Dynamic Network Analysis of the Risks of Mega Infrastructure Projects from a Sustainable Development Perspective

Round 1

Reviewer 1 Report

I have read the paper on dynamic network analysis of the risks of mega infrastructure projects from a sustainable development perspective with great interest. Today, the analysis of risk factors, especially megaprojects, is essential in the construction sector as well as from the point of sustainability.

In my opinion, the article is not ready for publication. Authors should review the paper according the specific comments given below.

The results section lacks a relationship with the theory (for example, with the formulas presented in Table 3) and a detailed analysis of the results. For the reader, it should be clear how the risk factors and risk interactions were determined by the methods chosen.  

To achieve sustainable development of MIPs and implement appropriate risk mitigation strategies, the authors identified risks related to the sustainable development of MIPs, calculated and analyzed risk interactions in various stages of MIPs, and clarified the key risks of each stage (which could lead to the failure of MIPs).

The importance of the competency of the expert is crucial. Usually, experts are defined as individuals who work in practice as a practitioner or in science in the field (in both cases in mega infrastructure projects). Semi-structured interviews were conducted with 8 experts to determine risk relationships and weights. Given the mega infrastructure projects (MIPs) being evaluated, the number of experts is very small. Although the experience of the experts is extensive (see Table 1), if the risk factors and risk interactions of all stages of the MIPs life cycle (ie, decision-making stage, design stage construction stage and operation stage) are evaluated, we have one (government, designer, and operator) or two (construction contractor) representatives for a particular stage. Can the opinion and assessment of one expert be presented as a judgement/summary of the key risks of each stage of MIPs?

Are the authors familiar with the methods for checking the consistency ratio of experts’ opinions?

 “This study makes both theoretical and practical contributions. The results are of practical importance to participants of MIPs as it will assist in improving their understanding of MIPs’ risk and risk interaction relationships from a sustainable development perspective. They can then formulate reasonable risk mitigation strategies and promote the realization of sustainable development of MIPs.”

The conclusions are too ambitious considering that only eight experts were involved in the evaluation. The results obtained are more indicative than representative. The limitations of study in terms of the generalization of the findings should be added.

There are a number of errors (highlighted in yellow in the text) that need to be corrected.

Finally, I cannot recommend the paper for publication in the present form. The authors should improve the article and demonstrate the exclusivity and validity of their research.

 

Comments for author File: Comments.pdf

Author Response

Response to the comments: (March 19^th, 2022)

Buildings-1623818: Dynamic Network Analysis of the Risks of Mega Infrastructure Projects from a Sustainable Development Perspective

Dear reviewer:

We are grateful for your efforts and constructive comments. We have studied all the comments carefully and have made corrections and improvements accordingly. We hope that the revised manuscript meets the requirements of the journal.

Many thanks and best regards!

Here are our responses to the reviewers' comments point-by-point:

 

Point 1: The results section lacks a relationship with the theory (for example, with the formulas presented in Table 3) and a detailed analysis of the results. For the reader, it should be clear how the risk factors and risk interactions were determined by the methods chosen.

 

Response 1: Thanks for your comments. We have added the reasons for selecting critical risks and critical risk links to the text.

 

[Risks with low indegree and high outdegree values may be source risks and have a greater direct effect on other risks. Risks with high indegree and low outdegree values are regarded as cumulative risks because they have various sources and are more sensitive to the occurrence of other risks [4].]

[In this study, the top 40% of risk nodes at each stage sorted by indegree are selected as indegree key risks because 40% of the risk nodes have a larger indegree than the mean of all indegrees (21.4391). Similarly, the top 50% of the risk nodes at each stage sorted by outdegree are selected as outdegree key risks. Table 4 presents the indegree and outdegree key risks in each stage.]

[In this study, the top 40% of risk nodes at each stage sorted by RRGD-In were selected as RRGD-In key risks because 40% of the risk nodes have a larger RRGD-In than the mean of all RRGD-In (26.8024). Similarly, the top 45% of risk nodes at each stage sorted by RRGD-Out are selected as RRGD-Out key risks. ] 

[The higher the node betweenness centrality, the greater the control over the interaction relationships flowing through the network. ]

[In this study, the top 25% of the risks ranked by node betweenness centrality in each stage were selected as betweenness centrality key risks because 25% of the risk nodes have higher betweenness centrality than the mean of all betweenness centrality (0.0226).]

