Agile Innovation in Agribusiness: Implementing Scrum in the Peruvian Cocoa Production Chain

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the opportunity to read this interesting article. Indeed, the problem of accumulation of heavy metals in agricultural products is very relevant now. The danger of cadmium in food is underestimated by many experts. The authors devoted their article to a technique to avoid the accumulation of cadmium in cocoa beans. However, the essence of the method is not entirely clear. The authors indicate that biofertilizers were used, but there is no detailed description of these fertilizers and the method of their application. It is necessary to indicate this in the text of the article. It is also unclear how the reduction of cadmium content was achieved during fermentation. The article can be published after significant revision.

 

Author Response

COMMENTS:

Thank you for the opportunity to read this interesting article. Indeed, the problem of accumulation of heavy metals in agricultural products is very relevant now. The danger of cadmium in food is underestimated by many experts. The authors devoted their article to a technique to avoid the accumulation of cadmium in cocoa beans. However, the essence of the method is not entirely clear. The authors indicate that biofertilizers were used, but there is no detailed description of these fertilizers and the method of their application. It is necessary to indicate this in the text of the article. It is also unclear how the reduction of cadmium content was achieved during fermentation. The article can be published after significant revision.

RESPONSE:

Thank you very much for the review of the article and for taking the necessary time to provide your valuable comments. They undoubtedly strengthen the level and rigor we aim to achieve with this research.

Regarding the observation made, clarification was provided on the use of biofertilizers in SPRINT 4. The mixture of the components used in the process can be seen, expressed in kilograms per hectare for the application. Likewise, the presence of these fertilizers was substantiated, as well as how they contribute to the reduction of cadmium in cocoa according to the different combinations applied in each treatment.

With respect to SPRINT 3, the objective during fermentation was not to generate a reduction in cadmium levels, but rather to ensure optimal fermentation conditions in which the color characteristics of the cocoa beans required by the market could be evidenced. As demonstrated in SPRINT 3, optimal fermentation control ensures proper colorimetry.

The modification made based on your observation can be found on pages 10-11

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for asking me to review this manuscript. The paper aims to apply SCRUM methodology to the Peruvian cocoa production chain to improve efficiency, bean quality, and sustainability. The topic is practically relevant and the cadmium reduction data in Sprint 4 is promising. That said, the manuscript has major flaws that I outline below.

1. Research Gap and Framing

The author claims that gap is basically that SCRUM has not been applied in this context before. But that is not a sufficient gap. The authors need to explain what frameworks or approaches are normally used in the literature for managing agricultural production chains, what problems those approaches have, and why SCRUM is expected to do better. Without this, the reader cannot understand why SCRUM was chosen.

More critically, there is a disconnect between the framing and the actual findings. The paper is framed around SCRUM, but the results and discussion are almost entirely about agronomic outcomes: stem diameter, plant height, fermentation temperature, and cadmium levels. SCRUM is not really evaluated as a variable. So the question is: what is the real contribution here? Whatever sits at the center of the findings and discussion is the actual research problem, and the introduction should be built around that. If the contribution is about bio-organic treatments and cadmium mitigation, then the introduction needs to focus on that problem, show why it matters, review how the literature has addressed it, and identify the gap there. If the contribution is about SCRUM, then the results need to empirically show how SCRUM outperformed alternative approaches. Right now the paper promises one thing and delivers another.

2. Research Questions and Hypotheses

There are no explicit research questions, hypotheses, or a research model anywhere in the paper. The authors test specific relationships (treatment affects cadmium, fertilizer affects growth, etc.), and these should be hypothesized before being tested. If the claim is that X leads to Y, this needs to be stated upfront and grounded in the literature. I recommend formulating clear hypotheses that map onto the sprint objectives, and presenting a conceptual model that connects the variables.

3. Methodology

Several issues here. First, the paper conflates SCRUM (a project management tool) with the experimental design. Using sprints to organize work does not make the experiments more rigorous. These two layers need to be separated. Second, key experimental details are missing: no randomization, no replication strategy, no plot sizes, no discussion of confounding variables. Treatment labels change across sprints (M1-M3 vs T1-T6) without explanation. The "alive-alive" and "dead-dead" scenarios are never defined. Statistical tests are not clearly identified, there is no mention of post-hoc tests, and the SPC control limits in Sprint 3 are so wide that having all points within limits is not very meaningful. Also, there is no comparison with a non-SCRUM baseline, so the authors cannot attribute any outcome to SCRUM itself. Finally, Section 4 is titled "Methodology" but presents results, which is a structural error. The authors should consolidate the methodology into one clear section and separate the results properly.

