0
Research Papers

Pairwise Elicitation for a Decision Support Framework to Develop a Flood Risk Response Plan PUBLIC ACCESS

[+] Author and Article Information
MiguelAndres Guerra

Civil and Environmental Engineering,
Virginia Tech,
1368 MacArthur Street,
Blacksburg, VA 24060
e-mail: MAGuerra@vt.edu

Yekenalem Abebe

School of Engineering,
University British Columbia,
3333 University Way,
Kelowna, BC V1V1V7, Canada
e-mail: yeke15@vt.edu

1Corresponding author.

Manuscript received September 30, 2017; final manuscript received April 29, 2018; published online August 14, 2018. Assoc. Editor: Faisal Khan.

ASME J. Risk Uncertainty Part B 5(1), 011004 (Aug 14, 2018) (7 pages) Paper No: RISK-17-1094; doi: 10.1115/1.4040661 History: Received September 30, 2017; Revised April 29, 2018

There are several ways of quantifying flood hazard. When the scale of the analysis is large, flood hazard simulation for an entire city becomes costly and complicated. The first part of this paper proposes utilizing experience and knowledge of local experts about flood characteristics in the area in order to come up with a first-level flood hazard and risk zoning maps, by implementing overlay operations in Arc GIS. In this step, the authors use the concept of pairwise comparison to eliminate the need for carrying out a complicated simulation to quantify flood hazard and risk. The process begins with identifying the main factors that contribute to flooding in a particular area. Pairwise comparison was used to elicit knowledge from local experts and assigned weights for each factor to reflect their relative importance toward flood hazard and risk. In the second part of this paper, the authors present a decision-making framework to support a flood risk response plan. Once the highest risk zones have been identified, a city can develop a risk response plan, for which this paper presents a decision-making framework to select an effective set of alternatives. The framework integrates tools from multicriteria decision-making, charrette design process to guide the pairwise elicitation, and a cost-effective analysis to include the limited budget constraint for any city. The theoretical framework uses the city of Addis Ababa for the first part of the paper. For the second part, the paper utilizes a hypothetical case of Addis Ababa and a mock city infrastructure department to illustrate the implementation of the framework.

FIGURES IN THIS ARTICLE
<>

Heavy rains and floods are common challenges that many cities face. Often, these floods disrupt the normal behavior of the city and its inhabitants. Furthermore, floods can disrupt the provision of services such as electricity, clean water, street sewer, transit mobility, and others. This paper proposes a method based on pairwise input to: (1) quantify flood hazards and (2) support decision-making for alternatives in facing these challenges. Flood risk assessment is the practice of identifying the nature plus extent of a certain flood risk through analysis of potential hazards and consideration of existing vulnerability as well as exposure of people, environment, property, livelihoods, and services. Therefore, risk mapping involves two major steps: first, identifying technical parameters of hazards such as their intensity, location, frequency, and probability of occurrence; and second, identifying vulnerability and exposure of the element under investigation that has socio-economic, environmental, or physical dimensions [1]. The proposed method is applied in Addis Ababa, Ethiopia.

Experts quantify flood hazard in various ways, including two-dimensional or three-dimensional hydrodynamic simulation, which is a well-established and most accurate method of quantification. However, this standard method requires high input data and specialized software, and while it yields the accurate results, it may not be the most appropriate method for a large-scale analysis as such a method becomes complicated and costly [2]. In order to deliver accurate outcomes, the simulations require the city to input high quality data about the flood characteristics for the area and may also require advanced technology [3] not available in every city worldwide. Additionally, due to limited resources, cities tend to develop risk response plans when the risk for a potential hazard is imminent. In this manner, however, cities do not have enough time to collect and input the data for a simulation. For this reason, in order to determine a first-level flood hazard and risk zoning maps, the authors propose consultation with local experts who are knowledgeable with the area and experienced in the field of flood disaster and prevention, as well as implementation of overlay operations in Arc GIS.

