Advancing Sustainable Tourism Through Smart Wheelchair Optimization: A Mixed-Integer Linear Programming Framework for Inclusive Travel

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article “Advancing Sustainable Tourism through Smart Wheelchair Optimization: A Mixed-Integer Linear Programming Framework for Inclusive Travel” presents a framework that integrates MILP optimization with smart wheelchair technologies to improve accessible tourism planning in Thailand. However, the work also carries several limitations that deserve attention.

The abstract is too long and functions more as an extended summary than as a concise overview of objectives, methods, and findings.

One of the main limitations lies in the use of a static national accessibility database, which cannot capture dynamic or real-time conditions such as weather disruptions, infrastructure maintenance, or sudden changes in demand. This weakens the applicability of the model in practice, as tourism accessibility is highly variable.

Another limitation is the small scale of empirical validation: only 30 participants in Ayutthaya were surveyed, a number too small to generalize findings to Thailand’s broader tourism landscape or to ASEAN countries.

Ethical issues also are insufficiently addressed, especially regarding the use of smart wheelchairs with health-monitoring functions that raise questions of data privacy and security. Moreover, the MILP model is deterministic and does not incorporate stochastic or machine learning approaches that could better handle uncertainty.

In terms of structure, the article suffers from redundancy and imbalance. The Introduction is overly long and descriptive, with excessive discussion of sustainable tourism principles and SDGs before narrowing down to the specific research problem. This dilutes the focus and makes the contribution appear less distinct.

The Literature Review, although comprehensive, is too broad and sometimes drifts away from the central theme of accessible tourism, particularly in its lengthy technical discussion of smart wheelchair technologies. While the Methodology is innovative, its technical presentation is heavy and difficult for non-specialist readers, especially since the choice of model weights is not explained with sufficient justification.

The Results section provides promising outputs, such as reduced travel time and improved accessibility indices, but the discussion of these results lacks depth. Figures and tables are informative but not always connected to meaningful policy implications.

The Discussion often restates findings instead of critically examining limitations, trade-offs, or the feasibility of large-scale implementation. The tone is overly positive, giving little room for reflection on practical constraints such as funding, infrastructure gaps, or cultural attitudes toward disability.

The broader framing of the study within sustainable development and sdgs relies heavily on well-established discourse, which reduces the perception of novelty.

Comments on the Quality of English Language

The English is understandable but verbose, with awkward phrasing, inconsistent verb tenses, and several grammatical errors. Examples include phrases like “discussion the result,” which indicate that the manuscript would benefit from careful language editing by a native or professional academic editor.

Author Response

1. The abstract is too long and functions more as an extended summary than as a concise overview of objectives, methods, and findings.

Answer

Thank you for your valuable feedback regarding the abstract of our manuscript. We appreciate your observation that the original abstract was overly lengthy and functioned more as an extended summary than a concise overview. In response to your comment, we have revised the abstract to provide a more succinct and focused summary of the study’s objectives, methods, and key findings, while maintaining clarity and academic rigor. (Line 22-40)

 

2. One of the main limitations lies in the use of a static national accessibility database, which cannot capture dynamic or real-time conditions such as weather disruptions, infrastructure maintenance, or sudden changes in demand.

Answer

Thank you for your insightful feedback regarding the limitation of our study’s reliance on a static national accessibility database, which cannot capture dynamic or real-time conditions such as weather disruptions, infrastructure maintenance, or sudden changes in demand. We acknowledge that this is a significant constraint, as the current database, limited by its lack of integration with external systems in Thailand, does not support real-time data updates. To address your concern, we have revised Section 5.3 (Limitations and Future Research) to explicitly discuss the implications of this limitation and propose actionable solutions for future work.

The revised section now emphasizes the potential for integrating real-time data streams, such as IoT-enabled smart wheelchairs or machine learning-based frameworks, which have shown a 12% reduction in travel time through dynamic route adjustments in urban tourism settings. Additionally, we propose incorporating stochastic elements into the MILP framework to address demand uncertainties, drawing on policy-oriented optimization models that achieve 15% inclusive improvements in Mobility-on-Demand services [73]. These revisions clarify how future iterations of the framework could leverage dynamic data to enhance adaptability and robustness under varying conditions. (Line 1018-1071)

 

3. This weakens the applicability of the model in practice, as tourism accessibility is highly variable. Another limitation is the small scale of empirical validation: only 30 participants in Ayutthaya were surveyed, a number too small to generalize findings to Thailand’s broader tourism landscape or to ASEAN countries.

