internal air humidity (%) Soil pH Soil moisture (%) Fan (Speed %) Heater (Power %) Water pump 1 for irrigation (%) Water pump 2 for fertilization (%) Water pump 3 for fertigation (%) Linguistic terms (MF.
Low.
Average.
High Low.
Moderate.
High Acidic.
Neutral.
Basic Dry, moderately wet.
Saturated Off.
Low.
High Off.
Low.
High Off.
Low.
High Off.
Low.
High Off.
Low.
High Type of MF Triangular Triangular Triangular Triangular Triangular Triangular Triangular Triangular Triangular Range .
The fuzzy graph for the air temperature parameter, with low, average, and high membership functions, is shown in Figure 2.
50 AC is the highest recorded temperature in this investigation.
The air humidity fuzzy graph, which ranges from 0% to 100%, is shown in Figure 3.
There are three types of air humidity: low, moderate, and high.
The fuzzy soil pH chart, which is divided into three categories: acidic, neutral, and basic, is shown in Figure 4.
The fuzzy chart for soil moisture, which is divided into three categories: dry, moderately wet, and saturated, is shown in Figure 5.
Figure 6 shows the fuzzy membership function for the heater, with power levels categorized as low, off, or high based on temperature deviations from the threshold.
Figure 7 displays the fuzzy membership function for the fan, with power levels determined by temperature and humidity levels relative to their optimal thresholds.
Development of a fuzzy logic-based greenhouse system for optimizing A (Achouak Touham.
A ISSN: 2088-8708 Figures 8, 9, and 10 illustrate the fuzzy membership functions for irrigation, fertilization, and Irrigation control depends on soil moisture and humidity levels, with power levels set to low, off, or high accordingly.
Fertilization is regulated based on soil pH, adjusting power levels based on deviations from the optimal range.
Fertigation considers both soil pH and moisture, combining conditions to determine the appropriate control power level.
The fuzzy rule base used to connect the input-output membership functions is shown in Figure 11.
In our study, we have 81 rules.
The fuzzy rules were primarily formulated based on expert knowledge from local farmers, agronomists, and greenhouse technicians, combined with observations from previous studies and real-world behavior of the crop under varying climatic and soil conditions .
Ae.
The IF-THEN guidelines listed in Table 2 govern how the fuzzy inference system functions.
To connect different language variables, these rules use certain fuzzy logic operators like AuANDAy or AuOR.
Ay Figure 2.
The membership function of air Figure 3.
The membership function of air Figure 4.
The membership function of soil pH Figure 5.
The membership function of soil moisture Figure 6.
The membership function of the heater Figure 7.
The membership function of the fan Figure 8.
The membership function of irrigation Figure 9.
The membership function of fertilization Int J Elec & Comp Eng.
Vol.
No.
October 2025: 4555-4568
Int J Elec & Comp Eng
ISSN: 2088-8708
Figure 10.
The membership function of fertigation pump Figure 11.
Fuzzy rule base Table 2.
Some fuzzy rules Temperature Low Humidity Low Soil pH Acidic Soil moisture Dry Low Moderate Neutral Moderately wet Average High Neutral Saturated Average Low Basic Dry High High Basic Saturated Actuators decision Fan=Off Heater=High Water pump 1=High Water pump 2=High Water pump 3=High Fan=Low Heater=High Water pump 1=Low Water pump 2=Off Water pump 3=Off Fan=Low Heater=Low Water pump 1=Off Water pump 2=Off Water pump 3=Off Fan=Low Heater=Low Water pump 1=Low Water pump 2=Low Water pump 3=Low Fan=High Heater=Off Water pump 1=Off Water pump 2=Low Water pump 3=Off Block diagram of the fuzzy control system The overall structure of the fuzzy logic-based greenhouse control system is represented in the block diagram in Figure 12.
The process consists of five main stages:
Environmental sensors .
: the system starts by collecting real-time data from four key sensors that measure: internal air temperature, internal air humidity, soil pH, and soil moisture.
Fuzzification module: the sensor readings .
risp input value.
are passed to the fuzzification module.
Here, each input is mapped to corresponding linguistic variables .
, off, low, hig.
using membership functions.
Fuzzy inference engine: based on the fuzzified inputs, a set of expert-defined fuzzy rules (IFAeTHEN statement.
is applied.
These rules model the decision-making process and determine the appropriate control actions under varying environmental conditions.
Development of a fuzzy logic-based greenhouse system for optimizing A (Achouak Touham.
A ISSN: 2088-8708 Defuzzification module: The fuzzy outputs produced by the inference engine are converted into precise .
values using the centroid defuzzification method.
This ensures that the system provides continuous, smooth control actions.