In this step, the authors use the concept of pairwise comparison to eliminate the need for carrying out a complicated simulation to quantify flood hazard and risk. The process begins with identifying main factors that contribute to flooding in a particular area. These factors are mostly similar everywhere. However, the contribution of each factor toward the final flood hazard is different from place to place. The authors use pairwise comparison to elicit knowledge from local experts and assign weights for each factor in order to reflect their relative importance toward flood hazard and risk.

Once a flood risk map for a city is built, a city can identify the highest risk zones and develop a list of alternatives to respond to the hazard, should it occur. Phase-two work involves the process of devising alternatives to mitigate the risk, and the analysis to decide which alternatives should be implemented. This analysis could be quite complicated because there are many stakeholders with different preferences, making the most efficient alternatives not evident when the resources available are limited.

The second part of this paper proposes a decision-making tool to help a government council make decisions more effectively, putting aside personal biases and preferences while focusing on the attributes of each of these alternatives. After identifying the vulnerable areas, a team of experts will brainstorm many alternatives to face the flood (or disaster). To remove personal biases from the decision, the group identifies the attributes that will be valued in the alternatives to be chosen. The group making the decision has a very different objective hierarchy because of its members' own natural roles in the city council. Principles of the charrette are used to help the team decide the weights of each attribute. Finally, the alternatives are organized according to their effectiveness, area of impact, and the cost/resources of implementing them. In this manner, the council can select the alternatives from a list ranked according to the area of impact per dollar (resources).

Charrette Design Process.

A charrette is a collaborative design process that aims to deliver a project with the support of all stakeholders, while also significantly reducing the design delivery time [4]. The term “charrette” dates back to 19th century France and refers to the cart where architectural students placed their time-limited final design drawings for their revision [5]. Appropriately termed, a charrette design process compresses the large amount of time typically required to deliver a design project to just a few days.

The National Charrette Institute (NCI) at Michigan State University proposes a three-phase approach. The first phase is the preparation of the charrette, which consists of gathering a multidisciplinary team and conducting site visits to the community in order for the team to familiarize itself with the context of the project [6]: the community, the location, the public institutions, the cultural characteristics of the community, etc. The second phase is the charrette itself, a design process that mandates three feedback loops—generating concepts, generating alternatives, and refining the final design [4]—between the design team and the different stakeholders (i.e., city designers, authorities, groups advocating specific concerns, future users) through private and public meetings. At the end of the second phase, the final project design is presented to the community for an open discussion about its implementation [7], allowing the design team to shape feasible design that addresses as many stakeholder concerns as possible. The third phase is the plan implementation in which the design team finishes the design details—stakeholder feedback included—and develops all the documentation required to implement the project [8].

A known example to illustrate the efficient application of charrette design process is the four-week charrette that took place as a response to hurricane Katrina in 2005 [9]. The charrette was sponsored by the Mississippi Renewal Forum and the Congress for the New Urbanism, where over two hundred professionals collaborated to develop recovery plans for eleven affected cities and towns in Mississippi [10]. These designs were in fact implemented in the reconstruction of this coastal region.

The objective of this study is to present a decision support tool for prevention and mitigation of floods, where human risk is reduced by the natural hazard influence. The authors used pairwise criteria to build a first-level hazard zones map and to develop a system that integrates multicriteria decision-making tools, charrette design process, and cost effectiveness analysis to help decision making under a context with limited resources (e.g., time to react, human action, money).

The authors developed a theoretical decision-making framework that integrates tools from pairwise elicitation, stakeholder participation (charrette), multicriteria decision making (MCDM) such as SMART and SWING SMART, and cost-effective analysis. Once the theoretical framework is laid down, the authors used the Addis Ababa city to develop the risk zones map and used a mock example for the city to apply and describe the second part of this framework.

