Evaluation of white oyster
mushroom production under different substrate types: a completely randomized
design vs. simulation experiment
Evaluación de la producción
de hongos ostra blanco bajo diferentes tipos de sustrato: un diseño
completamente aleatorizado vs. simulación del
experimento
Jessica Alexandra Marcatoma-Tixi
1
Luis Antonio Vera-Rojas 2
Kerly Doménica Damián-Espín 3
José Carlos Gallegos-Meza 4
Published Edwards Deming Higher
Technological Institute. Quito
- Ecuador Periodicity April - June Vol. 1, Num. 25, 2025 pp.
85-103 http://centrosuragraria.com/index.php/revista Dates of receipt Received: February, 2024 Approved: March 20, 2025 Correspondence author rjmarcatoma@espoch.edu.ec Creative Commons License Creative Commons License,
Attribution-NonCommercial-ShareAlike 4.0
International.https://creativecommons.org/licenses/by-nc-sa/4.0/deed.es
1 Engineer in Computer Statistics, Escuela
Superior Politécnica de Chimborazo, Riobamba, Ecuador rjmarcatoma@espoch.edu.ec, https://orcid.org/0000-0001-9531-3234 2 PhD in
Mathematics, Escuela Superior Politécnica de Chimborazo, Riobamba, Ecuador.
lvera@espoch.edu.ec, https://orcid.org/0000-0002-7087-8447 3 Student of
Statistics Engineering, Escuela Superior Politécnica de Chimborazo,
Riobamba, Ecuador. kerly.damian@espoch.edu.ec,
https://orcid.org/0009-0003-6269-9654 4 Student of Statistics Engineering, Escuela
Superior Politécnica de Chimborazo, Riobamba, Ecuador. jose.gallegos@espoch.edu.ec
https://orcid.org/0009-0007-6831-6131
Key words: Fungi, simulation,
designs, assumptions, Tukey, statistics.
Resumen:
El cultivo de hongos Ostra blanco ha cobrado
relevancia por sus beneficios económicos, nutricionales y ecológicos. Este
estudio tiene como objetivo identificar el sustrato más eficiente para
maximizar la producción de hongos mediante un Diseño Completamente
Aleatorizado. Se evaluaron tres sustratos (paja, aserrín y cáscara de maní) con
dos réplicas por tratamiento. El análisis de datos incluyó análisis de
varianza, comprobación de supuestos, pruebas post-hoc para determinar
diferencias significativas y analizó la posibilidad de simular este experimento
bajo cierta distribución. Los resultados del análisis exploratorio indicaron
que la paja presentó el mayor rendimiento promedio de 212.33 g, mientras que la
cáscara de maní mostró la menor variabilidad, asegurando mayor estabilidad en
la producción. Las pruebas estadísticas confirmaron el cumplimiento de los
supuestos de normalidad, varianza constante e independencia. La prueba de Tukey reveló diferencias significativas entre los
sustratos, destacando que la paja y el aserrín permiten maximizar la producción
de hongos Ostra blanco. Este estudio resalta la importancia de la selección del
sustrato en la optimización del cultivo de hongos y sugiere la simulación bajo
la distribución normal como una herramienta para reducir costos experimentales
y mejorar la eficiencia.
Palabras Clave: Hongos, simulación, diseños, supuestos, Tukey, estadística.
Introduction
The cultivation
of edible mushrooms, such as Pleurotus ostreatus or better known as white oyster mushrooms, has
gained relevance for its contribution to food security, environmental
sustainability and the generation of value in agricultural production chains.
The optimization of cultivation conditions is key to
maximize the productivity and quality of these mushrooms (Ramírez,
2019).
According to Castañeda and Martínez, (2019),
increasing the inoculation rate significantly improves the yield of white
Oyster mushrooms and the different culture media influence in vitro mycelial
growth, providing fundamental information for industrialized processes.