Answer

Thank you for your valuable feedback highlighting two key limitations in our study: the reduced practical applicability of the model due to the static nature of the national accessibility database and the limited generalizability stemming from the small-scale empirical validation involving only 30 participants in Ayutthaya. We greatly appreciate your insightful comments, which have guided our revisions to strengthen the manuscript. To address these concerns, we have revised Section 5.3 (Limitations and Future Research) to provide a more detailed discussion of these limitations and propose actionable solutions.

As this study developed a prototype model, we propose integrating real-time data streams, such as IoT sensor data from smart wheelchairs, and incorporating stochastic modeling to enhance adaptability, drawing on studies that achieved 15% inclusivity improvements in mobility services.

Concerning the small sample size, we recognize that the empirical validation with 30 participants in Ayutthaya limits generalizability to Thailand’s broader tourism landscape or ASEAN countries. This initial study served as a prototype, tested with a small group of tourists in Ayutthaya. To address this, we have outlined plans for future research to scale up the framework, involving larger and statistically robust sample sizes, particularly targeting the growing "Tourism for All" demographic, including elderly travelers and wheelchair users. We propose conducting pilot studies across diverse Thai regions (e.g., Bangkok, Isan, Phuket). Additionally, we plan to collaborate with tourism operators to implement and evaluate the framework in real-world settings, incorporating additional variables such as user satisfaction and infrastructure capacity to enhance practical relevance. (Line 1018-1071)

 

4. Ethical issues also are insufficiently addressed, especially regarding the use of smart wheelchairs with health-monitoring functions that raise questions of data privacy and security.

Answer

Thank you for your valuable feedback highlighting the insufficient discussion of ethical issues, particularly regarding data privacy and security concerns arising from the use of smart wheelchairs with health-monitoring functions. We acknowledge the importance of this concern and have revised the manuscript to address it more comprehensively, recognizing that the smart wheelchair development in this study is currently at the prototype stage, as part of an experimental research phase. To address your comment, we have updated Section 5.4 (Ethical Considerations) to explicitly discuss the ethical implications of collecting and processing sensitive health and location data through smart wheelchairs. We recognize that, as the framework progresses to the next phase of real-world implementation with elderly and disabled users, robust ethical safeguards are essential. Accordingly, we have outlined a commitment to obtaining ethical approval and adhering to international data protection standards, such as the General Data Protection Regulation (GDPR) and Thailand’s Personal Data Protection Act (PDPA). We propose implementing stringent measures, including data encryption, anonymization, and user consent protocols, to protect participant data, drawing on best practices from IoT-based mobility studies [82]. Furthermore, we have strengthened the manuscript by incorporating plans for future research to integrate a comprehensive ethical framework. This includes conducting privacy impact assessments and engaging stakeholders, such as disabled travelers and advocacy groups, to ensure user-centric design and transparency. These revisions enhance the manuscript’s alignment with best practices for responsible innovation and address the ethical dimensions of the technology. (Line 1058-1070)

 

5. Moreover, the MILP model is deterministic and does not incorporate stochastic or machine learning approaches that could better handle uncertainty.

Answer

Thank you for your insightful feedback regarding the deterministic nature of the Mixed-Integer Linear Programming (MILP) model and its lack of stochastic or machine learning approaches to address uncertainty. We acknowledge this limitation and appreciate your suggestion to enhance the model’s ability to handle dynamic and uncertain conditions. To address this concern, we have revised Section 5.3 (Limitations and Future Research) to explicitly discuss the constraints of the deterministic MILP model, which does not account for variability in factors such as tourism demand fluctuations or infrastructure disruptions.

In response, we have proposed future research directions to incorporate stochastic elements into the MILP framework to better manage demand uncertainties, drawing on studies demonstrating a 15% improvement in efficiency for Mobility-on-Demand services. Additionally, we suggest integrating machine learning techniques, such as real-time predictive modeling, which have achieved a 12% reduction in travel time in urban tourism contexts. These enhancements aim to improve the model’s robustness and adaptability to dynamic environments. (Line 1019-1030)

 

6. In terms of structure, the article suffers from redundancy and imbalance. The Introduction is overly long and descriptive, with excessive discussion of sustainable tourism principles and SDGs before narrowing down to the specific research problem.