To develop the flood risk map, the authors invited local professionals from the Addis Ababa municipality and experts from Addis Ababa University and University of Stuttgart (virtually) for a charrette-style meeting. The charrette-style meeting was held with thirteen experts. In an ideal case, this procedure should be done with the help of flood experts who have studied the characteristics of flooding in the area. The 13 experts who participated in this study have backgrounds in areas such as natural disasters (3), architecture (3), construction engineering (3), and civil engineering (4).

The goal of the meeting was to define and assign the contributing factors for flooding in the city. The participants completed a pairwise comparison table, following Saaty's [11] pairwise comparison nine-point scale, to drive the relative weight for each of the factors. Furthermore, the desired attributes for the flood mitigation solutions and infrastructure development alternatives were discussed. This meeting helped to elicit expert knowledge that will be used in both parts of the framework. The outcome of the meeting, the pairwise comparison table, can be seen in Table 1.

Using a gis software, the authors drew a first-level flood hazard zoning map with the information obtained previously. For the second phase, the authors used a mock charrette session with Virginia Tech graduate students in related fields to implement the decision-making framework to develop the risk response plan in case the hazard occurs. The steps of the framework consist of (i) developing the desire attributes and their relative weights, (ii) using SMART and SWING SMART to assign the relative weights to each alternative for every attribute, (iii) calculating the impact factor per dollar invested, (iv) performing the cost-effective analysis, and (v) selecting the best combination of alternatives within the city's limited budget. The final set of infrastructure projects selected represents the city's flood response plan.

Hydrodynamic simulation is the conventional and most accurate method of quantifying flood hazard. This simulation requires high input data and specialized software; thus, it can become complicated and expensive when working with large systems like an entire city. This framework proposes eliciting information from local experts about which flood characteristics are the most influential in building a first-level flood hazard zoning map. The authors used pairwise comparison to elicit this knowledge.

Pairwise comparison is part of the analytical hierarchy process (AHP). AHP involves structuring objectives and criteria in a hierarchical order [12] and implementing pairwise comparison for two purposes: first, to determine the relative weight of criteria for the attainment of the fundamental objective; and second, to determine the relative rank of each alternative with respect to the criteria [11]. The final priority vector of alternatives is the cumulative effect of the two. However, in our case, a pairwise comparison principle is solely used to drive a scale of priorities for major flood contributing factors based on local experts' knowledge. Pairwise evaluation is performed using a nine point scale [11]. This step should be supported by experts in the particular field and in the location, and the evaluation should involve a large number of experts in order to obtain reliable data.

After the pairwise comparison, the next step is to create an n × n matrix, with n representing the number of criteria [13]. To calculate the relative weights of criteria, the matrix is first normalized by dividing each value with the sum of the columns. The next step is to calculate the normalized principal Eigen vector or priority vector, which gives us the weights of each criterion. This is done by simply summing the row values and dividing them by the number of criteria [14].

Since pairwise comparison involves subjective personal judgments, the consistency of these evaluations needs to be tested. This is accomplished by calculating the consistency ratio (CR). A CR provides a measure of how the pairwise comparison matrix was filled and its consistency [14]. A higher CR implies that the input judgments were not consistent. CR is calculated by dividing the consistency index (CI) with the random index. Random index is a predefined value that is only dependent on the number of factors under consideration. The CI is calculated using the equation below [14]:

CI=(largesteigenvaluen)/(n1)

The maximum threshold of the consistency ratio is ten percent and if the result of the analysis produces a CI of greater than ten percent, the pairwise comparison has to be revised [11].

Deriving Weights From Pairwise Comparisons: Application in Our Case Study.

The pairwise comparison step of AHP is used to determine the relative weight of each flood hazard and risk contributing factors. Figure 1 summarizes the urban flood risk contributing factors. Each factor in the figure is further subdivided into two or five equal subgroups and scored from one to five, with highest values reflecting a high contribution to flooding. For example, mean annual rainfall in Addis Ababa ranges from 1175 mm to 1300 mm, and therefore, the rainfall factor is subdivided into 25 mm increment ranges. Areas that are in the highest range between 1275 and 1300 mm/year receive a score of five, while areas under 1175–1200 mm/year receive a score of one. All flood contributing factors in Fig. 1 have been assigned a sub group scoring in a similar fashion. Therefore, the risk score is simply the sum of the weight multiplied by the sub factor scores.