Mushrooms of the genus White Oyster mushrooms, known as mushrooms, are valued
for their economic, nutritional and ecological relevance; their ability to grow
on low-cost substrates, such as agricultural waste, positions them as a
sustainable alternative in modern agriculture, promoting the circular economy
by reducing waste (Castañeda and Martínez,
2019); also, their high protein content and richness in vitamins and essential
minerals make them a strategic food to improve food security (Gómez et al,
2021).
In Paraguay,
the journal of Agrarian Sciences has investigated and published about the
cultivation of Pleurotus ostreatus
and Ganoderma lucidum using
wood and agricultural residues. The results showed that these substrates favor
the growth and production of both fungi (De Madrignac,
2019); in Spain, research published by International Journal of Molecular
Sciences has analyzed the impact of different lignocellulosic substrates on the
production of white oyster mushrooms, determining that the combination of wheat
straw and vine residues significantly improves the biological efficiency of the
fungus (Pineda, 2020).
In Ecuador, the
province of Chimborazo has ideal conditions for mushroom production due to its
climate and availability of agricultural residues such as sawdust and sugarcane
bagasse. However, the lack of technological infrastructure and technical
knowledge limits its expansion. In contrast, for Gómez et al, (2021), in Zamora
Chinchipe, the commercialization of mushrooms has generated economic
opportunities and logistical challenges that require solutions adapted to the
regional context.
The use of
fungi in the present experiment seeks to maximize their production according to
different types of crops that use substrates rich in lignocellulose as the main
source of nutrients. According to Hossain et al., (2021), sawdust is widely
used due to its moisture retention capacity and its supply of key nutrients for
mycelial growth. For Ali and Usman, (2020), straw contains cellulose and
hemicellulose and, although its moisture retention is limited, its low cost and
wide availability make it a popular substrate. Prasad and Singh (2023) mention
that peanut shells are an agricultural by-product that provides a source of
cellulose and lignin, in addition to favoring substrate aeration, which is
crucial for fungal growth.
Methodology
This study employed a Completely Randomized
Design (CRD), an experimental approach that randomly assigns treatments to
experimental units, widely used in research where experimental conditions are
kept homogeneous, allowing to minimize sources of external variability and to
clearly evaluate the effects of independent variables . Different types of substrate were evaluated in
the production of white oyster mushrooms. For a more robust analysis and to
simulate different production scenarios, 100 simulation replicates were
generated based on specific probability distributions, which allowed modeling
the variability in production and evaluating the behavior under controlled but
stochastic conditions.
Substrate preparation began with cleaning and
grinding, followed by pasteurization at 65-75°C for 4 hours to eliminate
unwanted microorganisms, a fundamental technique to ensure culture viability
and minimize contaminations (Prasad and Singh, 2023). The substrates were mixed
with water to a humidity of 60-65%, tested by the fist method, and placed in
heavy-duty polyethylene plastic bags as experimental units. The mycelium was
inoculated uniformly in each experimental unit. The sealed bags were placed in
a temperature-controlled incubation chamber (25 ± 2°C) and total darkness for
30 days, with weekly monitoring to record the progress of colonization (Sharma
et al., 2022). After complete colonization, the bags were transferred to the
fruiting area under specific conditions: temperature between 15-22°C according
to Sharma et al., (2022), a humidity of 85-90% by manual irrigation with
non-chlorinated water, natural diffuse light and constant ventilation to
prevent CO2 accumulation. The period of each fruiting lasted approximately 21
days, with weighing and recording of the harvested mushrooms, also evaluating
their visual quality. The accumulated production of each experimental unit was
used as a response variable in the statistical analysis (Ali et al., 2020).
Protocols of good agricultural practices were implemented, including the use of
sterilized tools and strict control of environmental conditions, documenting
the data in specific record sheets for each experimental unit. A Completely
Randomized Design (CRD), an experimental approach that randomly assigns the
different types of substrates to the experimental units, was used in this
study. The substrates evaluated were: sawdust, straw and peanut shells,
selected for their availability and potential to be used in sustainable crops
according to Naseer et al., 2022). The sample
consisted of 9 experimental units, distributed in three treatments with two
replicates each.
Response variable: Y: =" Production of white oyster
mushrooms" (grams).