Answer

Thank you for your valuable feedback regarding the structure of our manuscript, particularly the observation that the Introduction is overly long, descriptive, and contains redundant discussions of sustainable tourism principles and the Sustainable Development Goals (SDGs) before addressing the specific research problem. We appreciate your insight and have made substantial revisions to improve the clarity, conciseness, and balance of the Introduction.

To address your concerns, we have streamlined Section 1 (Introduction) by reducing its length and focusing on the core elements relevant to the research problem. Specifically, we have condensed the discussion of sustainable tourism principles and SDGs, retaining only the most pertinent points that directly contextualize the study’s focus on accessible tourism in Thailand. The revised Introduction now briefly outlines the global and regional significance of inclusive tourism, its alignment with SDGs 3, 8, and 11, and the role of assistive technologies like smart wheelchairs, before swiftly transitioning to the specific research problem of inadequate planning frameworks for mobility-impaired tourists in Thailand. Redundant details, such as the extended discussion of general sustainable tourism principles and overly descriptive examples, have been removed or consolidated to enhance clarity and focus. Additionally, we have restructured the Introduction to ensure a more balanced progression toward the research problem, gap, objectives, and contributions. The revised section now emphasizes the unique challenges faced by elderly and disabled travelers in Thailand, the potential of Mixed-Integer Linear Programming (MILP) and smart wheelchair technologies, and the study’s practical and methodological contributions. These changes aim to make the Introduction more concise, targeted, and aligned with the manuscript’s overall objectives. (Line 45-198)

 

7. This dilutes the focus and makes the contribution appear less distinct. The Literature Review, although comprehensive, is too broad and sometimes drifts away from the central theme of accessible tourism, particularly in its lengthy technical discussion of smart wheelchair technologies.

Answer

Thank you for your valuable feedback highlighting that the overly broad scope of the Literature Review dilutes the focus of the article and makes the research contribution less distinct, particularly due to the lengthy technical discussion of smart wheelchair technologies that sometimes drifts from the central theme of accessible tourism. We acknowledge this concern and have revised the manuscript to enhance its clarity and focus.

To address your comments, we have refined Section 2 (Literature Review) by narrowing its scope to emphasize content directly relevant to accessible tourism and the application of smart wheelchair technologies in Thailand’s tourism context. We have significantly reduced the lengthy technical discussion on smart wheelchairs, retaining only essential details about features pertinent to tourism planning, such as obstacle detection and autonomous navigation, which support the Mixed-Integer Linear Programming (MILP) framework. Additionally, we have restructured the Literature Review to underscore the research gap in accessible tourism planning within the ASEAN region, particularly Thailand, and the need for integrating assistive technologies with optimization methods. These revisions sharpen the article’s focus and highlight the distinct contribution of our study to inclusive and sustainable tourism. We believe these changes make the Literature Review more concise and aligned with the research objectives. (Line 200-507)

 

8. While the Methodology is innovative, its technical presentation is heavy and difficult for non-specialist readers, especially since the choice of model weights is not explained with sufficient justification.

Answer

Thank you for your constructive feedback regarding the technical presentation of the Methodology section and the insufficient justification for the choice of model weights in our Mixed-Integer Linear Programming (MILP) framework. We recognize that the overly technical language may pose challenges for non-specialist readers and that the rationale for model weights requires clearer explanation. We have made targeted revisions to address these concerns and enhance the accessibility and clarity of the manuscript.

To improve the readability of Section 3 (Methodology), we have simplified the technical presentation by reducing jargon and incorporating plain-language explanations of key concepts, such as the MILP framework and its application to accessible tourism planning. For instance, we now include a brief, intuitive description of how the MILP model optimizes travel itineraries by balancing accessibility, cost, and environmental sustainability, making it more approachable for a broader audience, including policymakers and tourism practitioners. Additionally, we have added a concise glossary of technical terms as an appendix to assist non-specialist readers. Regarding the choice of model weights, we have revised Section 3.2 (Model Formulation) to provide a detailed justification for the selection of weights in the MILP objective function. Specifically, we now explain that the weights (e.g., 0.4 for accessibility, 0.3 for cost, 0.2 for environmental impact, and 0.1 for cultural significance) were determined based on stakeholder consultations with tourism operators and disability advocacy groups in Thailand, as well as a pilot survey conducted during the 2024 Thailand Tourism Symposium [48]. We also reference studies on multi-objective optimization in tourism planning [75], which support the prioritization of accessibility and sustainability metrics. Furthermore, we have included a sensitivity analysis in Section 4.3 to demonstrate how variations in these weights impact itinerary outcomes, providing transparency and robustness to the model’s design. (Line 531-640)