4. Results and Discussion

The results need effect sizes with confidence intervals, not just p-values. The probability plots used in the figures are not standard for showing treatment effects; conventional formats like bar charts or boxplots would work better. In the discussion, the comparisons with literature are too shallow. The authors mostly say "our findings are consistent with" previous work without any quantitative comparison or discussion of where results diverge. The discussion needs much deeper engagement with the literature.

5. Practical Implications and Limitations

The paper claims the model is replicable but gives no practical guidance: no cost analysis, no implementation guidelines for farmers, no discussion of adoption barriers or scalability. For a paper in Sustainability, this is a notable gap. Additionally, the paper has no limitations section at all. The authors should acknowledge at minimum the single-site design, the absence of a non-SCRUM control, the short time horizon, and potential confounds from soil and climate variability. Directions for future research should follow from these.

6. References

The manuscript cites only 16 references, which is far too few. A paper like this needs at least 10 to 20 references in the introduction section alone. The authors need to significantly expand their literature review and cite more work in the introduction, state of the art, and discussion sections. A deeper engagement with the literature would help them better position their contribution and identify the actual research gap.

Author Response

COMMENT 1:

1. Research Gap and Framing

The author claims that gap is basically that SCRUM has not been applied in this context before. But that is not a sufficient gap. The authors need to explain what frameworks or approaches are normally used in the literature for managing agricultural production chains, what problems those approaches have, and why SCRUM is expected to do better. Without this, the reader cannot understand why SCRUM was chosen.

More critically, there is a disconnect between the framing and the actual findings. The paper is framed around SCRUM, but the results and discussion are almost entirely about agronomic outcomes: stem diameter, plant height, fermentation temperature, and cadmium levels. SCRUM is not really evaluated as a variable. So the question is: what is the real contribution here? Whatever sits at the center of the findings and discussion is the actual research problem, and the introduction should be built around that. If the contribution is about bio-organic treatments and cadmium mitigation, then the introduction needs to focus on that problem, show why it matters, review how the literature has addressed it, and identify the gap there. If the contribution is about SCRUM, then the results need to empirically show how SCRUM outperformed alternative approaches. Right now the paper promises one thing and delivers another.

RESPONSE:

Thank you very much for the review of the article and for taking the necessary time to provide your valuable comments. They undoubtedly strengthen the level and rigor we aim to achieve with this research.

Regarding Comment 1, the current real gap in the cocoa production chain in Peru (where this research is conducted) lies in the limited knowledge of farmers engaged in this type of crop, particularly in addressing the agronomic problems detailed in the study.

In response to the need for an agile framework that places people at the center of the process—and of the outcomes derived from the implementation of good practices—SCRUM was identified as the most appropriate method to drive the necessary change in cocoa cultivation. Through work packages (SPRINTS) designed as incremental and iterative responses within an agile framework, it is possible not only to implement technical improvements but also to foster knowledge development among the individuals carrying out these practices.

This clarification has been incorporated on page 2

COMMENT 2:

2. Research Questions and Hypotheses

There are no explicit research questions, hypotheses, or a research model anywhere in the paper. The authors test specific relationships (treatment affects cadmium, fertilizer affects growth, etc.), and these should be hypothesized before being tested. If the claim is that X leads to Y, this needs to be stated upfront and grounded in the literature. I recommend formulating clear hypotheses that map onto the sprint objectives, and presenting a conceptual model that connects the variables.

RESPONSE:

Regarding Comment 2, explicit hypotheses cannot be formulated, since SCRUM is an agile framework whose fundamental principle is continuous improvement. As an agile approach, it does not follow a strictly predefined process, as this would correspond to predictive project methodologies, which is not the focus of this research.

However, specific objectives were established for each SPRINT in order to analyze, evaluate, control, and improve the process, thereby adhering to the core characteristics of SCRUM.