Factor of interest: F: =" Substrate", which has three
levels:
In order to perform a more robust analysis and
simulate different production scenarios, 100 simulation replicates were
generated for each treatment. This simulation allowed modeling the variability
in production and evaluating the behavior of the treatments under controlled
but stochastic conditions.
Analysis techniques included:
Descriptive analysis: Measures of central tendency and dispersion of
production were calculated for each treatment.
Analysis of variance (ANOVA): Analysis of variance was used to determine if
there were significant differences in fungal production among the different
treatments. This analysis is appropriate for experimental designs with only one
factor of interest (Saini, 2023).
Validation of assumptions: To ensure the validity of the analysis of
variance, the following fundamental assumptions were checked for Gabriel,
(2021).
H0 : The residuals have a normal distribution.
H1 : The residuals do not have a normal
distribution.
The Shapiro-Wilk test statistic is defined as:
The test statistic is given by:
Where:
H0 : The residuals are independent.
H1 : There is correlation between the residuals.
The statistic for the gust test is calculated
as:
Post-hoc tests: If significant differences were found in the
ANOVA, a Tukey mean comparison test was analyzed to identify specific
differences between factor levels as recommended by Perez, (2020).
They were used to compare the differences
between sample means with the critical value determined by:
Data simulation and comparison with
distributions: Once
the experiment was completed, the response variable was modeled based on three
simulation models such as normal, uniform and exponential distribution under
the following equations.
It is important to note that 100 simulation
replicates were run for each model and then, through numerical indicators , the model with the best fit was chosen to
replicate the data obtained in the experiment.
Results
This section presents the results obtained from
the analysis of the completely randomized design, focused on evaluating the
effect of different types of substrate on the production of white oyster
mushrooms.
Table 1 shows the descriptive statistical
analysis of the three substrates evaluated for the production of white oyster
mushrooms. The results showed significant differences in the average yield of
the treatments.
Table 1. Comparative statistical analysis of substrates for white oyster mushroom
production.
|
Measure |
Straw |
Peanut shells |
Sawdust |
|
Media |
212.33 |
71.83 |
109.17 |
|
Standard error |
46.81 |
19.22 |
41.56 |
|
Median |
215.00 |
61.00 |
94.50 |
|
Standard deviation |
114.67 |
47.07 |
101.80 |
|
Sample variance |
13149.07 |
2215.77 |
10363.77 |
|
Kurtosis |
-2.35 |
-2.55 |
-2.46 |
|
Asymmetry coefficient |
0.06 |
0.29 |
0.27 |
|
Range |
279.00 |
104.00 |
229.00 |
|
Minimum |
78.00 |
24.00 |
16.00 |
|
Maximum |
357.00 |
128.00 |
245.00 |
|
Sum |
1274.00 |
431.00 |
655.00 |
Straw presented the highest average yield, with
an average of 212.33 g, far surpassing the other treatments. Sawdust obtained
an average yield of 109.17 g, while peanut shells showed the lowest yield, with
a mean of 71.83 g. These differences highlighted the potential of straw as a
more suitable substrate for white oyster mushroom production under the
conditions evaluated.
In terms of data variability, high standard
deviations were observed in all treatments, indicating a high dispersion in
individual yields. Straw had the highest standard deviation, followed by
sawdust, while peanut shells had the lowest dispersion.
On the other hand, the kurtosis results
indicated that the distributions of the three substrates are platykurtic, with negative kurtosis values. This implies
that the data are less concentrated around the mean compared to a normal
distribution. In addition, the skewness coefficients revealed a slight positive
skewness in all treatments, suggesting a tendency toward higher values than the
mean.
In terms of variable length, it was observed
that straw presented the highest range of values, while sawdust and peanut
shells had ranges of 229.00 g and 104.00 g, respectively.
Table 2.