To quantify the relative weights of each factor, the first step is to make a pairwise comparison of each criterion. Again, in an ideal case, this procedure should involve local flood experts who have experience in flood hazard modeling in the area. Table 1 below is developed together with colleagues from Addis Ababa University and University of Stuttgart, and Addis Ababa city experts who are currently working on flood risk assessment. According to the expert's judgement, stream buffer and slope are five and nine times more relevant than rainfall intensity and elevation, respectively. Flooding of areas located near riverbanks is a common phenomenon in the city (Table 1). On the other hand, rainfall intensity received seven times more importance than elevation, while land cover was assessed to have three times more influence than soil type. Based on the pairwise comparison, the relative weight of each factor is calculated.

According to the pairwise comparison table, the elevation, soil type, and rain fall distributions received the lowest weights of 2.65%, 4.8%, and 9.84% respectively, while slope and stream network buffer received 34.98% and 33.55% weights, respectively. Finally, Fig. 2 shows the result of GIS overlay operation to prepare flood risk map for Addis Ababa.

The modern environmental management and urban planning discourse stresses the importance of community involvement in major decision making processes [15]. Top-down approaches where every decision is made at the city council level without community participation in the design or implementation of decisions have demonstrated many drawbacks [1517]. Community involvement in the decision-making process is very important to overcoming these drawbacks [18,19]. In addition, the multidisciplinary nature of flood risk mitigation involves a wide range of stakeholders from the different departments that influence decision making in the city administration such as in the transportation planning department, housing development office, or water and sewage authority; and one way or another, all have a role to play in the process [20]. Although all of these parties should be involved in the process of proposing alternatives to respond to flood hazards, their involvement makes the process complicated, since stakeholders' interests may have conflicting objectives that may prevent them from prioritizing the alternatives based on a common goal.

The decision-making support tool in this paper integrates multiple tools from MCDM (SMART and SWING SMART), community engagement (charrette), and budget allocation (cost-effectiveness analysis). MCDM tools help to compare and rank multiple alternatives that involve unequal attributes. Thus, MCDM is well suited for ranking different flood risk mitigation alternatives. It can help stakeholders make decisions in a flood management problem by framing their values and preferences, and quantify these priorities to apply them to a particular decision context [21]. This framework also supports reconciling different stakeholders' objectives, priorities, and personal biases. The section continues with an introduction to the charrette design process, followed by a detailed explanation of the process to integrate these tools in the proposed framework.

Using Charrette Design to Determine the Attributes and Their Relative Weights.

The first step is to define the attributes that the alternatives represent (i.e., environmental impact, implementation cost), and to determine the relative weight of the attributes in achieving the desired outcomes (i.e., the environmental impact is 20% less relevant than the implementation cost). The charrette design process allows all the stakeholders to provide feedback in an all-inclusive feedback loop. This feedback is processed right away in order to maintain the momentum between the feedback loops of the stakeholders. This perspective taking of the stakeholders helps them to process the viewpoints of other stakeholders. The final design contains the considerations provided by all members.

The authors use the case of Addis Ababa city to illustrate these steps. The attributes defined by the city after a charrette are the implementation cost, maintenance cost, life cycle, number of people impacted, construction to implement, time to drain-off, environmental impact, and the area of impact of the city. In the charrette, the city also assigned the relative weights of the attributes according to the end objective of the response plan for flood hazards. Table 2 shows the portfolio of infrastructure projects alternatives for the response plan. Table 2 also illustrates the attributes and their respective weights.

Swing Smart Tool for Decision Making.