COMMENT 3

3. Methodology

Several issues here. First, the paper conflates SCRUM (a project management tool) with the experimental design. Using sprints to organize work does not make the experiments more rigorous. These two layers need to be separated. Second, key experimental details are missing: no randomization, no replication strategy, no plot sizes, no discussion of confounding variables. Treatment labels change across sprints (M1-M3 vs T1-T6) without explanation. The "alive-alive" and "dead-dead" scenarios are never defined. Statistical tests are not clearly identified, there is no mention of post-hoc tests, and the SPC control limits in Sprint 3 are so wide that having all points within limits is not very meaningful. Also, there is no comparison with a non-SCRUM baseline, so the authors cannot attribute any outcome to SCRUM itself. Finally, Section 4 is titled "Methodology" but presents results, which is a structural error. The authors should consolidate the methodology into one clear section and separate the results properly.

RESPONSE

SCRUM is the agile framework that provides a response to the problem addressed, which was segmented into four defined SPRINTs.

Regarding the experimentation carried out in each SPRINT, it was conducted in the field on cocoa cultivation plots, generating different analyses based on treatments and various combinations in order to analyze and evaluate the best results.

In SPRINT 1 and SPRINT 2, the treatment was applied to samples consisting of 62 cocoa seedlings at the field replanting stage. These received treatments with different amounts of Trichoderma (filamentous fungi used in agriculture as biological control agents and plant growth promoters), which undoubtedly impact the objectives of SPRINTs 1 and 2. Since these were samples with exact quantities, they were designated as “M.”

In contrast, in SPRINT 4, treatments were applied to entire plots (not numbered samples), and therefore they were designated as “T.”

This clarification can be found on page 6.

COMMENT 4

4. Results and Discussion

The results need effect sizes with confidence intervals, not just p-values. The probability plots used in the figures are not standard for showing treatment effects; conventional formats like bar charts or boxplots would work better. In the discussion, the comparisons with literature are too shallow. The authors mostly say "our findings are consistent with" previous work without any quantitative comparison or discussion of where results diverge. The discussion needs much deeper engagement with the literature.

RESPONSE

Normal probability plots were used instead of bar or box plots because they allow for the evaluation of data distribution, the normality assumption, and variability patterns across treatments. This approach provides a more comprehensive statistical interpretation of the experimental behavior, particularly when parametric analyses are applied.

Regarding the results shown in the figures, normal probability plots were selected as the most appropriate representation to display segmented evaluations under different conditions, as established in each SPRINT of the research. Moreover, this type of graph is more useful for verifying data normality, comparing distributions across treatments or samples, and analyzing the statistical behavior of the dataset.

This strengthens the comparison and evaluation process of the best alternative, as it allows for a clearer visualization of performance under differentiated conditions and the quality of the statistical fit, thereby adding greater robustness to the analysis conducted.

In the discussion section, emphasis is placed on the importance of the appropriate use of bio-inputs to generate a positive response pattern in the crop; therefore, the findings are compared with those reported in other studies.

COMMENT 5

5. Practical Implications and Limitations

The paper claims the model is replicable but gives no practical guidance: no cost analysis, no implementation guidelines for farmers, no discussion of adoption barriers or scalability. For a paper in Sustainability, this is a notable gap. Additionally, the paper has no limitations section at all. The authors should acknowledge at minimum the single-site design, the absence of a non-SCRUM control, the short time horizon, and potential confounds from soil and climate variability. Directions for future research should follow from these.

RESPONSE

Regarding this observation, the necessary correction and clarification were made on page 13.

COMMENT 6:

6. References

The manuscript cites only 16 references, which is far too few. A paper like this needs at least 10 to 20 references in the introduction section alone. The authors need to significantly expand their literature review and cite more work in the introduction, state of the art, and discussion sections. A deeper engagement with the literature would help them better position their contribution and identify the actual research gap.

RESPONSE

There could be many references; however, for them to be effective, they must maintain relevance and significance in relation to what is proposed in the research article. For this reason, each reference included in the article was considered important for both the development and the analysis of the study.

It is also important to note that this article presents research in which a methodology is incorporated into a new sector, highlighting its practical applicability and usefulness.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review this interesting paper. It deals with an important issue for a certain group of local cocoa producers in the global environment. The study proposes the implementation of the agile SCRUM methodology, adapted to the Peruvian cocoa agribusiness sector, using a group of farmers as a case study, with the aim of strengthening production processes and contributing to cadmium mitigation under real production conditions. 