[In this study, the top 25% of risk links at each stage are sorted by link betweenness centrality. These links are selected as key risk links because 25% of them have a larger link betweenness centrality than the mean of all link betweenness centrality values (0.0043). ]

[After determining each risk node metric ranking, all rankings are integrated to obtain the final key risks in each stage of the MIPs. Careful attention should be paid to the risks that appear in three or more metric ranking lists. They play multiple roles with different functions to support risk networks. Combining the results of high node degree, high RRGD, and high betweenness centrality, Table 8 lists the key risks.]

[Key risk links were selected by betweenness centrality, as shown in Table 7. Ideally, by severing some links, the propagation effect of the risk interaction relationship will be reduced, thereby reducing the overall risk exposure [8]. Because there are too many key risk links at each stage, to understand their actual meaning and effect on the sustainable development of MIPs, this study classifies them and determines the main challenges faced by MIPs at each stage under the effect of risk interaction. The risk links in the same challenge have similar characteristics and can be solved similarly.]

Point 2: The importance of the competency of the expert is crucial. Usually, experts are defined as individuals who work in practice as a practitioner or in science in the field (in both cases in mega infrastructure projects). Semi-structured interviews were conducted with 8 experts to determine risk relationships and weights. Given the mega infrastructure projects (MIPs) being evaluated, the number of experts is very small. Although the experience of the experts is extensive (see Table 1), if the risk factors and risk interactions of all stages of the MIPs life cycle (ie, decision-making stage, design stage construction stage and operation stage) are evaluated, we have one (government, designer, and operator) or two (construction contractor) representatives for a particular stage. Can the opinion and assessment of one expert be presented as a judgement/summary of the key risks of each stage of MIPs? Are the authors familiar with the methods for checking the consistency ratio of experts’ opinions?

 

 

[Despite the small sample size, the findings are important as these experts hold senior positions in their respective organizations and have an average of 14.375 years of practical or research experience in the MIP field (Table 1). Therefore, these experts were able to provide insights that offered meaningful contributions.

Furthermore, the experts have rich work experience and knowledge covering the life cycles of MIPs: a government officer from Chongqing Commission of Housing and Urban-Rural Development participated in decision-making, construction, and operation stages of multiple MIPs; a designer from China Southwest Architectural Design and Research Institute has served as the chief designer of an MIP; two construction contractor from China State Construction International Investments Limited, one of them has worked with a consulting company, a design company, and a contractor company, participating in the design, construction, and operation stages of MIPs; an operator from Hong Kong-Zhuhai-Macao Bridge Authority participated in the construction, operation, and maintenance of the main part of the bridge; three university professors have research experience on the risk management process encompassing the complete life cycle of MIPs. Such expansive experience assures the reliability of the experts’ feedbacks for the study. ]

 

[ Therefore, the weightage of the link was unanimously raised to 4. At the end of the multi-round feedback, all experts agreed upon the effects of all risk links, and the final Weighted Risk Matrix (WRM; Eq. (2)) was developed. ]

 

 

Point 3: “This study makes both theoretical and practical contributions. The results are of practical importance to participants of MIPs as it will assist in improving their understanding of MIPs’ risk and risk interaction relationships from a sustainable development perspective. They can then formulate reasonable risk mitigation strategies and promote the realization of sustainable development of MIPs.”

The conclusions are too ambitious considering that only eight experts were involved in the evaluation. The results obtained are more indicative than representative. The limitations of study in terms of the generalization of the findings should be added.

 

Response 3: Thanks for your comments. We add this limitation in the conclusion section.

 

[Firstly, only eight experts were invited to evaluate the risk relationships. They might not be able to fully represent the actual conditions. But the findings are helpful in the risk management of MIPs, and researchers can increase the interview sample for further investigations. ]

 

Point 4: There are a number of errors (highlighted in yellow in the text) that need to be corrected.

 

Response 4: Thanks for your comments. The parts marked in yellow have been revised in the manuscript.