Once the decision-making group has assigned the weights for each of the attributes, the next step is to use the SWING SMART tool to assign a score to each of the alternatives according to each of the attributes. This process consists of comparing the relative weight among all the alternatives (portfolio of infrastructure project) according to each attribute. For each attribute, the alternative that contributes least receives a zero; and the attribute that contributes the most receives a one hundred. All the other alternatives are located among this 0–100 scale for the specific attribute. The only exception to not assign zero to the lowest impact alternative is when the alternative's outcome is an absolute value. This is explained in the example described in the next paragraph. This process must be repeated for all the attributes defined in the charrette stage.

For the attribute “life cycle,” the alternatives of “building rain gardens” (ID: A1) and “improving the sewer system” (ID: E) receive scores of one hundred and zero, respectively, due to the highest and lowest impacts of the outcomes for this attribute. All the remaining alternatives are placed in this range according to their relative impact. However, in a few cases, there is no infrastructure project alternative with a zero score. This occurs when the attributes represent absolute values, such as the number of people impacted, that should not be represented by relative weights. For example, when evaluating the alternatives for the “number of people it impacts” attribute, the risk map indicates that there are one and a half million people in the medium, moderate, and high-risk zone of the city. Thus, one hundred is assigned to every alternative that will impact this whole population. Accordingly, the other alternatives are assigned a score with a direct proportion to the number of people they will impact.

This process allows to make the decision under more technical criteria, and therefore it helps to avoid the natural biases that the decision-making members have. Once all the alternatives have been scored for each attribute, a final weighted score is calculated. This weighted score considers the importance that each attribute has compared to all the attributes considered in the decision-making chart. The alternatives can be ordered now by their scores, which represent an order from the most beneficial to the less beneficial option. Table 3 shows the scores of all the alternatives for every attribute. The last column ranks the order of the alternatives according to their impact scores.

Cost-Effectiveness Analysis.

At this stage, the alternatives are ranked according to their overall impact. In many circumstances, this would be enough for a decision-making process; however, the portfolio of alternatives should be selected based on those outcomes, which offer the greater benefits per unit of resource invested (i.e., dollar). In other words, the objective is not to look for just the best option but the best feasible alternative that also includes investments considerations. Cost-effectiveness analysis helps to compare the costs to implement different courses of action and the benefits of the course of action outcomes [22,23].

A new organization that considers the effectiveness of the alternatives per dollar invested is needed. This new rearrangement is obtained using a calculation of the ratio of the alternative impact scores and the cost to implement each alternative. Table 4 shows the cost-effectiveness arrangement of the infrastructure projects alternatives.

Limited Budget Considerations for Decision Making.

A city cannot select all the “appealing” alternatives because cities are constrained by a finite budget. The arrangement of alternatives according to their cost-effective impact helps in selecting the best combination of alternatives under a limited budget. The city will select as many alternatives as long as the cumulative cost of the alternatives does not exceed the budget. This selected set of alternatives represents the courses of action for the city's hazard and risk response plan. Table 5 highlights the set of alternatives (ID: 10, 7, 11, 1) that a city should select with a hypothetical budget of $5.5M. Table 5 also presents another selection of alternatives if the hypothetical budget is only $5M. In this second scenario, the city could not include the fourth alternative, building rain gardens, because the cumulative cost would surpass the available budget. In this second hypothetical scenario, the most efficient selection would be to include ID-2, underground infiltration trenches, instead of ID-1, rain gardens. This selection of alternatives is efficient because the money that has not been used from an allocated budget will return to the city central account and may be reassigned somewhere else. For this reason, cities always try to maximize the use of their budget.

The work presented in this study represents a support tool for the decision-making process which cities must go through when facing potential natural disasters. The first step is to determine the area that the phenomenon can affect, because the rest of the tool is built upon these results. Time and budget are very important constraints in this process. The authors used pairwise comparison to identify the more vulnerable areas, since it is an accurate tool when the time constraint must be considered.

