Towards Resilient Re-Routing Procedures in Ports: Combining Sociotechnical Systems and STAMP
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThank you for the opportunity to read and evaluate an interesting and timely article. Nevertheless, there are places that need to be corrected:
1. Punctuation marks are needed in keywords;
2. It is recommended not to create subsections in the introduction. The introduction should clearly show what has been done by other scientists, what the gap is, and at the same time highlight the purpose, objectives, structure, etc. of your article. What is currently in the introduction as subsections could be moved to the literature analysis section.
3. Citations must meet the methodological requirements of the journal.
4. It is recommended not to create small subsections, i.e. subsections smaller than 1 page.
5. The title of subsection 1.3 must be clarified/concretized.
6. In the methodology section, it is recommended to first present what method you will apply and why, and describe it, and then move on to the case study.
7. It is not recommended to start/end the section/subsection with bullet points/figures/tables. Summary, critical insights of the authors of the article, and a logical transition to the next subsection are required.
8. The analysis of results and discussion sections should contain more critical insights of the authors of the article.
9. The limitations of the study should be presented in the methodology or conclusions section.
10. The conclusions should be presented as a separate part of the article, and not as subsections 4.4. The conclusions should also be corrected, because they currently contain a statement of facts about what was done, and not the specific results themselves.
11. It is recommended to expand the list of references.
Author Response
Comment 1: Punctuation marks are needed in keywords;
Response 1: Punctuation added between keywords (line 28-29).
Comment 2: It is recommended not to create subsections in the introduction. The introduction should clearly show what has been done by other scientists, what the gap is, and at the same time highlight the purpose, objectives, structure, etc. of your article. What is currently in the introduction as subsections could be moved to the literature analysis section.
Response 2: As recommended, subsections in the introduction have been removed and the introduction has been re-structured as: context, existing work, research gap, aim and structure (lines 39-59). The literature analysis has been extended and key references added (lines 62-194).
Comment 3:Citations must meet the methodological requirements of the journal.
Response 3: We believe citations now meet journal requirements: [1] instead of 1
Comment 4:It is recommended not to create small subsections, i.e. subsections smaller than 1 page.
Response 4:We have removed the smallest subsections in introductory material and discussion
Comment 5:The title of subsection 1.3 must be clarified/concretized.
Response 5:As this was a small subsection it has been assimilated into a larger section so the title has been removed (see lines 181-189).
Comment 6:In the methodology section, it is recommended to first present what method you will apply and why, and describe it, and then move on to the case study.
Response 6:The Method now starts with a description of what method will be applied and why (see 191-205). We appreciate why the recommendation for case study after methods is made, but the reader needs to first understand the system under study in order to understand what we did.
Comment 7:It is not recommended to start/end the section/subsection with bullet points/figures/tables. Summary, critical insights of the authors of the article, and a logical transition to the next subsection are required.
Response 7:We have addressed this all relevant places and can see that the article has much improved as a result - Thank you for this useful comment. See lines 348-357, 418-426, 440-444, 450-457.
Commetn 8: The analysis of results and discussion sections should contain more critical insights of the authors of the article.
Response 8: See comment 7. Also insight has been added to the Discussion, following the table, lines 467-484.
Comment 9. The limitations of the study should be presented in the methodology or conclusions section.
Response 9: The limitations of the study now under Conclusions, lines 546-558. Some methodological points are also in the discussion where they fit better, lines 518-532.
Comment 10: The conclusions should be presented as a separate part of the article, and not as subsections 4.4. The conclusions should also be corrected, because they currently contain a statement of facts about what was done, and not the specific results themselves.
Response 10: Conclusions are now a main section, and we have removed statements of what was done.
Comment 11: It is recommended to expand the list of references
Response 11: The list has been expanded by 10 references.
Reviewer 2 Report
Comments and Suggestions for AuthorsThis paper combines the STS and STAMP methods to solve the problem of truck congestion caused by disruptions in container terminals. The research method is appropriate, and research content has good practical application value. The following modifications and improvements are needed:
- The paper mentioned that the research method used has superiority over traditional methods, but there is a lack of detailed description of traditional methods, it is suggested to supplement the literature review.
- Based on the analysis of previous research, the author needs to further supplement the contributions and innovations.
- It is suggested to supplement the process of the analysis framework combined with STS and STPA.
- This research only interviewed two senior port management personnel, lacking information from truck drivers and terminal operators and other personnel, and the number of personnel is too small.
- The paper analyzes the complementarity of STS and STPA, but the description of how the two are integrated for analysis is not detailed.
Author Response
Comment 1: The paper mentioned that the research method used has superiority over traditional methods, but there is a lack of detailed description of traditional methods, it is suggested to supplement the literature review.