 

[[1]. MIPs]

[ life cycle of MIPs]

[Figure 1]

[Step1: Identify MIPs' risks]

[Weak contribution on local economy]

[cultural heritage]

[consequences]

[ Project Management Journal ]

[ Project Management Journal ]

[International Journal of Project Management]

[2020, 12, (1), 6. ]

[ International Journal of Environmental Research and Public Health]

[International Journal of Project Management]

Author Response File: Author Response.docx

Reviewer 2 Report

The paper presents a complex mathematical description of what eight experts have to say about the way that risks in mega infrastructure projects interact.  The identification of the risks categorised by the stage of the project (design – operation) and by sustainability dimension could be of value as a check list.  The analysis of the interaction of the risk does probably provide people who are familiar with risks associated with the projects (relatively experienced practitioners) with a useful short hand language for a lot of things they already know.  This is of course the primary value of this kind of study, it creates a way of talking about, and keeping in the forefront of a person’s mind the fact that for example, the realisation of an environmental disaster, is going to have economic consequences in the form of environmental remediation costs and social compensation payouts.  Although it could be argued that people who understand infrastructure projects understand that.  Actually as I began a review of this paper I wondered at its usefulness.  However on reaching section 5.3 I concluded that there is value (possibly even significant value) in undertaking the exercise.  This rather complicated set of analysis comes to the conclusion that greater transparency and public engagement are critical to the sustainability of major infrastructure.  This is pretty well understood, but the rather stark terms in which the flow on of outcomes which arise from failing to deal with this rather obvious risk mitigation strategy is described is I think on balance useful.

There are some issues with the paper which I have detailed below as I came to them.  I hope they are of assistance in dealing with improving the situation of the study in the introduction and literature review.  Which are quite weak in comparison with the strong and well written conclusion.

Some lack of clarity in the sentences starting on line 20,’Therefore,…’ but hopefully it will become clear what is meant when the methodology is more clearly explained.

The sentence ‘Develop risk networks…”, line 23 appears to be missing a word.

Line 45 Sentence ‘present multidimensional and interrelated.’ Is incomplete.

A range of loose language eg page 2 line 46 what is ‘spontaneous probablity’?

Page 2 line 52 what does the first sentence mean?  The sentence is incomplete.

Line 58, I think [8] should be the author name.

There are some issues with the introduction in terms of the logic and the way the study is expressed. 

The point is made that risks interrelate and presumably when realised they cascade into more significant disasters.  According to the authors should be taken into account in risk management strategies.  I actually think this should be taken into account when considering if the project is too risky to proceed something which the authors do not comment on.  In my view the actions taken to mitigate the risks is probably going to be much the same once the project commences.  This point is supported by the general finding of the paper which is that the primary action that can reduce risk is to increase transparency and fully involve the public.  The authors may wish to consider the nuances here in their review of the introduction.

The next point that the paper makes is that risk realisation can result in big harm to social, ecological and economics of areas and the project itself. Plus sustainable infrastructure is a SDG.  Unbalanced projects may increase risk.  I am not sure these three things go together in the way that they are causally linked in the paragraph on page 2 line 52 – 68. 

The question of risk is always tricky to write about because it is often the case that managing unrealised risks is quite a different proposition to managing risks what manifest (where risk management fails).  This paper makes the mistake of jumping between these two domains without a lot of clarity provided to the reader about what exactly is being discussed.  For example line 90 reads ’ This study aims to identify which key risks and key risk relationships will threaten’. The word ‘will’ suggests that what is being talked about is risks that are going to manifest, not those which need to be managed so that they do not occur.  Consideration of how risk (something that might occur) should be allocated and managed to prevent it occurring is completely different from consideration of how to manage the disaster that ensues if risk management fails and the risk manifests as an actual problem (which means it is no longer a risk it is a reality).

Line 100 – presumably the methodology is not proposed but has been carried out?

Figure 2, what are the numbers on the Y axis of the graph indicative of? – ah density.

Lines 421 – 424 underline this issue of the paper. 

Line 427 I think the word should be hub not hup.

The description of the impact of the results under 4.3 seemed less explanatory of the meaning of the results limiting the discussion to what the numbers do, rather than including what it means.  It would be good to see this expanded a little.

Line 478 there are fewer environmental risks in the decision making stage.  Perhaps but the findings from the rest of the paper suggest that environmental risk are realised by mistakes made in these stages (deciding to build the freeway over the area that is internationally protected wetland for example).  This is a good example of where the line between risk assessment, risk reduction, and disaster management needs extra clarity.  It also should be said that the issue with environmental risk is that they clearly will not be realised until the thing is built, the wetland destroyed and/or people are living under the freeway overpass (with attendant noise and air pollution) , because the somewhat cerebral process of planning a project do not have environmental effects – except possibly the extra consumption of coffee.