Experiment data matrix
|
Substrates |
Production
(grams) |
|||||
|
1: Straw |
297 |
284 |
357 |
146 |
78 |
112 |
|
2: Peanut Shells |
123 |
87 |
128 |
34 |
35 |
24 |
|
3: Sawdust |
245 |
186 |
166 |
23 |
16 |
19 |
The initial exploratory analysis showed that
production means varied among treatments, with the "Straw" substrate
showing the highest values, while treatment 2 showed the greatest variability.
The analysis of variance of the experiment is
shown below.
Table 3. Analysis of Variance
(ANOVA)
|
FV |
SS |
GL |
MS |
Fo |
p-value |
|
Substrate |
63554,77778 |
2 |
31777,38889 |
3,7053 |
0,0492 |
|
Error |
128643 |
15 |
8576,2 |
- |
- |
|
Total |
192197,7778 |
17 |
11305,75163 |
- |
- |
Since the observed value was greater than the
critical value, H0 was rejected and it was concluded that the substrate factor
has an effect on white oyster mushroom production. This means that at least one
pair of substrate types generated a different average mushroom production.
Table 4.
Summary of ANOVA
|
Significance
level |
0,05 |
|
Observed value |
3,705 |
|
Critical value |
3,682 |
|
Critical region |
(3.68; )∞ |
|
Decision |
Reject H0 |
Verification of assumptions
A graphical procedure to verify compliance with
the assumption of normality is the QQ Plot of Probability of Normality on the model
residuals.
Figure 1. Summary of ANOVA
In Figure 1, it was observed that the points
are approximately on the bisector, that is, the sample quantiles are very
similar to the population quantiles. Therefore, it is graphically concluded
that the sample of residuals comes from a normal distribution.
To contrast this result, the Shapiro-Wilks
statistical test was applied.
Table 5. Summary
of the Shapiro-Wilks Test
|
Significance level |
0,05 |
|
Observed value |
0,9012 |
|
Critical value |
0,982 |
|
Critical region |
(0,98; ∞) |
|
Decision |
Do not reject H0 |
Since the observed value was less than the
critical value, H0 was not rejected and it was concluded that the
data come from a normal distribution F (x) is normal. And this agrees with what
was observed in the QQ plot
Homocedasticity
Two techniques were used for variance testing
in relation to the treatments used, the graphical
procedure used as base information the predicted data against the residuals.
Figure 2. Graph of averages
In Figure 2, it was
observed that the points are randomly distributed in a horizontal band.
Therefore, graphically it was concluded that the assumption of constant
variance is fulfilled, the second technique applied Bartlett's statistical test
for homogeneity of variancesAnd it can be seen that
in the variables weight of 100 g mature, weight of production g/plant, and
yield of gold coffee with and without adjustment, statistical differences were
found between treatments; not so in the variables Weight 100 g of dry parchment
coffee/plant and the variable Conversion cherry coffee to gold coffee.
Table 6. Summary of Bartlett's Test
|
Significance level |
0,05 |
|
Observed value |
0,3207 |
|
Critical value |
5,991 |
|
Critical region |
(5, 991; ∞) |
|
Decision |
Do not reject H0 |
Since the observed
value was less than the critical value, H0 was not rejected and it
was concluded that the assumption of constant variance was met. And this agrees
with what was observed in the graphical technique.
Independence
The assumption of independence
in the residuals was verified by plotting the order
The weight of humus
treatment D1 was 1.6 kg, followed by agricultural gypsum with 1.4 kg and in
last place the control treatment with 0.15, well below the other treatments.
Figure 3. Independence Graph
In Figure 3, a clearly defined trend or random
pattern in the points was observed. Therefore, graphically it was concluded
that it complies with the independence assumption; on the other hand, the
statistical test of Rachas was applied with the
following findings.
Table 7. Summary
of the Gust Test
|
Significance level |
0,05 |
|
Observed value |
-1,915 |
|
Critical value |
1,96 |
|
Critical region |
|
|
Decision |
Do not reject H0 |
Since the
observed value was less than the critical value, H0 was not rejected and it was
concluded that the assumption of independence is fulfilled. And this is in
agreement with what was observed in the graphical technique.
The following is
the comparison of means test for the choice of treatment to maximize mushroom
production in the sense that the assumptions of the ANOVA model were fully met
with a significance of .
