Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Generative AI has already revolutionized the world of graphic design, images as well as text. The next step for this powerful technology is to conquer the realm of 3D animation and in some cases, it has already managed to output good quality, but it does require a lot of input and tinkering since the solutions are still within their early stages.

While Generative AI still struggles to define accurate 3D models, there are various tools that have improved within the last few months and they will redefine 3D modeling, some of them have already implemented 3D animation with specific character rigging tools that are more then decent and usable within production.

The proposed research paper is aimed to study based on a quantitative approach to impact the AI effect on animation production, mainly on the use of network data envelopment analysis to evaluate efficiency across different production stages in a novel contribution.

The introduction of the paper is well defined and includes a good number of published papers and studies associated with AI-generated content (AIGC). The research questions are clear and focused on the research direction of the proposed work.

The literature review incorporates a high number of references that discuss various aspects regarding 3D animation production.

Within the paper, the authors have employed the use of a mixed-method approach to evaluate the impact of AI on 3D animation production. They have conducted qualitative research through expert interviews to identify the key indicators of AI`s influence regarding the production time and AI usage hours. Subsequently, Network Data Envelopment Analysis (NDEA) was used to quantitatively assess efficiency improvements across different production stages.

The authors have proposed a four-stage framework for 3D animation production. This has been aligned with industry standards and modern production practices. Each stage has distinct inputs, outputs as well as workflows. 

I consider that the proposed framework is well defined to provide a detailed analysis of AI's impact on each stage, considering both traditional workflows and contemporary AI integration patterns.

The results presented by the authors highlight that there are significant variations within the overall efficiency scores across different types of 3D animation projects. Commercial projects generally exhibited higher efficiency, while educational and entertainment projects faced greater challenges. The efficiency distribution revealed distinct patterns based on project type, scale, and complexity.

Within the discussion section, the authors have provided a good overview of AI tools impact within the 3D animation production. The discussions effectively address the research questions and hypotheses, highlighting the key findings and their implications for the field of 3D animation production.

 

The conclusions are based on the research findings and the authors have effectively analyzed the data collected within the interviews and NDEA. I am sure that within the near future all 3D animation productions will be heavily influenced by AI tools, even if there are various ethical considerations as well as economic implications associated with the use of these tools.

 

Comments for author File: Comments.pdf

Author Response

Thank you very much for taking the time to review this manuscript! We are grateful for the reviewer's thorough evaluation and supportive comments regarding our research questions, literature review, methodology, and conclusions. Your feedback has helped improve the manuscript's clarity and presentation.

Comments 1: The topic of the paper is not original, as many researchers investigate various aspects associated with the disruptive AI tools used within all creative fields, including animation production. But the topic is very relevant to the field as there are many gaps within this research field considering that AI tools are just starting to be widely adopted within animation production as they are an emerging technology.

Response 1: We sincerely appreciate the reviewer's insightful observation. While we acknowledge that AI in creative fields is an active research area, we agree that the rapid evolution and recent widespread adoption of AI tools in animation production creates vital research opportunities. As noted in our revised introduction, this timing makes our empirical investigation particularly valuable for understanding AI's practical impact on production efficiency.

Comments 2: Within Figure 2, I would suggest the authors increase the font size within the Figures to be the same as the text within the body of the paper to ensure that the text is readable.

Response 2: We thank the reviewer for this constructive suggestion regarding Figure 2's readability. We have optimized the font sizes in Figure 2 to enhance overall readability and ensure all text elements are clearly legible in the final PDF version.

Reviewer 2 Report

Comments and Suggestions for Authors

It is recommended to highlight the innovative aspects of the study.

Even though DEA is applied for the analysis, it is better to include some clarification for the methodology of the choce of the weights assigned to each stage of the production.

Although the ideal use of AI is proposed to be between 30-70% but it is not explained why this range is optimal or how it is influenced by various factors.

The figures and tables would be more informative when they are properly explained.

Minor: Although the impact of AI is analyzed, the differences between the stages are not supported by clear qualitative analysis, which limits the understanding of its actual effect.

Author Response

We are grateful for the reviewer's thorough comments which have helped us significantly improve the manuscript's clarity and depth.