Actuator outputs: The final crisp control signals are sent to the corresponding actuators: fan, heater, water pump .
or irrigatio.
, fertilization pump, and fertigation pump .
or combined watering and fertilizatio.
This entire process runs continuously in real-time, ensuring optimal growing conditions inside the greenhouse by responding adaptively to environmental changes.
Figure 12.
Block diagram RESULTS AND DISCUSSION The section begins by describing the experimental setup, and then presents a detailed analysis of the results achieved.
Experimental setting This study, conducted in 2023/2024 at Tahri Mohammed University.
Bychar.
Algeria, aimed to enhance the fertigation process for farmers.
The experimental setup used a glass greenhouse model in Figure 13 measuring 80 y 40 y 40 cm, divided into two sections: one for the plant, sensors, and actuators, and the other for housing three tanks .
ater, organic fertilizer, and mixin.
and the microcontrollers.
Moreover, our greenhouse includes:
Oe An Arduino Uno board, based on the ATMega328 microcontroller.
Oe An ESP8266 integrated circuit.
Oe A DHT11 temperature and humidity sensor.
Oe A soil moisture sensor.
Oe An HY-SRF05 ultrasonic sensor.
Oe A PH sensor (PH-4502C).
Oe Two 12V fans.
Oe A heater.
Oe Three 12V water pumps, one for each tank.
Figure 13.
Our experimental greenhouse Int J Elec & Comp Eng.
Vol.
No.
October 2025: 4555-4568
Int J Elec & Comp Eng
ISSN: 2088-8708
Experimental study for the optimization of fertilization measures In Laboratory No.
04 at Tahri Mohamed University Ae Bychar.
Algeria, two 500 ml solutions were prepared: tap water with a pH of 6.
97 and a liquid organic fertilizer .
ade by soaking sheep manure in wate.
with a pH of 4.
Both measurements were taken at 25 AC, as shown in Figure 14.
Figure 14.
Measured the mixture with a pH meter Steps of the experiment The experimental steps included:
After measuring the pH values, 10 ml of the liquid fertilizer was added to the tap water using a pipette and mixed thoroughly to prepare the fertigation solution.
The pH electrode is placed into the mixture, and the pH value is read directly from the connected electronic meter's display.
After each measurement, the pH electrode is rinsed with distilled water, left for ten seconds, and then wiped with Joseph paper to prepare it for the next measurement after additional fertilizer is added.
We measure the pH of the mixture and repeat the process until we obtain the following results in Table 3.
Table 3.
pH results and the volume of the mixture using a pH meter The 1st case The 2nd case The 3rd case Volume of the sample before adding the fertilizer 500 ml 510 ml 520 ml pH of the sample before adding the Volume of 10 ml 10 ml 10 ml Volume of the sample after adding the fertilizer 510 ml 520 ml 530 ml pH of the sample after adding the Analysis of results The experiment showed that adding 20 ml of organic fertilizer to 500 ml of water with an initial pH 94Ae6.
97 adjusts the mixtureAos pH to fall within the optimal range .
Ae6.
for tomato plants.
Adding more than 20 ml caused the pH to fall below the suitable range.
This result was confirmed through repeated testing in Figure 15.
Figure 15.
Results using a pH meter Proof Assuming we have 500 ml of water that we wish to use for fertilization, this is equivalent to 100%, so 20 ml of liquid fertilizer is equivalent to 4% .
: Total volume of water=500 ml and volume of fertilizer added=20 ml.
Development of a fuzzy logic-based greenhouse system for optimizing A (Achouak Touham.
A ISSN: 2088-8708 Now, to find the percentage of fertilizer in the solution:
ycEycyceycayceycuycycayciyce ycuyce yceyceycycycnycoycnycyceyc = ( ycOycuycoycycoyce ycuyce yceyceycycycnycoycnycyceyc ycNycuycycayco ycycuycoycycoyce ycuyce ycycuycoycycycnycuycu ) y 100 Substitute the values:
ycEycyceycayceycuycycayciyce ycuyce yceyceycycycnycoycnycyceyc = ( 20 ycoyco 500 ycoyco ) y 100=4% We know that 20 ml of fertilizer is needed for 500 ml of water, which is equivalent to 4%.
So, the amount of fertilizer required for any volume of water ycO can be calculated based on this ratio .
ycEyceycycayceycuycycayciyce ycuyce yceyceycycycnycoycnycyceyc = ( ycO y4 And, the volume of fertilizer added is:
ycOycuycoycycoyce ycuyce yceyceycycycnycoycnycyceyc = ( ycO y20 Results This study applied a real benchmark over a spring season (May 14Ae.
, using captured data on internal air temperature, humidity, soil pH, and soil moisture.