The entire document needs to be checked to make sure that where author’s names should appear in the text they do so and have not been replaced with numbers eg line 531 reads ‘However, this discovery also explains why some previ-531 ous scholars have focused on the social risks of MIPs, such as [20, 43].’

For the discussion section to be meaningful and readable could I suggest that where codes for the variables are used they are explained.  It is a very painful process to scroll up to where these variables are described in order to understand what is being said.  I might add that doing this would perhaps allow the authors opportunity to consider the value of what they are saying a little better.  For example the finding that in the decision stage if something is done that is not legal that creates problems down the track for the project is not really all that much of a revelation – although it probably is important to say, same with the apparent discovery that delays in decisions is the major problem that can occur in the decision process.  The statement that this risk being realised causes ‘risk accidents’ line 548 is a little confusing.

Actually as I read down through the discussion this process of saying what the risks that were being talked about was actually about was completed.  However it would be better to connect the two identifiers so that people reading the model outputs were constantly able to connect to the actual variable being discussed.

Line 582 the sentence starting with “Therefore, ..” is incomplete

Author Response

Response to the comments: (March 19^th, 2022)

Buildings-1623818: Dynamic Network Analysis of the Risks of Mega Infrastructure Projects from a Sustainable Development Perspective

Dear reviewer:

 We are grateful for your efforts and constructive comments. We have studied all the comments carefully and have made corrections and improvements accordingly. We hope that the revised manuscript meets the requirements of the journal.

Many thanks and best regards!

Here are our responses to the reviewers' comments point-by-point:

 

Point 1: Some lack of clarity in the sentences starting on line 20,’Therefore,…’ but hopefully it will become clear what is meant when the methodology is more clearly explained.

Response 1: Thanks for your comments. We have revised this issue in the manuscript as follows.

 

[Therefore, to analyse the dynamic risk interaction of MIPs, initially, through literature analysis and expert interviews, combined with the four dimensions of sustainable development and the four stages of MIPs, 98 risk factors of MIPs were identified.]

 

Point 2: The sentence ‘Develop risk networks…”, line 23 appears to be missing a word.

Response 2: Thanks for your comments. We have revised this issue in the manuscript as follows.

 

[Risk networks were developed for each stage of MIPs , and improved social network analysis was applied to these risk networks.]

 

 

Point 3: Line 45 Sentence ‘present multidimensional and interrelated.’ Is incomplete.

 

Response 3: Thanks for your comments.

 

[MIPs involve many potential risks, which are multidimensional and interrelated [4].]

 

 

Point 4: A range of loose language eg page 2 line 46 what is ‘spontaneous probablity’?

 

Response 4: Thanks for your comments. Risk spontaneous probability refer to the occurrence probability of a risk, ignoring the influence of all other risks and their cause-effect relationships.

 

[Risk interaction may lead to abnormal risk propagation, due to which the occurrence probability and negative impacts of risks are amplified, significantly jeopardizing the project objectives [5].]

 

 

Point 5: Page 2 line 52 what does the first sentence mean? The sentence is incomplete.

 

Response 5: Thanks for your comments. The introduction in the manuscript has been altered and this sentence has been deleted.

 

Point 6: Line 58, I think [8] should be the author name.

 

Response 6: Thanks for your comments. The introduction in the manuscript has been altered and this sentence has been deleted.

 

 

Point 7: There are some issues with the introduction in terms of the logic and the way the study is expressed.

The point is made that risks interrelate and presumably when realised they cascade into more significant disasters. According to the authors should be taken into account in risk management strategies. I actually think this should be taken into account when considering if the project is too risky to proceed something which the authors do not comment on. In my view the actions taken to mitigate the risks is probably going to be much the same once the project commences. This point is supported by the general finding of the paper which is that the primary action that can reduce risk is to increase transparency and fully involve the public. The authors may wish to consider the nuances here in their review of the introduction.

Response 7: Thanks for your comments. The purpose of this study is not to propose more innovative strategies, but to identify key risk factors and risk relationships at various stages of a MIP project through a study that considers dynamic risk interactions. These key risks may be inconsistent with those identified by assessment methods that do not consider dynamic interactions. Therefore, in response to these risks, targeted risk strategies are proposed to more effectively achieve project risk management and control.