Post-Hoc Testing
Table 8. Comparison
of Tukey's Test
|
Difference Population |
Sample
Difference |
Tukey |
Decision |
Interpretation |
|
u1 - u2 |
140,5 |
138,7516 |
Reject H0: Significant |
Substrates
1 and 2 generate different average levels of oyster mushroom production. |
|
u1 - u3 |
103,1667 |
138,7516 |
Do
Not Reject H0: Not Significant |
Substrates
1 and 3 generate equal average levels of oyster mushroom production. |
|
u3 - u2 |
37,3333 |
138,7516 |
Do
Not Reject H0: Not Significant |
Substrates
3 and 2 generate equal average levels of oyster mushroom production. |
Substrate 1
(straw) and substrate 3 (sawdust) generated similar average levels, which
guarantees the maximization of the production of white oyster mushrooms.
However, to determine the best substrate option, it is necessary to consider
other factors, such as cost, availability, decomposition
time and management conditions, among other economic, operational and other
aspects.
Simulation
From the
experimental design, a random variable was simulated in order to model the
behavior of mushroom production under different substrates, without the need to
perform the experiment again.
During the
simulation process the response variable white oyster mushroom production was
compared using three simulation models (uniform, exponential and normal), under
the following evaluated parameters:
Table 9. Simulation Process Comparison
|
Experiment parameters |
||||||
|
Parameter |
Uniform
Distribution |
Exponential Distribution |
Normal Distribution |
|||
|
|
A |
B |
|
|
|
|
|
Values |
24 |
357 |
0,006 |
153,466 |
153,466 |
99,060 |
|
Simulation statistics |
||||||
|
Statistics |
139,411 |
251,320 |
161,505 |
0,006 |
204,268 |
7,590 |
|
Difference between parameters and statistics |
||||||
|
Differences |
115,411 |
105,079 |
161,498 |
153,460 |
50,802 |
91,470 |
Table 9 shows the
comparison of the parameters obtained from the experimentation and the
simulation statistics the results showed that it is possible to simulate the
response variable mushroom weight without the need to experiment again under
the normal distribution.
In the present
project, the impact of different substrates on the production of white oyster
mushrooms in the province of Chimborazo was evaluated, with the objective of
identifying the substrate that maximizes the crop production and if there is
any distribution with which it can be simulated for future applications
avoiding costs, time, among others. The results obtained showed that mushroom
production varied significantly according to the type of substrate used. A
study carried out in Colombia used a completely randomized design with four
treatments and four replications, using oak, conacaste
and liquid amber mulch as substrates. It was observed that oak reached a fungal
production of 70.57%, while the other substrates did not allow fungal growth,
(Acevedo, 2017). In comparison, in our study, certain agricultural residues as
in the straw substrate and sawdust substrate generate similar average levels
and that maximize the production of white Oyster mushrooms. However, to decide
the best substrate option, during the experimental process, two key aspects
were identified to be considered when choosing between them: economic and time.
which is in agreement with previous research
highlighting the influence of the culture medium on mushroom performance.
Conclusions
The exploratory
analysis of the three substrates evaluated for the production of white oyster
mushrooms showed that straw is the most efficient substrate, with an average
yield significantly higher than the other treatments. However, this substrate
also presented the greatest variability in its results, suggesting the need for
more rigorous control of growing conditions to optimize its performance.
Although peanut shells had the lowest average yield, they showed the least
variability among the three substrates evaluated, which makes them a viable
option for productions that prioritize consistency over quantity. Sawdust is an
intermediate alternative in terms of both yield and variability. Regarding the
application of the completely randomized design, it was confirmed that it
complies with the assumptions of normality, constant variance and independence,
which validates the results of the ANOVA. This analysis revealed that the type
of substrate has a significant effect on the production of white oyster
mushrooms. Given that at least one substrate type produces oyster mushrooms
differently from another in terms of quantity, the question arises: which
substrate types are significantly different from each other? Tukey's method
allows us to answer this question, as it is more precise at this level of
significance, since it keeps it constant in each comparison of means. That
said, it is concluded that the straw substrate and the sawdust substrate
generate similar mean levels and maximize the production of white oyster
mushrooms. However, in order to decide the best substrate option, two key
aspects were identified during the experimental process that should be
considered when choosing between them: quantity and time. From a quantitative
production perspective, the sawdust substrate is more favorable, while in terms
of time, the straw substrate offers greater results in a short period.