Response 1: The literature review has been expanded and now includes more detail on traditional (more “reductionist”) approaches (lines 61-90). The argument is now clearly made and supported that i) there is a need for more holistic approaches to resolve the intractable challenge of truck congestion, and ii) STS and STPA are established approaches for sociotechnical systems analysis, but have shortcomings that can be addressed by combining them. We have also added support by explaining that other authors are also seeing the value in combining system analysis approaches, using STAMP using HFACS as an example.
Comment 2: Based on the analysis of previous research, the author needs to further supplement the contributions and innovations.
Response 2: We hope that this has now been addressed adequately by the above. References added and literature section expanded (lines 61-90).
Comment 3: It is suggested to supplement the process of the analysis framework combined with STS and STPA.
Response 3: We did not make clear that the description of how STPA was used to structure the interviews – which is clearly laid out in the cited reference Leveson and Thomas (2018) – also contains the STPA analysis. We have now clarified this and explained how alterations were made to make interview results suitable also for analysis by STS, conducted according to cited ref 20 (see Interview 1 in Methods).
Comment 4: This research only interviewed two senior port management personnel, lacking information from truck drivers and terminal operators and other personnel, and the number of personnel is too small.
Response 4: The study’s purpose was not to obtain a representative set of stakeholder views but to conduct a deep, system-level analysis of the truck re-routing procedure through the integration of Sociotechnical Systems (STS) and Systems-Theoretic Process Analysis (STPA) frameworks. The two interviewees were selected purposively because they hold central coordinating responsibilities and system-wide visibility across terminal, traffic-control, and safety operations. Consistent with methodological guidance for expert studies in complex systems, this narrow but information-rich sample provided detailed insights. This was also discussed before preparing the study and after it, and we noted several reasons to stick with this approach, which we also detail here:
- This study aims to demonstrate how STS and STPA can be usefully combined as the findings complement each other, not to produce a complete empirical representation of all stakeholders or arrive at evidence about what causes port truck congestion in general. In such theory-building case studies, limited but information-rich sampling is methodologically acceptable.
- In sociotechnical and control-theoretic studies, the purpose of expert interviews is not breadth but depth: to capture how the system functions at a decision and coordination level. When the interviewees have system-wide visibility (e.g., regular contact with traffic control, terminal coordination, and procedure responsibility), even a small number can provide high information value.
- On reflection we saw no reason why management bias would affect the sociotechnical misalignments identified, driver behaviour or aspects of the system control structure. In STPA system loss identified is naturally defined by port management – it is their procedural outcomes that we are interested in improving – and all possible scenarios are considered affecting control actions and information flows, i.e. all perspectives are considered using a systematic process.
- The interviews were not used in isolation — they were cross-checked with operational documents, procedures, and driver surveys carried out by the port. This triangulation compensates for the small number of informants.
- The analysis focused on decision-making and control structure design — a level where front-line drivers and operators are recipients rather than controllers of the formal system. Thus, port-management experts were the appropriate informants for the system layer being modeled.
- While we agree that additional stakeholder consultation would further strengthen the findings, this was not feasible: the research was completed in 2021 under pandemic-related restrictions, and port procedures have since evolved, making retrospective member checking potentially misleading. (Please also consider that we were given 10 days to respond to the comments.)
We have clarified these methodological boundaries and rationale in the revised manuscript (Methods section, lines 193–205).
We have also added a section making clear this limitation, at the end of the Conclusions (line 546 onwards), to support future work considering system stakeholders.
In Methods at the start we now emphasize that the study is explorative, make arguments for why this is a methodologically sound approach given the study aims (see lines 191-206), but also clearly point this out that this could be seen as a limitation that future studies should address (see Conclusions).
Comment 5: The paper analyzes the complementarity of STS and STPA, but the description of how the two are integrated for analysis is not detailed.
Response 5: Thank you for this – we see now it was not entirely correct to use the phrase “integrated” – we have now replaced with the more correct term “combined”. We structure the interviews with STPA and analyse with STPA and STS, as is explained in the Method. Please note that the findings are combined rather than integrated, although we do actually attempt this in the Discussion.
Reviewer 3 Report
Comments and Suggestions for AuthorsThis manuscript proposes and demonstrates a method that integrates the Social Technology Systems (STS) theory with the System Theory Accident Model and Process (STAMP/STPA) for diagnosing and improving the truck re-routing procedures during the closure of the container terminal at the Port of Rotterdam. Through semi-structured interviews and document analysis, the research identified 92 unsafe control behaviors (UCAs) and 407 loss scenarios, and ultimately formed 16 design recommendations. The manuscript claims that this integrated framework can simultaneously provide "diagnostic insights" and "normative interventions", thereby enhancing the resilience, fairness and safety of the port transportation system. It is suggested that the author revise the manuscript according to the following suggestions.