To further address these issues in future research, we propose conducting additional stakeholder workshops to refine weight assignments and incorporating machine learning techniques to dynamically adjust weights based on real-time user feedback, as suggested by recent advancements in adaptive optimization. These revisions aim to make the methodology more transparent, justifiable, and accessible while maintaining its innovative rigor. (Line 1048-1070)

 

9. The Results section provides promising outputs, such as reduced travel time and improved accessibility indices, but the discussion of these results lacks depth. Figures and tables are informative but not always connected to meaningful policy implications.

Answer

Thank you for your insightful feedback regarding the Results section, specifically noting that while the outputs, such as reduced travel time and improved accessibility indices, are promising, the discussion lacks depth and the figures and tables are not consistently linked to meaningful policy implications. We greatly appreciate your comments and have made significant revisions to Section 4 (Results) and Section 5 (Discussion) to address these concerns, enhancing the depth of the analysis and strengthening the connection to policy implications.

To address the lack of depth in the discussion of results, we have expanded Section 4 (Results) and Section 5.1 (Discussion of Results) to provide a more comprehensive analysis of the key findings, including the 15–20% reduction in travel time and the 25% improvement in accessibility scores achieved by the Mixed-Integer Linear Programming (MILP) framework. Specifically, we now elaborate on the practical significance of these outcomes for various stakeholders, such as tourism operators, policymakers, and mobility-impaired travelers. For instance, in Section 4.2 (Optimal Travel Itineraries), we clarify how the inclusion of secondary attractions with high accessibility scores (e.g., smaller temples in Ayutthaya) reduces overcrowding at major sites like the Grand Palace, aligning with Thailand’s Green Tourism Initiative 2030 [41]. In Section 5.1, we further discuss how these results support cost savings for operators through optimized routing and enhance user satisfaction and safety. We have revised their descriptions and integrated them into a new subsection, Section 4.4 (Policy Implications), which explicitly links the results to actionable policy recommendations. For example, Figure 7 (Simulated Accessibility Network) now includes an expanded caption and discussion in Section 4.1, highlighting how structural fragmentation in Thailand’s accessibility infrastructure (e.g., poor connectivity between transport hubs and cultural sites) underscores the need for targeted investments in multimodal transport systems, particularly in secondary cities like Chiang Mai and Ayutthaya. Similarly, Table 1 (Accessible Tourism Demand Scenarios) and Figure 8 (Demand Scenarios, 2030–2050) are now accompanied by a detailed explanation in Section 4.3, connecting the projected increase in the Accessibility Index (from 0.45 in 2030 to 0.85 in 2050) to policy measures such as incentivizing low-emission transport and prioritizing accessibility upgrades at cultural and natural attractions. These measures align with Thailand’s 2025 “Amazing Thailand Grand Tourism and Sports Year” campaign [8]. Additionally, Figures 10–12 (Sensitivity Analysis) in Section 4.4 now explicitly tie the trade-offs between accessibility, cost, and demand to recommendations for policymakers, such as raising minimum accessibility thresholds (A_min) to improve inclusivity while investing in infrastructure to mitigate itinerary limitations under high-demand scenarios. (Line 789-912)

In Section 5.2 (Implications), we have further strengthened the policy relevance by outlining how the MILP framework and user evaluation results can inform national and regional tourism strategies. For instance, we emphasize the potential of the national accessibility database to support community-based tourism initiatives and the scalability of the framework for ASEAN countries, contributing to SDGs 3, 8, and 11. The user evaluation findings, particularly the high satisfaction with GPS connectivity and individualized programs , are now linked to recommendations for wheelchair manufacturers to develop lighter, cost-effective designs and for tourism operators to integrate smart wheelchairs into accessible tour packages. (Line 988-1017)

 

10. The Discussion often restates findings instead of critically examining limitations, trade-offs, or the feasibility of large-scale implementation. The tone is overly positive, giving little room for reflection on practical constraints such as funding, infrastructure gaps, or cultural attitudes toward disability.