Comments 1: It is recommended to highlight the innovative aspects of the study.

Response 1: We sincerely appreciate this valuable suggestion. We have substantially enhanced the introduction section to explicitly highlight our research contributions. Specifically, we have added three key innovative aspects: (1) Development of a comprehensive four-stage evaluation framework that captures complex interactions between AI technology and traditional animation workflows (2) Implementation of a modified NDEA methodology specifically adapted for creative production processes (3) Empirical validation of optimal AI usage ranges across different production stages, providing concrete guidelines for industry implementation

Comments 2: Even though DEA is applied for the analysis, it is better to include some clarification for the methodology of the choice of the weights assigned to each stage of the production.

Response 2: We thank the reviewer for this important observation. We have added detailed explanations about our weight assignment methodology in Section 3.3. The new content clarifies that our resource-based weighting method determines stage weights based on the proportion of total resources consumed in each stage. This approach ensures stages consuming more resources have proportionally larger impact on overall efficiency evaluation, aligning with practical significance in animation production.

Comments 3: Although the ideal use of AI is proposed to be between 30-70% but it is not explained why this range is optimal or how it is influenced by various factors.

Response 3: We appreciate this crucial point. We have expanded our discussion to provide comprehensive explanation of the optimal AI usage range. The analysis now presents our empirical findings in Section 4.3, showing that projects with highest efficiency scores demonstrated AI usage ratios between 30-70%. Section 5.3 further examines how this "sweet spot" balances AI enhancement with human creativity, and discusses various factors influencing AI integration effectiveness, including project characteristics, organizational capabilities, and standardization levels.

Comments 4: The figures and tables would be more informative when they are properly explained.

Response 4: Following this suggestion, we have enhanced the explanations for all figures and tables throughout the manuscript. For example, we have expanded the description of Figure 2 to clearly explain each component of our four-stage framework and its relationships. Similarly, we have improved the interpretation of the efficiency patterns shown in Figure 5's heat map to better convey the significance of our findings.

Comments 5: Minor: Although the impact of AI is analyzed, the differences between the stages are not supported by clear qualitative analysis, which limits the understanding of its actual effect.

Response 5: We thank the reviewer for this observation. We have strengthened the qualitative analysis of inter-stage differences in Section 5.3. The enhanced discussion now provides detailed examination of why efficiency patterns vary across stages, particularly exploring how the nature of tasks (creative vs. standardized) and current AI capabilities influence efficiency outcomes in different production phases.

Reviewer 3 Report

Comments and Suggestions for Authors

This study explores the impact of artificial intelligence (AI) on the efficiency of 3D animation production through Network Data Envelopment Analysis (NDEA).  The findings indicate that AI's influence on post-production the high efficiency which proved most effective in standardized processes rather than in creative tasks. These insights suggest that organizations should start with standardized processes to maximize efficiency. This research contributes to both the theoretical understanding and practical application of AI in creative production. Overall, the submission is written well and readable. Followings are some comments on the improvements.

1. There are only two mostly related works [51-52]. Please add 2 more references to cover the broader papers discussing efficiency analysis. That is, other methods not using Network DEA.

2. In Sec. 4.3, the used AI powered 3D animation tools for each stage should be mentioned.

3.  Figure 4 legend is misplaced.  

Author Response

We are grateful for the reviewer's constructive comments which have helped enhance the comprehensiveness and clarity of our manuscript.

Comments 1: There are only two mostly related works [51-52]. Please add 2 more references to cover the broader papers discussing efficiency analysis. That is, other methods not using Network DEA.

Response 1: We thank the reviewer for this valuable suggestion. We have expanded our literature review in Section 2.2 to include additional efficiency analysis methodologies. Specifically, we have added Yu (2018), which examines creative industry efficiency using traditional DEA approaches, and Li et al. (2022), who combine Three-stage DEA with Stochastic Frontier Analysis (SFA) to evaluate efficiency in cultural and creative industries. These additions provide a more comprehensive view of various efficiency analysis methods and strengthen our methodological framework.

Comments 2: In Sec. 4.3, the used AI powered 3D animation tools for each stage should be mentioned.