1. Only two port management experts were interviewed, and key actors such as drivers, transportation companies, dock operators, and police were not included. This "management perspective" bias may have overemphasized the official narrative. The STPA control model was constructed by researchers post hoc and has not been cross-group validated, or it may have overlooked the "informal workarounds" and trust/power relationships in front-line operations. At least conduct 1-2 focus groups or online workshops, involving drivers and transportation representatives, to conduct member checks on their UCA and loss scenarios, thereby reducing the bias in the single perspective of managers.
2. STS and STPA are "parallel" rather than "integrated": The results of STS are inserted as background explanations and cultural, power, and goal conflict variables are not explicitly reflected in the control loops or process models of STPA. In the control structure, explicitly add a "culture/equity" controller node, and write the STS dimensions (goals, norms, trust) as the "process model" and "feedback channel" in the control loop, achieving the true integration of STS and STPA.
3. The research is limited to the specific road network and regulations of Masfractlake, and does not discuss the differences in systems, technologies or cultures of other ports (such as Antwerp and Singapore). The generalized statement is somewhat excessive. Without taking into account the regulatory effects of different seasons, time periods, peak transportation periods or special policies related to COVID-19 on the behavior of drivers, the scenarios may be insufficiently comprehensive.
Author Response
Comment 1: Only two port management experts were interviewed, and key actors such as drivers, transportation companies, dock operators, and police were not included. This "management perspective" bias may have overemphasized the official narrative. The STPA control model was constructed by researchers post hoc and has not been cross-group validated, or it may have overlooked the "informal workarounds" and trust/power relationships in front-line operations. At least conduct 1-2 focus groups or online workshops, involving drivers and transportation representatives, to conduct member checks on their UCA and loss scenarios, thereby reducing the bias in the single perspective of managers.
Response 1:
We appreciate the reviewer’s concern regarding potential perspective bias. The study’s purpose, however, was not to obtain a representative set of stakeholder views but to conduct a deep, system-level analysis of the truck re-routing procedure through the integration of Sociotechnical Systems (STS) and Systems-Theoretic Process Analysis (STPA) frameworks. The two interviewees were selected purposively because they hold central coordinating responsibilities and system-wide visibility across terminal, traffic-control, and safety operations. Consistent with methodological guidance for expert studies in complex systems, this narrow but information-rich sample provided detailed insights. This was also discussed before preparing the study, and we noted several reasons to choose this approach.
- This study aims to demonstrate how STS and STPA can be usefully combined as the findings complement each other, not to produce a complete empirical representation of all stakeholders or arrive at evidence about what causes port truck congestion in general. In such theory-building case studies, limited but information-rich sampling is methodologically acceptable.
- In sociotechnical and control-theoretic studies, the purpose of expert interviews is not breadth but depth: to capture how the system functions at a decision and coordination level. When the interviewees have system-wide visibility (e.g., regular contact with traffic control, terminal coordination, and procedure responsibility), even a small number can provide high information value.
- On reflection we saw no reason why management bias would affect the sociotechnical misalignments identified, driver behaviour or aspects of the system control structure. In STPA system loss identified is naturally defined by port management – it is their procedural outcomes that we are interested in improving – and all possible scenarios are considered affecting control actions and information flows, i.e. all perspectives are considered using a systematic process.
- The interviews were not used in isolation — they were cross-checked with operational documents, procedures, and driver surveys carried out by the port. This triangulation compensates for the small number of informants.
- The analysis focused on decision-making and control structure design — a level where front-line drivers and operators are recipients rather than controllers of the formal system. Thus, port-management experts were the appropriate informants for the system layer being modeled.
- While we agree that additional focus groups would further strengthen the findings, this was not feasible: the research was completed in 2021 under pandemic-related restrictions, and port procedures have since evolved, making retrospective member checking potentially misleading. (Please also consider that we were given 10 days to respond to the comments.)
We have clarified these methodological boundaries and rationale in the revised manuscript (Methods section, lines 193–205).
We have also added a section making clear this limitation, at the end of the Conclusions (line 546 onwards), to support future work considering system stakeholders.
Comment 2: STS and STPA are "parallel" rather than "integrated": The results of STS are inserted as background explanations and cultural, power, and goal conflict variables are not explicitly reflected in the control loops or process models of STPA. In the control structure, explicitly add a "culture/equity" controller node, and write the STS dimensions (goals, norms, trust) as the "process model" and "feedback channel" in the control loop, achieving the true integration of STS and STPA.