Answer

Thank you for your insightful feedback regarding the Discussion section, particularly the observation that it often restates findings rather than critically examining limitations, trade-offs, or the feasibility of large-scale implementation, and that its overly positive tone leaves limited room for reflection on practical constraints such as funding, infrastructure gaps, or cultural attitudes toward disability. We greatly appreciate your comments and have revised the manuscript to address these concerns, ensuring a more balanced, critical, and reflective discussion.

To address the issue of restating findings, we have restructured Section 5 (Discussion) to focus on a deeper critical analysis of the results, limitations, and implementation challenges. Specifically, in Section 5.1 (Discussion of Results), we have reduced repetitive summaries of findings, such as the 15–20% reduction in travel time and 25% improvement in accessibility scores, and instead emphasize their implications, trade-offs, and challenges. For example, we now critically examine the trade-offs highlighted in the sensitivity analysis (Section 4.4), such as how stricter accessibility thresholds (A_min) improve inclusivity (Accessibility Index rising from 0.65 to 0.80) but reduce itinerary diversity by excluding less accessible sites, which may limit cultural engagement for some travelers. We also discuss the environmental trade-offs in the High Demand scenario, where increased travel distances for global medical tourists lead to a 10–15% rise in carbon emissions, highlighting the need for sustainable infrastructure investments to balance inclusivity and environmental goals. (Line 950-986)

To address practical constraints, we have expanded Section 5.3 (Limitations and Future Research) to provide a more reflective discussion on the feasibility of large-scale implementation. We now explicitly address funding challenges, noting that the development and maintenance of a national accessibility database and infrastructure upgrades require significant public and private investment. We also discuss infrastructure gaps, particularly in secondary cities like Chiang Mai and Ayutthaya, where fragmented transport networks and limited accessibility at cultural sites (e.g., Doi Suthep) pose ongoing barriers. Additionally, we have added a new discussion on cultural attitudes toward disability, acknowledging that societal stigma and lack of awareness in Thailand can hinder the adoption of inclusive tourism practices, as noted in community-based tourism studies. We propose stakeholder engagement with disability advocacy groups and public awareness campaigns to address these cultural barriers, aligning with Thailand’s “Tourism for All” initiatives. To temper the overly positive tone, we have revised the language throughout Section 5 to adopt a more balanced perspective, explicitly acknowledging uncertainties and challenges. For instance, we note that while the MILP framework and smart wheelchair integration show promise, their scalability depends on overcoming financial, logistical, and cultural barriers, and we highlight the need for pilot studies across diverse regions (e.g., urban Bangkok, rural Isan) to validate feasibility. We also propose future research to incorporate dynamic data sources, such as IoT-enabled sensors, to address the static nature of the current database, and to explore cost-sharing models to make smart wheelchairs more affordable, drawing on market forecasts for assistive technologies. (Line 1020-1071)

 

11. The broader framing of the study within sustainable development and sdgs relies heavily on well established discourse, which reduces the perception of novelty.

Answer

Thank you for your valuable feedback regarding the broader framing of the study within the context of sustainable development and the Sustainable Development Goals (SDGs).

To address this concern, we have revised key sections, including the Introduction (Section 1), Discussion (Section 5), and Conclusion, to better emphasize the novel aspects of the study while maintaining its alignment with sustainable development and SDGs. Specifically, we have reduced the reliance on general SDG discourse by focusing on the study’s innovative integration of Mixed-Integer Linear Programming (MILP) with smart wheelchair technologies and a national accessibility database, which represents a pioneering approach in the context of accessible tourism in Thailand and Southeast Asia. In Section 1 (Introduction), we now highlight the originality of combining data-driven optimization with assistive technologies to address specific accessibility challenges in Thailand’s tourism landscape, such as the fragmented infrastructure in secondary cities like Chiang Mai and Ayutthaya. This positions the study as a novel contribution to the field, distinct from broader sustainable tourism frameworks. (Line 950-969)