Response 2: Following this helpful suggestion, we have significantly enhanced Section 4.3 with detailed descriptions of AI tools used in each production stage. We now provide comprehensive information about specific tools such as Stable Diffusion and Midjourney for pre-production concept generation, Tripo and Meshy for asset creation, DeepMotion's MotionGPT for animation production, and various AI-enhanced rendering tools for post-production. The section now offers readers a clear understanding of how AI tools are practically implemented across different production stages.

Comments 3: Figure 4 legend is misplaced.

Response 3: We appreciate the reviewer's attention to detail. We have corrected the placement of the legend in Figure 4 to improve clarity and readability of the visualization.

Reviewer 4 Report

Comments and Suggestions for Authors

The authors should submit a more polished version.

The manuscript fails to clearly articulate a well-defined scientific question.

The author's viewpoints lack depth and critical analysis. For example, in discussing the impact of AI on production efficiency, the analysis is rather superficial. There is a lack of in-depth exploration of the underlying mechanisms and factors contributing to the observed efficiencies or inefficiencies

The figures and tables in the manuscript are of low quality. Figure 4, for instance, lacks proper labeling and a clear legend, making it challenging to interpret the data presented. The heat map in Figure 5 is not visually appealing and does not effectively convey the information about efficiency patterns across production stages.

The logical flow of the manuscript is still unclear.

Author Response

We are particularly grateful for the reviewer's incisive and professional critique. Your deep understanding of research methodology and insightful observations have guided us in making fundamental improvements to the manuscript. The thorough review comments have not only helped enhance the paper's scientific rigor and analytical depth, but also led to substantial improvements in its theoretical foundation and practical implications. The manuscript has become significantly more robust and comprehensive thanks to the reviewer's valuable expertise and constructive suggestions.

Comments 1: The authors should submit a more polished version. The manuscript fails to clearly articulate a well-defined scientific question.

Response 1: We sincerely appreciate this crucial observation. We have substantially revised our manuscript to articulate our scientific contributions more clearly. In the introduction section, we have strengthened the discussion of our research significance and theoretical framework, particularly emphasizing how our study advances understanding of AI integration in creative production processes through empirical analysis and methodological innovation. The enhanced introduction better establishes the theoretical foundation for our existing research questions and highlights the practical value of our findings.

Comments 2: The author's viewpoints lack depth and critical analysis. For example, in discussing the impact of AI on production efficiency, the analysis is rather superficial. There is a lack of in-depth exploration of the underlying mechanisms and factors contributing to the observed efficiencies or inefficiencies.

Response 2: We thank the reviewer for this important feedback. We have significantly enhanced our analysis depth throughout the manuscript, particularly in the discussion section. Our enhanced analysis now:

• Examines how organizational context and capabilities influence AI integration success
• Explores the relationship between standardization and efficiency across different project types
• Investigates why efficiency patterns vary across production stages, considering task nature and current AI capabilities
• Analyzes how project characteristics and resource allocation strategies affect efficiency outcomes

Comments 3: The figures and tables in the manuscript are of low quality. Figure 4, for instance, lacks proper labeling and a clear legend, making it challenging to interpret the data presented. The heat map in Figure 5 is not visually appealing and does not effectively convey the information about efficiency patterns across production stages.

Response 3: We appreciate these specific concerns about visualization quality. We have thoroughly revised all figures:

• Improved Figure 4's layout with clear labels, proper legend placement, and enhanced readability
• Redesigned Figure 5's heat map with optimized color scheme and clearer visual hierarchy
• Added comprehensive captions explaining key patterns and relationships
• Enhanced all figures' resolution and visual clarity
• Standardized formatting across all visualizations

Comments 4: The logical flow of the manuscript is still unclear.

Response 4: Following this valuable observation, we have restructured the manuscript to ensure clearer logical progression:

• Strengthened connections between research questions and methodology
• Enhanced transitions between sections to better guide readers
• Improved alignment between theoretical framework and empirical analysis
• Added clear linkages between findings and conclusions
• Restructured the discussion section to build arguments progressively

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Your responses are well-structured and thoroughly explained. The revisions you made significantly enhance the quality of your manuscript.

Reviewer 4 Report

Comments and Suggestions for Authors

Current version is ok.