 

[Mega infrastructure projects (MIPs) refer to large-scale public engineering projects with funds exceeding $1bn, usually commissioned by the government and delivered by capable private contractors and suppliers, that provide basic public services for social production and residents' lives [1]. MIPs are usually in the transportation, water conservation, telecommunication, power generation, and other such sectors that significantly affect national politics, economy, national security, public health, environmental protection, and society [2]. In contrast to small or medium-sized infrastructure projects, MIPs have a larger investment scale, longer implementation period, complex uncertain factors, and numerous stakeholders [3]. MIPs involve many potential risks, which are multidimensional and interrelated [4]. Risk interaction may lead to abnormal risk propagation, due to which the occurrence probability and negative impacts of risks are amplified, significantly jeopardizing the project objectives [5]. Failure to adequately analyse these complex relationships results in poor risk assessments, ineffective risk-mitigation initiatives and strategies [6]. Therefore, it is important to explore the interaction of risks to ensure successful risk management of MIPs.]

 

 

Point 8: The next point that the paper makes is that risk realisation can result in big harm to social, ecological and economics of areas and the project itself. Plus sustainable infrastructure is a SDG.  Unbalanced projects may increase risk.  I am not sure these three things go together in the way that they are causally linked in the paragraph on page 2 line 52 – 68. 

 

Response 8: Thanks for your comments. We have reorganized the logic of the introduction in the manuscript.

 

[Mega infrastructure projects (MIPs) refer to large-scale public engineering projects with funds exceeding $1bn, usually commissioned by the government and delivered by capable private contractors and suppliers, that provide basic public services for social production and residents' lives [1]. MIPs are usually in the transportation, water conservation, telecommunication, power generation, and other such sectors that significantly affect national politics, economy, national security, public health, environmental protection, and society [2]. In contrast to small or medium-sized infrastructure projects, MIPs have a larger investment scale, longer implementation period, complex uncertain factors, and numerous stakeholders [3]. MIPs involve many potential risks, which are multidimensional and interrelated [4]. Risk interaction may lead to abnormal risk propagation, due to which the occurrence probability and negative impacts of risks are amplified, significantly jeopardizing the project objectives [5]. Failure to adequately analyse these complex relationships results in poor risk assessments, ineffective risk-mitigation initiatives and strategies [6]. Therefore, it is important to explore the interaction of risks to ensure successful risk management of MIPs.

Many scholars agree that considering risk interactions is crucial while studying risk management for MIPs and have carried out a series of relevant studies [4, 7, 8]. For example, Chang et al. determined the interrelationship between political risk factors for international high-speed rail projects, to enable international contractors to better understand the political risks of projects [9]. Lu and Zhang identified the key safety risk factors during the construction stage of subways and drew relationships between the subway construction safety risk factors, which are conducive to reasonable risk management by the government and managers [10]. Chen et al. examined the diversity and interdependence of construction schedule risks and generated more reliable risk identification and risk inferences [11]. Etemadinia and Tavakolan analysed the risks and risks relationships in the design stage of construction projects and identified the key risks that have a greater impact on project goals [12]. While research on risk interactions has received some attention, most studies have focused on the development of static risk networks that only consider certain phases of MIPs and a single category of risk factors.

However, few studies have examined the dynamics of risk interactions over the life cycle of MIP and considered risk interactions among different categories. The risks of MIPs dynamically evolve throughout the project life cycle, with not only changes in the risk factors at each stage but also in the means and intensity of interactions between risks [13]. When the risk interaction changes drastically, the accuracy of the static risk network analysis decreases, underestimating the importance of some nodes. This is because in a static network important nodes at a certain time seem to be less important [14]. Furthermore, the effectiveness of risk control strategies proposed for key risks is greatly reduced. Therefore, it is necessary to transform the aggregated static risk network into a dynamic analysis of the risk network in different stages to determine the key risks.

Moreover, MIPs involve a variety of risks, including not only cost, schedule and quality risks, but also social, economic and environmental risks associated with the project [1]. Hence, it is imperative that sustainable development is addressed as a core issue in MIPs risk management. To achieve sustainable project delivery, economic losses, environmental issues, and social disputes cannot be handled separately [15]. For example, large-scale hydropower infrastructure projects reduce energy consumption and bring huge economic benefits. However, some of these projects have caused serious environmental degradation and ecological disasters. This has intensified confrontations between the public, enterprises, and the government. Tao et al. argue that the interrelationship between different risk categories increases the complexity of the risk system [16]. Hence, it is necessary and challenging to integrate multiple dimensions of sustainable development to conduct MIPs risk interaction research.]