In Section 5.2 (Implications), we have further clarified the study’s unique contributions by emphasizing how the MILP framework’s ability to balance accessibility, cost, environmental impact, and cultural preservation through real-time data from smart wheelchairs fills a critical research gap. We now underscore that, unlike existing studies that focus on general accessibility barriers or broad SDG alignments [8, 16], this research offers a scalable, technology-driven solution tailored to mobility-impaired tourists, with potential applications across ASEAN countries. Additionally, we have included a discussion in Section 5.1 (Discussion of Results) that contrasts our approach with conventional tourism planning models, highlighting the novelty of incorporating secondary attractions with high accessibility scores and low environmental impact (e.g., smaller temples in Ayutthaya) to redistribute tourist flows, as supported by sensitivity analyses [Section 4.4].To further enhance the perception of novelty, we have tempered the use of generalized sustainable development rhetoric throughout the manuscript, focusing instead on specific, actionable outcomes. We also propose future research in Section 5.3 to explore innovative extensions, such as integrating machine learning with MILP to adapt itineraries dynamically, further distinguishing the study from established frameworks. (Line 988-1070)

Reviewer 2 Report

Comments and Suggestions for Authors

1. Consider tightening the Abstract section by focusing on the most novel points: for example, emphasize the sustainability and inclusivity outcomes more succinctly and reduce less critical detail (e.g. some survey statistics) to improve readability.

2.  In Abstract section, It mentions improvements over “conventional itineraries” – please clarify what baseline is meant by “conventional itineraries”.

3. One suggestion for Introduction section is to emphasize the environmental dimension of sustainable tourism a bit more explicitly. The text rightly focuses on social equity and economic growth, but since the MILP includes “cultural–environmental” factors, it might help to preview why environmental considerations matter for tour planning.

4. For Introduction section, since sustainability is a core theme, consider adding an explicit mention (in either objectives or contributions) of how the model addresses environmental or cultural preservation goals, not just cost and accessibility.

5. One improvement for the Review Literature section could be to conclude the review with a brief paragraph summarizing the gaps identified and how your study fills them, which would transition nicely into the Methods. This synthesis is partly there, but a concise closing statement would highlight novelty.

6. One suggestion is to provide more intuition on how the MILP incorporates sustainability: specifically, explain how the environmental/cultural term (e_k) is quantified in practice. The objective function (Equation 1) includes α, β, γ weights but doesn’t define e_k in the text. A brief description (even qualitative) of what e_k represents (e.g. “number of natural attractions visited” or “penalty for non-green routes”) would help readers grasp the model.

7. The methodological framework is well structured. The conceptual figure (Fig. 3) helps; ensure its caption is descriptive. Overall, the four-step description is clear. One suggestion is to provide more intuition on how the MILP incorporates sustainability: specifically, explain how the environmental/cultural term (e_k) is quantified in practice. The objective function (Equation 1) includes α, β, γ weights but doesn’t define e_k in the text. A brief description (even qualitative) of what e_k represents (e.g. “number of natural attractions visited” or “penalty for non-green routes”) would help readers grasp the model.

8. The connectivity constraints (each node has one ingoing/outgoing) imply a tour visiting all nodes. If the intent is a full circuit through all sites, mention that. If it is a subset, note how the model chooses nodes (e.g. budgets might force skipping nodes). Also, consider referencing how you handle subtour elimination (the brief mention of “flow conservation” could be expanded; a citation [79] is given but it could be better explained for completeness).

9. In Data Collection (3.4), it’s excellent that you built a multi-source database. It would improve rigor to explain how the accessibility scores (a_ij) were derived from the survey and GIS data. For example, did you use a scoring rubric or expert ratings? Citing a methodology or standard (if any) for these scores would add credibility. Also mention the scale of the database (number of nodes, edges, cities covered) to give an idea of scope.

10. The case study description is clear. One suggestion: briefly mention how travel times and networks were modeled (e.g. road network travel time from GIS?), since the MILP needs those inputs. If travel times are assumed or estimated, a sentence clarifying the source or calculation method would be useful.

11. In 4.2, the optimized itinerary results are promising. The equation and constraints are restated from methods; to avoid repetition, you might refer back to the model description rather than reprinting them. Focus instead on interpretation of the outcome: for example, explain why secondary attractions became included (perhaps they have high accessibility scores, which the model valued).