The objective was to maintain these parameters within thresholds defined in consultation with local farmers as shown in Table 4 and to optimize tomato plant fertigation using a bio-fertigation approach.
Results plotted in MATLAB show that all four parameters were effectively controlled, ensuring efficient fertigation.
Table 4.
Thresholds of the four climatic parameters for the tomato greenhouse Temperature Humidity Soil pH
Soil moisture Minimum threshold
15 AC
Maximum threshold
25 AC
In our work, we used six actuators in the:
First fan to regulate the temperature.
Second fan to regulate the humidity.
Heater to regulate the temperature.
Water pump 1 to distribute water.
Water pump 2 to distribute bio-fertilizer.
Water pump 3 to distribute liquid bio-fertilizer .
ater bio-fertilize.
Figures 16, 17, 18 and 19 present monitoring of the air temperature, air humidity, soil pH, and soil moisture over a sequence of hours, respectively.
These figures illustrate the indoor air temperature, indoor air humidity, indoor soil pH, and indoor soil moisture.
After applying our fuzzy logic algorithm, we observed that the four internal microclimatic parameters of our greenhouse were well adjusted within the specified Specifically, the measured parameters, over the sequence of hours, remained between the minimum and maximum thresholds.
Additionally, the actuators, including the heater, fans, and three water pumps, functioned effectively to regulate the internal microclimate and fertigate the tomato plants.
In our study, we have 81 cases.
For Case 1.
If (.
emperature<15 AC) AND .
umidity<60%) AND .
oil moisture<70%) AND .
H<.
) OR (.
emperature<15 AC) AND .
umidity<60%) AND .
oil moisture<70%) AND .
H>6.
), then the actuators for the heater, water pump 1, and water pump 3 are activated.
After a few seconds, the actuators stop working, and the captured parameters stabilize between the minimum and maximum thresholds.
Case 2.
If (.
emperature is optima.
AND .
umidity<60%) AND .
oil moisture<70%) AND .
H<.
) OR (.
emperature is optima.
AND .
umidity is optima.
AND .
oil moisture<70%) AND .
H>6.
), then the actuators for water pump 1, water pump 2, and water pump 3 are activated.
After a few seconds, the actuators stop working, and the captured parameters stabilize between the minimum and maximum thresholds.
Int J Elec & Comp Eng.
Vol.
No.
October 2025: 4555-4568
Int J Elec & Comp Eng
ISSN: 2088-8708
Case 3.
If (.
emperature>25 AC) AND .
umidity<60%) AND .
oil moisture is optima.
AND .
H<.
) OR (.
emperature>25 AC) AND .
umidity<60%) AND .
oil moisture is optima.
AND .
H>6.
), then the actuators for the fan, water pump 1, and water pump 2 are activated.
After a few seconds, the actuators stop working, and the captured parameters stabilize between the minimum and maximum thresholds.
Figure 16.
Monitoring the air temperature of our greenhouse Figure 17.
Monitoring the air humidity of our greenhouse Figure 18.
Monitoring the soil pH of our greenhouse Development of a fuzzy logic-based greenhouse system for optimizing A (Achouak Touham.
A ISSN: 2088-8708 Figure 19.
Monitoring the soil moisture of our greenhouse Table 5 summarizes the output responses of the fuzzy inference system based on four different environmental conditions measured by sensors.
Each test case includes values for internal air temperature, air humidity, soil pH, and soil moisture, which are processed using fuzzy logic rules to control five actuators:
fan, heater, irrigation pump, fertilization pump, and fertigation pump.
For example, in test case 1, low temperature and highly acidic soil activate the heater, fertilization, and fertigation systems, while the fan remains off.
In contrast, test case 5 shows a hot and humid environment with alkaline soil, triggering a highspeed fan and fertilization, but no irrigation or fertigation due to sufficient moisture.
The system demonstrates adaptive control based on combined climate and soil conditions.
To demonstrate how the optimization of our fuzzy logic-based system enhances the efficiency of bio-fertigation, we calculated the percentage improvements in water efficiency, fertilizer usage, and crop yields using .
ycOycaycyceyc ycycaycycnycuyci (%) = ( ycOycaycyceycycycycayccycnycycnycuycuycayco OeycOycaycyceycyceycycycyc ycOycaycyceycycycycayccycnycycnycuycuycayco yayceycycycnycoycnycyceyc ycycaycycnycuyci (%) = ( ycUycnyceycoycc ycnycuycaycyceycaycyce (%) = ( .