The abstract is informative, and the introduction presents the study's framework in sufficient detail. 

The literature review is not comprehensive, but it provides background for the study. The changed circumstances with the limitation of maximum permitted amounts of cadmium in the EU have caused problems, especially for smaller Peruvian farms. The sources are relevant and up-to-date. 

The methodology is appropriate; SCRUM has been used for similar studies. It is properly explained and used in a customized way for this case study. SCRUM methodology is an effective tool for optimizing productive, administrative, educational, and innovation processes. 

The analysis is detailed enough, authors researched four issues: Evaluation of stem diameter and mortality under bio-organic treatments; Evaluation of plant height and mortality using biofertilizers; Fermentation monitoring: temperature and bean quality; and Cadmium reduction through agricultural prototypes and experimental treatments. 

The findings are discussed and duly justified in the next chapter. The conclusions are thoroughly supported by the results presented in the article and those referenced in the secondary literature. 

In the conclusions, authors might add the study limitations and proposals for future research. 

 

Author Response

COMMENT:

In the conclusions, authors might add the study limitations and proposals for future research. 

RESPONSE:

Thank you very much for reviewing the article and for taking the necessary time to provide your comments; they undoubtedly enhance the level and rigor we aim to achieve with this research.

Regarding the limitations and proposals for future research, they have been incorporated into the document. The limitations are presented at the end of the Discussion section, and the proposals for future research have been included in the Conclusions section. These can be found on pages

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

The authors correctly place the study in its current context: the problem of cadmium in cocoa and its regulatory consequences (EU), existing strategies for reducing Cd (agroforestry, microorganisms, genotypes), and the growing use of SCRUM outside of IT. The introduction and state of the art cover both agronomic literature and works on SCRUM in non-IT contexts. This is a significant advantage of the article. However, the section on SCRUM remains at a descriptive level, without in-depth theoretical reflection (no reference to literature on agile in operations management, agile governance, or socio-technical systems). Furthermore, it is not clearly shown what scientific gap the combination of SCRUM + Cd bioremediation fills (beyond a general statement of innovation). The authors should more clearly define the research gap at the intersection of agile management and agronomic experimentation, rather than treating SCRUM solely as an organizational tool.

The methodology is described in detail in operational terms (Sprints 1–4, SCRUM roles, artifacts, biological variables, SPC for fermentation, Cd measurements). However, from the perspective of scientific methodology, there are significant shortcomings: no explicitly formulated research questions, no formal hypotheses, no classic experimental design scheme (number of replicates, randomization, sample size, control of confounding factors). SCRUM is described as a research methodology, although in practice it serves as an experiment management framework, which is not clearly distinguished. The authors declare an “applied and experimental approach,” but do not specify: the type of experiment (field trial? quasi-experiment?), the statistical structure of the sample, or the procedures for validating the results. It is necessary to add a separate subsection: Research Questions and Hypotheses, and to clarify the experimental design in accordance with agronomic research standards.

The discussion is logically organized according to Sprints: vegetative growth, fermentation, Cd reduction, SCRUM methodological contribution. The authors do not overinterpret the results concerning plant mortality, correctly point out the limitations of biofertilization, and compare their own data with the current literature. However, the “Methodological contribution” section is too declarative—it lacks critical reflection on the limitations of SCRUM in field research. SCRUM is presented almost exclusively in a positive light, without discussing implementation costs, scalability, or dependence on team competence.

The results are empirically rich. The section on Sprint 4 (Cd) is particularly strong, showing a clear effect of the MycoUP + MBB combination. However, I would like to point out that the statistical presentation is chaotic in places, there are no summary tables comparing all Sprints, no confidence intervals for Cd are given, and no effect sizes are presented. It would be good to add a summary table summarizing all the main results.

The bibliography includes current sources (2020–2025), including MDPI, Heliyon, Land, Agronomy, and Sustainability. It should be noted that there are duplicates (Wigunanda 2023) and single citations with incomplete data. In addition, it would be worthwhile to supplement the bibliography with classic works on agile outside IT (e.g., Rigby, Denning, Conboy).

The conclusions regarding improved growth, fermentation stabilization, and Cd reduction are well supported by empirical data. However, the conclusion about a “replicable methodological model” is too strong for a single case study without interregional validation.