12. Scenario analysis (4.3) is well-illustrated. When introducing Fig. 8 and Table 1, ensure the text describes how the low/medium/high scenarios were defined (e.g. Table 1 has numeric demand values – clarify if “1.6” means 1.6 million travelers or some index). Readers should understand the scenario assumptions. The narrative does a good job summarizing each scenario’s outcome. You might strengthen it by commenting on sustainability: for instance, note that the high-demand scenario puts strain on infrastructure and suggest any environmental implications (e.g. greater emissions from the longer routes).

13. Sensitivity analysis (4.4) effectively shows trade-offs. The explanation of Fig. 10–12 is clear. One improvement: define any terms (e.g. “normalized cost units”) and ensure the magnitude of changes is meaningful (is +10% cost a realistic variation?). The text describes trends well, but consider adding a sentence on how these insights could guide planners (e.g. “This suggests policymakers might prioritize reducing barriers (raising A_min) to improve average accessibility, even if that excludes some sites.”).

14. The discussion does a good job relating your findings to existing studies. It is strong in connecting the optimization gains to similar multi-objective tourism planning work. To deepen the sustainability perspective, you might explicitly mention environmental or cultural findings here. For example, if your model included any measure of cultural site preservation (e_k), discuss what the results imply for that goal. Even if you did not quantify emissions or environmental impact, you could note that “by potentially reducing overall travel distances, the optimized itineraries may also yield indirect environmental benefits, an area for future quantification.”

15. The implications section (5.2) is comprehensive. You tie results back to Thai policy (e.g. infrastructure in secondary cities) and SDGs, which is excellent. One suggestion is to also briefly address how tourism operators or wheelchair manufacturers could use these insights (e.g. promoting accessible tour packages or improving wheelchair cost-effectiveness), to broaden the practical reach.

16. The implication to SDGs (3, 8, 11) is reiterated, which is fitting. You might compress some repetitive parts (the SDGs and numbers were mentioned several times in abstract, discussion, and conclusion). For example, the final paragraphs could focus more on high-level lessons learned and less on re-stating percentages already given.

Author Response

Find the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

I have read the manuscript with great interest and appreciation. The paper is  well-structured, and conceptually coherent. It demonstrates a high level of scholarly diligence, offering a robust integration of quantitative optimization techniques with sustainability and accessibility objectives. The argumentation is logically developed, the empirical components are clearly articulated, and the work is thoroughly grounded in relevant literature. The manuscript provides valuable and original insights into the intersection of accessible tourism, optimization modeling, and sustainable development, thereby making a significant contribution to both theoretical understanding and practical policy implications.

Nevertheless, I would like to raise a minor but important concern regarding the figures included in the manuscript. Specifically, it is not entirely clear who created the graphical materials presented as Figures, nor whether the authors hold the rights to reproduce and publish them. For full compliance with copyright and academic transparency standards, the source or authorship of each figure should be explicitly indicated in the captions - clarifying whether the graphics were produced by the authors, adapted from other works, or reproduced with permission.

Apart from this issue, the article is of excellent quality, highly informative, and a pleasure to read.

Author Response

We sincerely express our deepest gratitude for your thoughtful and comprehensive review of our manuscript. Your kind remarks regarding the structure, coherence, and scholarly rigor of the paper, as well as its contributions to the fields of accessible tourism, optimization modeling, and sustainable development, are greatly appreciated. We are truly honored by your recognition of the manuscript’s originality and its potential to inform both theoretical and practical advancements in the field.

We have carefully considered your valuable feedback, particularly your concern regarding the clarity of authorship and copyright status of the graphical materials included in the manuscript. To address this, we wish to clarify that all figures presented in the paper are original works created by the authors as part of our research. These graphics are wholly owned by us, and we confirm that there are no copyright restrictions associated with their use or reproduction. To ensure full compliance with academic transparency standards, we have revised the captions for all figures to explicitly indicate that they were produced by the authors. Additionally, in response to your comment, we have replaced Figure 5 with an updated version that depicts the practical testing of a smart wheelchair prototype in real-world settings designed to support accessible tourism. These tests were conducted in Bangkok and Ayutthaya Province, aligning with the research objectives of evaluating accessibility infrastructure. The updated figure further strengthens the empirical grounding of our work and provides clearer visual representation of the prototype in action.