 

Point 9: The question of risk is always tricky to write about because it is often the case that managing unrealised risks is quite a different proposition to managing risks what manifest (where risk management fails).  This paper makes the mistake of jumping between these two domains without a lot of clarity provided to the reader about what exactly is being discussed.  For example line 90 reads ’ This study aims to identify which key risks and key risk relationships will threaten’. The word ‘will’ suggests that what is being talked about is risks that are going to manifest, not those which need to be managed so that they do not occur.  Consideration of how risk (something that might occur) should be allocated and managed to prevent it occurring is completely different from consideration of how to manage the disaster that ensues if risk management fails and the risk manifests as an actual problem (which means it is no longer a risk it is a reality).

 

Response 9: Thanks for your comments. This article focuses on the consideration of how risk (something that might occur) should be managed to prevent it from occurring.

 

[This study aims to analyse the dynamic changes of risk interaction from different dimensions of sustainable development and different stages of the life cycle of MIPs. It expands the research perspective of MIPs risk interaction and makes up for the lack of consideration given to the dynamic characteristics and complexity of risks in the literature. This study uses the improved social network analysis (SNA) to calculate and analyse the dynamic changes in the multi-level network indicators for the established multi-stage risk network, and determine the key risks and risk relationships in each stage. This can help stakeholders develop risk response strategies to significantly reduce the complexity of risk networks to prevent or minimize adverse impacts and create more stable and sustainable projects. ]

 

Point 10: Line 100 – presumably the methodology is not proposed but has been carried out?

 

Response 10: Thanks for your comments. We have revised this issue in the manuscript as follows.

 

[Section 3 describes the research process, lists the risks, and proposes an improved SNA approach.]

 

Point 11: Figure 2, what are the numbers on the Y axis of the graph indicative of? – ah density.

Response 11: Thanks for your comments. We have made modifications to the figure.

 

Point 12: Lines 421 – 424 underline this issue of the paper.

Response 12: Thanks for your comments. We have revised this issue in the manuscript.

 

[If ‘bribery and corruption’ is not controlled at the beginning of MIPs, this risk will cause greater consequences as MIPs progresses.]

 

Point 13: Line 427 I think the word should be hub not hup.

Response 13: Thanks for your comments. The word should really be ''hub''. We have revised this issue in the manuscript

 

Point 14: The description of the impact of the results under 4.3 seemed less explanatory of the meaning of the results limiting the discussion to what the numbers do, rather than including what it means.  It would be good to see this expanded a little.

 

Response 14: Thanks for your comments. We extended our analysis of the results in this part.

Link betweenness centrality is a typical indicator for identifying key links [8]. It helps identify the hub links in the network that play the role of the key channels for risk transmission. Risk managers can prevent risk interaction by controlling these risk links. In this study, the top 25% of risk links at each stage are sorted by link betweenness centrality. These links are selected as key risk links because 25% of them have a larger link betweenness centrality than the mean of all link betweenness centrality values (0.0043). Table 7 ranks the top ten key risk links in each stage by link betweenness centrality.

 

There are 47 key risk links in the decision-making stage, the top three links ranked by link betweenness centrality value include EC1R4 - SO1R2 (0.0625; costs caused by disputes with the community - obstructed participation of local resident and community), EC1R1 - SO1R5 (0.0349; wrong market demand forecasts [overrate] - opportunism decision making), and SO1R4 - CO1R2 (0.0331; bribery and corruption - decision-making of all stakeholders).

There are 66 key risk links in the design stage, the top three links ranked by link betweenness centrality value include EC2R4 - SO2R7 (0.0390; land acquisition and resettling cost overruns - design deficiency), SO2R3 - EC2R3 (0.0342; no access of the disabled - opportunism decision making), and EN2R2 - CO2R3 (0.0303; discoordination between contractor and public - weak sustainability awareness).

The construction stage consists of 161 key risk links, the top three links ranked by link betweenness centrality value include SO3R12 - CO3R1 (0.0436; discoordination between contractor and public - inadequate communication and coordination among stakeholders), EC3R1 - SO3R2 (0.0293; construction and installation cost overruns - bribery and corruption), and SO3R12 - CO3R3 (0.0271; discoordination between contractor and public - weak sustainability awareness).