The manuscript is written in generally understandable academic English, and the main arguments, methodology, and results can be followed without major difficulty However, the text would clearly benefit from professional language polishing by a native or near-native academic editor.

 

 

 

Author Response

COMMENT 1:However, the section on SCRUM remains at a descriptive level, without in-depth theoretical reflection (no reference to literature on agile in operations management, agile governance, or socio-technical systems). Furthermore, it is not clearly shown what scientific gap the combination of SCRUM + Cd bioremediation fills (beyond a general statement of innovation). The authors should more clearly define the research gap at the intersection of agile management and agronomic experimentation, rather than treating SCRUM solely as an organizational tool.

RESPONSE:

Thank you very much for reviewing the article and for taking the necessary time to provide your comments; they undoubtedly enhance the level and rigor we aim to achieve with this research.

The theoretical reflection on SCRUM was revised and strengthened by linking it with other methodological principles, highlighting the importance of its application within the present study. Please refer to the attached document on page 4

COMMENT 2

SCRUM is described as a research methodology, although in practice it serves as an experiment management framework, which is not clearly distinguished. The authors declare an “applied and experimental approach,” but do not specify: the type of experiment (field trial? quasi-experiment?), the statistical structure of the sample, or the procedures for validating the results. It is necessary to add a separate subsection: Research Questions and Hypotheses, and to clarify the experimental design in accordance with agronomic research standards.

RESPONSE

Explicit hypotheses cannot be formulated, as SCRUM is an agile framework whose fundamental principle is continuous improvement. Being agile in nature, it does not follow a strictly predefined linear process, as this would correspond to predictive project approaches, which are not the focus of this research. However, specific objectives were defined for each SPRINT in order to analyze, evaluate, control, and improve, thus adhering to SCRUM’s core characteristics. Additionally, each SPRINT includes not only a defined objective, but also a justification, the artifacts used, the increment produced, and the corresponding rationale.


COMMENT 3:
SCRUM is described as a research methodology, although in practice it serves as an experiment management framework, which is not clearly distinguished. The authors declare an “applied and experimental approach,” but do not specify: the type of experiment (field trial? quasi-experiment?), the statistical structure of the sample, or the procedures for validating the results. It is necessary to add a separate subsection: Research Questions and Hypotheses, and to clarify the experimental design in accordance with agronomic research standards.

RESPONSE:

Regarding the limitations of the research development, these have been included at the end of the Discussion section on page 11. Proposals for future research have also been incorporated at the end of the Conclusions section on page 13.


COMMENT 4:
In addition, it would be worthwhile to supplement the bibliography with classic works on agile outside IT (e.g., Rigby, Denning, Conboy).

RESPONSE: 

There could be many references; however, for them to be effective, they must maintain relevance and significance in relation to what is proposed in the research article. For this reason, each reference included in the article was carefully selected and considered important for both the development and the analysis of the study. It is also important to note that this article presents research in which a methodology is incorporated into a new sector, demonstrating its practical applicability and usefulness.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

the authors have made the necessary changes

Author Response

Dear Reviewer,
We sincerely appreciate your time, expertise, and willingness to review our manuscript. Your comments have undoubtedly provided a valuable contribution to improving this work.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Most of the concerns I raised in the previous round remain unaddressed, and unfortunately I cannot recommend the article for publication in its current form.

A research paper should clearly define the research problem, explain the importance of addressing it, review the state of the art in the literature, and justify the research gap. It should then explain the solutions proposed to address this problem, test those solutions rigorously, and demonstrate through analyzed data whether the proposed solutions were effective. The entire article should convey a coherent narrative. The research design used to test the proposed solutions must be rigorous, and the novelty and contributions of the article must be clearly justified. All of these essential aspects remain largely missing from this article. In the previous round, I provided detailed recommendations for addressing the weaknesses of this paper, but the authors appear to have declined to act on them.

As a final remark, citing only 15 articles across the entire paper is not justifiable. A broader engagement with the literature demonstrates that the authors have thoroughly reviewed prior work and are able to position their contribution within the existing body of knowledge. As it stands, no clearly defined research stream or literature base is discussed in this article to show where it belongs or how it contributes to the field.

Author Response

Dear Reviewer,
Taking into account your comments, the necessary corrections have been made for your further review. We appreciate your time and your willingness to evaluate our work.

Author Response File: Author Response.pdf