Response 2:
We appreciate this thoughtful observation and fully agree that closer integration of sociotechnical and control-theoretic perspectives is a valuable direction for future research. However, we deliberately kept the two analytical paradigms conceptually distinct in this study. STS and STPA are based on different ontological and epistemological assumptions: STS interprets social and technological dynamics in open systems, whereas STPA models control relationships within bounded systems. Integrating them does not necessarily mean embedding cultural constructs directly into control loops, which could compromise the formal semantics and analytical rigor of STPA.
Our intent was instead to pursue complementarity rather than fusion: the STS analysis diagnoses why breakdowns occur (through misalignments of goals, norms, and infrastructures), while the STPA analysis identifies how control actions and feedback structures can be redesigned to mitigate them. This sequential integration, or rather combination, enables each paradigm to contribute its respective strengths without distorting its theoretical foundations.
As a result of the reviewers comment we have clarified this rationale in the revised Discussion section (lines 518-523). We acknowledge the reviewer’s suggestion points to a need for methodological innovation—particularly for future work exploring how cultural or equity variables might be formally represented as higher-level control constraints or emergent properties within system-theoretic models. This will take some development.
Please also note we see now it was not entirely correct to use the phrase “integrated” – we have now replaced with the more correct term “combined”. We structure the interviews with STPA and analyse with STPA and STS, as is explained in the Method. Please note that the findings are combined rather than integrated, although we do actually attempt this in the Discussion.
Comment 3: The research is limited to the specific road network and regulations of Masfractlake, and does not discuss the differences in systems, technologies or cultures of other ports (such as Antwerp and Singapore). The generalized statement is somewhat excessive. Without taking into account the regulatory effects of different seasons, time periods, peak transportation periods or special policies related to COVID-19 on the behavior of drivers, the scenarios may be insufficiently comprehensive.
Response 3: We agree that the analysis was context-specific and have revised the manuscript to clarify this boundary more explicitly. The purpose of the study was not to produce a generalizable model of port behavior, but to use the Maasvlakte case as an analytic subject for demonstrating how STS and STPA can be integrated to analyze complex sociotechnical systems. The value of the findings therefore lies in transferability of insights—for example, the identification of systemic misalignments and feedback gaps that are likely to appear in other ports under different operational conditions—rather than in statistical generalization.
We also note that the empirical work was conducted in 2021, when COVID-19 regulations and traffic conditions were in flux; controlling for these exogenous factors would have been impractical and beyond the scope of the study’s system-analytic focus. Instead, we concentrated on structural relationships (information flow, control dependencies, and actor coordination) that remain relevant across operational contexts, irrespective of seasonal or pandemic-related variations. We have clarified this point and added a brief reflection on transferability and contextual limitations in the revised Discussion (lines 523-532).
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThanks for the corrections. Here are some more tips on what needs to be corrected:
1. The introduction should emphasize the gap between other scientists and highlight the need for your topic. At the end of the introduction, it is recommended to briefly describe the structure of the article itself.
2. Since there are subsections, it is recommended to make a short introduction at the beginning of section 2.
3. In lines 190 and 206, the same section and its title.
4. In line 207, the section number is missing, because now it is ".1. Case study: Truck rerouting in the Port of Rotterdam"
5. In line 263, the reference to the source is incorrectly provided (the same 277).
6. 627 and 629, the numbering of references is incorrect. Accordingly, this most likely affects the numbering in the text.
Author Response
Comment 1
The introduction should emphasize the gap between other scientists and highlight the need for your topic. At the end of the introduction, it is recommended to briefly describe the structure of the article itself.
Response 1
Added text to do this in lines 42-45; 56-59; and 68-72.
Comment 2
Since there are subsections, it is recommended to make a short introduction at the beginning of section 2.
Response 2
The start of Section 2 has also been changed accordingly, accounting also for the changes to the Introduction (line 74-77).
Comment 3
In lines 190 and 206, the same section and its title.
Response 3
We now give specific sections and titles – now lines 212, 227
Comment 4.
In line 207, the section number is missing, because now it is ".1. Case study: Truck rerouting in the Port of Rotterdam"
Response 4
“3.1. Case study: Truck rerouting in the Port of Rotterdam” – we think this was a confusion caused by track changes – switching off “show changes” makes it clearer.
Comment 5. In line 263, the reference to the source is incorrectly provided (the same 277).
Response 5
Thank you – changed from 22 to the correct reference number, 29.
Comment 6.
627 and 629, the numbering of references is incorrect. Accordingly, this most likely affects the numbering in the text.
Response 6
We have reviewed the numbering in the article and checked numbers are correct – but we couldn’t find anything wrong here.
Reviewer 2 Report
Comments and Suggestions for AuthorsAuthors have revised the paper according to the review comments and explained some issues. The paper is expressed clearly and has generally achieved the research purpose. Please further check typos, such as "3methods" on line 206.
Author Response
Please further check typos, such as "3methods" on line 206.
Thank you. We have checked and corrected for typos.