We believe these revisions address your concerns comprehensively and enhance the manuscript’s clarity and adherence to academic standards. Once again, we are immensely grateful for your insightful and constructive feedback, which has undoubtedly strengthened our work. Please do not hesitate to provide further guidance if additional refinements are needed.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The revised manuscript shows clear improvement, but several limitations remain.
The model still relies on a static national accessibility database, limiting real-time applicability.
Empirical validation is based on only 30 participants in one city, reducing generalizability.
The environmental impact and ethical framework are conceptually discussed but not yet implemented or empirically tested.
Overall, the study should be viewed as a prototype or exploratory contribution.

Comments on the Quality of English Language

The English is understandable but verbose, with awkward phrasing, inconsistent verb tenses, and several grammatical errors. Examples include phrases like “discussion the result,” which indicate that the manuscript would benefit from careful language editing by a native or professional academic editor.

Author Response

Dear Reviewer1,

We sincerely express our gratitude for your insightful and valuable feedback on our manuscript, titled “Advancing Sustainable Tourism through Smart Wheelchair Optimization: A Mixed-Integer Linear Programming Framework for Inclusive Travel” (sustainability-3911585-Rev 15-10-2025). Your comments have significantly contributed to enhancing the quality and clarity of our work, and we greatly appreciate the opportunity to address the identified limitations.

In response to your feedback, we have carefully revised the manuscript to address the concerns regarding the static national accessibility database, the limited empirical validation, and the conceptual discussion of environmental and ethical frameworks. Below, we outline the specific revisions made to address each point:

1. Static National Accessibility Database Limiting Real-Time Applicability: We acknowledge the limitation of relying on a static database managed by the Tourism Authority of Thailand, which requires verification for accuracy. To address this, we have introduced a temporary database solution in Section 3 (Methodology, Subsection 3.2, lines 535-542), explaining how the AI platform downloads data for real-time processing to optimize personalized travel routes. Additionally, Section 5 (Discussion, lines 1064-1070) and Section 6 (Conclusion, lines 1176-1181) now elaborate on this approach and propose future integration of IoT-enabled smart wheelchair sensors for dynamic updates, enhancing real-time applicability.
2. Limited Empirical Validation with 30 Participants in One City: To address the concern about generalizability, we clarified in Section 4 (Results, lines 933-937) that the prototype testing with 30 participants in Ayutthaya focused on assessing system stability. In Section 5 (Discussion, lines 1001-1007), we added a detailed discussion on the sample size limitation and outlined plans for larger-scale studies across diverse regions (Bangkok, Isan, Phuket) with over 100 participants to achieve statistical significance. Section 6 (Conclusion, lines 1111-1123) further emphasizes the exploratory nature of the study and the need for broader validation.
3. Environmental Impact and Ethical Framework Not Implemented or Empirically Tested: We have addressed this by incorporating preliminary environmental impact estimates in Section 3 (Subsection 3.3, lines 654-659), noting a 10% reduction in carbon emissions using electric-powered Smart Wheelchairs, with plans for empirical testing. A proposed ethical framework for data privacy, aligned with ISO/IEC 27001 and GDPR principles, was added to Section 5 (lines 1051-1060), including measures like encryption and user consent mechanisms. Both sections outline future empirical validation plans. Section 6 (lines 1176-1181) acknowledges the conceptual nature of these aspects and commits to further testing.
4. Study as a Prototype or Exploratory Contribution: We have explicitly framed the study as a prototype in the Abstract (lines 39-42) and Section 6 (Conclusion, lines 1176-1181), emphasizing its exploratory contributions and outlining future research directions to address current limitations, such as real-time data integration, scalability, and empirical testing of environmental and ethical aspects.

All revisions are clearly marked in yellow with red text in the revised manuscript for your convenience. Additionally, we updated the Institutional Review Board Statement (lines 1200-1201) to note that ethical approval for expanded studies has been sought, ensuring compliance for future research phases. A new subsection (5.3 Future Research Directions, lines 1111-1123) consolidates plans to enhance the framework’s adaptability, generalizability, and sustainability. We believe these revisions strengthen the manuscript by addressing the reviewer’s concerns while reinforcing its contribution to inclusive and sustainable tourism. We are deeply grateful for your constructive feedback, which has guided these improvements, and we hope the revised manuscript meets your expectations for publication.

Thank you for your time and consideration.