Therefore, the choice will depend on the specific priorities and needs of the
reader or user interested in these results. From the simulation applied to the
experimental variable, it was concluded that it is possible to model the
production of white oyster mushrooms using a normal distribution. This finding
represents a significant advance, since it allows predicting the production
behavior without the need to repeat the experiment physically, optimizing time
and resources. The simulation not only reinforces the results obtained in the
experimental design, but also provides a valuable tool to predict and improve
mushroom production. For future lines of experimentation with the objective of
obtaining more accurate conclusions, it is recommended to implement a
completely randomized block design (CRBD), also considering the time factor.
This approach will allow obtaining more robust results.
References
Acevedo Cárdenas, R. C. (2017).
Assessment and growth of Pleurotus ostreatus cultivation in four substrates generated from
agricultural production processes, in the municipality of Málaga
Santander. Universidad Nacional Abierta y a Distancia (UNAD).
https://repository.unad.edu.co/bitstream/handle/10596/13043/63398539.pdf.
Ali, M., Khan, S., and Usman, M.
(2020). Optimization of oyster mushroom (Pleurotus ostreatus) cultivation
using agricultural residues. Journal of Mycology, 56(2), 78-89.
Castañeda, C., and Martínez, D.
(2019). Increased productivity of Pleurotus
ostreatus by increasing inoculation rate. Revista
Mexicana de Micología, 47(1), 205-213.
De Madrignac (2019). Cultivation of
Pleurotus ostreatus and Ganoderma lucidum in Paraguay using agricultural
residues. Revista de Ciencias Agrarias, 36(1), 22-35.
Gabriel, J. (2021). Experimental
designs (2nd ed.). Ecuador.
Gómez, L., and Ramírez, J.
(2021). Assessing normality in experimental data: Methods and applications. International
Journal of Data Analysis, 10(1), 45-58.
Gómez, M., Pérez, R., and
Sánchez, L. (2021). Evaluation of the production and
protein value of Pleurotus ostreatus
in Ecuador. Avances en Ciencia e Ingeniería, 13(2), 45-55.
Hossain, M., Akter, S.,
and Rahman, M. (2021). Evaluation of
lignocellulosic substrates for sustainable mushroom production. International
Journal of Agricultural Research, 27(1), 33-42.
Naseer, F., Ahmad, I., and Saleem, M. (2022). Comparative analysis of yield performance in
oyster mushrooms grown on various substrates. Journal of Agricultural
Science, 29(2), 89-99.
Perez, E. S. (2020). Importance
of using proper post hoc test with ANOVA.
Pérez, J., and Ramírez, L. (2023). Generation of random variables for simulation
models. Journal of Simulation and Modeling, 15(2), 123-136.
Pineda, I. (2020). Influence of
lignocellulosic substrates on Pleurotus eryngii production in Spain. International Journal of
Molecular Sciences, 21(5), 1347-1358.
Prasad, R., Verma, T., and Singh, D. (2023). Effects of substrate sterilization
techniques on oyster mushroom yield. Mushroom Science and Technology, 12(1),
15-23.
Ramírez, A., and Gómez, L. (2019). Productive capacity of Pleurotus
ostreatus using alfalfa as substrate. Revista
Mexicana de Ciencias Agrícolas, 9(2), 145-152.
Saini, J., and Singh, P. (2023). Statistical
designs in agricultural experiments: A review. Journal of Biostatistics, 11(4),
120-134.
Sharma, P.,
Gupta, R., and Bhatt, J. (2022). Influence of environmental conditions on the fructification of Pleurotus
ostreatus. Mycology Advances, 24(3), 112-124.