The operation stage consists of 73 key risk links, the top three links ranked by link betweenness centrality value include SO4R2 - SO4R4 (0.0915; damages on residents' health - damages on participation of local residents), EC4R2 - SO4R4 (0.0734; devaluation of residents' assets [decrease in residents' income] - damages on participation of local residents), and EC4R4 - EC4R2 (0.0589; Weak solvency ability - devaluation of residents' assets [decrease in residents' income]).

 

Point 15: Line 478 there are fewer environmental risks in the decision making stage.  Perhaps but the findings from the rest of the paper suggest that environmental risk are realised by mistakes made in these stages (deciding to build the freeway over the area that is internationally protected wetland for example).  This is a good example of where the line between risk assessment, risk reduction, and disaster management needs extra clarity.  It also should be said that the issue with environmental risk is that they clearly will not be realised until the thing is built, the wetland destroyed and/or people are living under the freeway overpass (with attendant noise and air pollution) , because the somewhat cerebral process of planning a project do not have environmental effects – except possibly the extra consumption of coffee.

Response 15: Thanks for your comments. We have added an explanation of the reasons for less environmental risk in the decision-making phase.

 

[In the decision-making stage, the largest direct inter-group interaction is from social risks to economic risks (52.3068), followed by coordination risks to social risks (36.6606), and then by coordination risks to economic risks (27.8209). The results show that social risks, economic risks, and coordination risks are closely related, whereas environmental risks seem to be relatively isolated. This may be because the environmental impact of a project is not exposed during the planning and start-up stages, and the project participants do not consider its environmental risks. However, the insufficient environmental impact assessment (EIA) in the initial stage leads to mistakes in decision-making, which leads to a large number of environmental risks in the construction and operation phases of the project. For example, if a highway is planned to be constructed in an international wetland reserve, until the project construction and operation stages begin, the wetland will be damaged and the surrounding residents will be affected by noise, air pollution, and other environmental impacts.]

 

Point 16: The entire document needs to be checked to make sure that where author’s names should appear in the text they do so and have not been replaced with numbers eg line 531 reads ‘However, this discovery also explains why some previ-531 ous scholars have focused on the social risks of MIPs, such as [20, 43].’

 

Response 16: Thanks for your comments. We've checked the entire document to make sure he entire document needs to be checked to make sure that where author's names should appear in the text they do so and have not been replaced with numbers.

 

Point 17: For the discussion section to be meaningful and readable could I suggest that where codes for the variables are used they are explained.  It is a very painful process to scroll up to where these variables are described in order to understand what is being said.  I might add that doing this would perhaps allow the authors opportunity to consider the value of what they are saying a little better.  For example the finding that in the decision stage if something is done that is not legal that creates problems down the track for the project is not really all that much of a revelation – although it probably is important to say, same with the apparent discovery that delays in decisions is the major problem that can occur in the decision process.  The statement that this risk being realised causes ‘risk accidents’ line 548 is a little confusing.

 

Response 17: Thanks for your comments. We have added an explanation of the risk code to the Discussion section.

 

Point 18: Actually as I read down through the discussion this process of saying what the risks that were being talked about was actually about was completed.  However it would be better to connect the two identifiers so that people reading the model outputs were constantly able to connect to the actual variable being discussed.

 

Response 18: Thanks for your comments. We have added an explanation of the risk code to the Discussion section.

 

Point 19: Line 582 the sentence starting with “Therefore, ..” is incomplete

 

[Therefore, training project participants must improve their knowledge about sustainability practices and foster a culture of sustainable development within the organization.]

 

Round 2

Reviewer 1 Report

I appreciate the great efforts of the authors that they have made in response to my questions and concerns. The revision clarifies the points I raised and helps me (and hopefully readers) to understand the manuscript.

This is an interesting paper and overall very good for what is aiming to achieve. The authors present a dynamic network analysis of the risks of mega-infrastructure projects from a sustainable development perspective. Today, the analysis of risk factors, especially megaprojects, is essential in the construction sector and from a sustainability point of view.

While I remain critical of the small number of experts, the article would be of interest to both researchers and practitioners.

In my opinion, the article is suitable for publication in its current form.