) y 100 yayceycycycnycoycnycyceycycycycayccycnycycnycuycuycayco Oeyayceycycycnycoycnycyceycyceycycycyc yayceycycycnycoycnycyceycyceycycycyc Oeyayceycycycnycoycnycyceycycycycayccycnycycnycuycuycuycayco ) y 100 ) y 100 The implementation of the fuzzy logic-based bio-fertigation system led to measurable improvements in key agricultural performance metrics.
As shown in Table 6, water usage was reduced by approximately 27.
fertilizer usage decreased by over 58.
82%, and crop yield increased by more than 47.
5% compared to the traditional method.
These results confirm the system's ability to optimize resource use and improve Table 5.
Input data and output results of the fuzzy inference system
Test
Temperature (AC) Humidity (%) Soil
Soil
moisture (%) Fan
(%)
Heater
(%)
Irrigation (%) Fertilization (%) Table 6.
Percentage improvements results Parameter Water usage (L/h.
Fertilizer usage .
g/h.
Crop yields .
ons/h.
Traditional method Fuzzy logic system Int J Elec & Comp Eng.
Vol.
No.
October 2025: 4555-4568
Improvement (%) Fertigation (%) Int J Elec & Comp Eng
ISSN: 2088-8708
Statistical estimators The performance of our contribution can be assessed using four criteria: mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), and mean absolute percentage error (MAPE).
Equations are used to calculate these metrics from the obtained data set in order to compare the ideal values with the measured ones.
Equations .
Ae.
, in that order .
, .
Ae.
Ocycuycn=.
ycEycoycn Oe ycEycuycn | .
ycAycIya = Ocycuycn=1.
cEycoycn Oe ycEycuycn )2 .
ycIycAycIya = OoycAycIya ycAyaya = ycu ycAyaycEya = ycu Ocycuycn=1 | ycEycoycn OeycEycuycn ycEycoycn Where PCo and PCe are, respectively, the measured parameter and the optimized parameter.
The results are summarized in Table 7, which presents a comparison between the conditions before and after applying control to our greenhouse using the proposed fuzzy logic algorithm.
Table 7.
Statistical estimators results MAE
MSE
RMSE
MAPE (%)
Before After Where T.
P and S are, respectively, air temperature, air humidity, soil pH, and soil moisture.
We observe that the values of MAE.
MSE.
RMSE, and MAPE after implementing our algorithm are reduced to zero, indicating that the system meets acceptable thresholds for accuracy and precision.
CONCLUSION
The results of this study demonstrate the effectiveness of the fuzzy logic-based control system in optimizing the greenhouse microclimate, particularly by maintaining balanced internal air temperature, humidity, soil pH, and soil moisture.
The system successfully automated actuator responsesAiincluding fan, heater, and irrigation/fertilization/fertigation pumpsAibased on real-time sensor inputs.
Statistical evaluations using MAE.
MSE.
RMSE, and MAPE confirmed the model's accuracy and reliability.
Moreover, the implementation of this algorithm led to measurable improvements in resource efficiency, with significant reductions in water and fertilizer usage.
Most importantly, the optimized environmental conditions resulted in a noticeable increase in tomato yield, confirming the practical benefit of the system in precision agriculture.
These findings highlight the potential of fuzzy logic as a robust and intelligent solution for sustainable crop Looking ahead, in our future work, we plan to include additional parameters, add other crops, combine our fuzzy logic system with machine learning or optimization algorithms and integrate cloud computing, and AI to further optimize crop growth through real-time data analytics.
Finally, we aim to explore hydroponic greenhouses, vertical farming, and urban farming as part of our ongoing research.
ACKNOWLEDGMENTS
The authors express gratitude to individuals and institutions for their guidance, support, and contributions, including Touhami Nawal.
Professor Benahmed Khelifa, the INIE and LSEE laboratory teams, and the anonymous reviewers whose feedback improved the paper's quality.
FUNDING INFORMATION
According to the authors, this work was not supported by any funding source.
Development of a fuzzy logic-based greenhouse system for optimizing A (Achouak Touham.
A ISSN: 2088-8708
AUTHOR CONTRIBUTIONS STATEMENT
This journal uses the Contributor Roles Taxonomy (CRediT) to recognize individual author contributions, reduce authorship disputes, and facilitate collaboration.
Name of Author Achouak Touhami Amina Bourouis Amel Mahammedi Sana Mechraoui Sana Touhami C : Conceptualization M : Methodology So : Software Va : Validation Fo : Formal analysis ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue I : Investigation R : Resources D : Data Curation O : Writing - Original Draft E : Writing - Review & Editing ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue ue Vi : Visualization Su : Supervision P : Project administration Fu : Funding acquisition CONFLICT OF INTEREST STATEMENT There are no conflicts of interest, according to the authors.
DATA AVAILABILITY
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
REFERENCES