After identifying the vulnerable areas, the team brainstorms many alternatives to face the potential risk. As a way to start removing the personal biases from the decision, the group identifies the attributes that will be valued in the alternatives to be chosen. The group members who are making the decision have a very different objective hierarchy because of their own natural roles in the city council. Principles of the charrette design process are used to help the team decide the weights of each attribute. Each alternative is ranked based on the effectiveness of its implementation. The second constraint is the budget, and therefore it is of interest to select alternatives where the dollar invested produces the greatest benefits. This report presents solid concepts and principles for a tool to support the decision-making process.

This is a theoretical framework development that while it stands on strong research, it still needs to be applied in real cases of flooding for different cities. This validation with real cases will improve the usability of the decision-making framework. At the same time, a comparison of this method with hydrodynamic simulation is needed to determine the accuracy and practicality of the results. To avoid these limitations, the researchers have laid down the immediate research path. The next steps are to: (1) build comparisons with two-dimensional or three-dimensional hydrodynamic simulation; and to (2) validate this framework by developing case studies of applications to cities with different characteristics and flooding vulnerabilities. This second step should include (2.1) a study presenting and discussing the pairwise elicitation thoroughly for simplifying the existing techniques, and (2.2) a study applying the integrated tools for making decisions on flood response plans to real flood scenarios, and perhaps expanding into other natural disaster response plans. This future work will help to define the reliability and validity of this approach and define the conditions for best practices of this approach. Finally, this paper proposes a novel approach to develop efficient flood risk response plans for cities, providing an alternative to current methods that in large-scale can become costly and impractical.

This paper has been made possible due to the support of multiple parties. The authors want to thank: Dr. Christian Wernz, decision-making Professor at Virginia Tech (now Associate Professor at Virginia Commonwealth University) for inspiring the authors to do research related to decision-making, and for guiding the draft of this research; Dr. Ulrich Dittmer, Professor at University of Stuttgart, for guiding the pairwise elicitation process; the professionals and experts from Addis Ababa University and Addis Ababa Infrastructure Office who participated in the charrette-style meeting; the graduate students from Virginia Tech who participated in the mock charrette session to implement the decision-making framework in order to develop the flood risk response plan in case the hazard occurs; and the editors Amanda Wright Cron and Reema Azar for their professionality in editing and proofreading this paper.

Jalayer, F. , De Risi, R. , De Paola, F. , Giugni, M. , Manfredi, G. , Gasparini, P. , Topa, M. E. , Yonas, N. , Yeshitela, K. , Nebebe, A. , Cavan, G. , Lindley, S. , Printz, A. , and Renner, F. , 2014, “ Probabilistic GIS-Based Method for Delineation of Urban Flooding Risk Hotspots,” Nat. Hazards, 73(2), pp. 975–1001.
Kasvi, E. , Alho, P. , Lotsari, E. , Wang, Y. , Kukko, A. , Hyyppä, H. , and Hyyppä, J. , 2015, “ Two-Dimensional and Three-Dimensional Computational Models in Hydrodynamic and Morphodynamic Reconstructions of a River Bend: Sensitivity and Functionality,” Hydrol. Process, 29(6), pp. 1604–1629. [CrossRef]
Lee, J.-W. , Hong, S.-Y. , Kim, J.-E. E. , Yoshimura, K. , Ham, S. , and Joh, M. , 2015, “ Development and Implementation of River-Routing Process Module in a Regional Climate Model and Its Evaluation in Korean River Basins,” J. Geophys. Res. Atmos., 120(10), pp. 4613–4629. [CrossRef]
Lennertz, W. R. , and Lutzenhiser, A. , 2014, The Charrette Handbook: The Essential Guide for Accelerated, Collaborative Community Planning, American Planning Association, Chicago, IL.
Hackett, E. J. , and Rhoten, D. R. , 2009, “ The Snowbird Charrette: Integrative Interdisciplinary Collaboration in Environmental Research Design,” Minerva, 47(4), pp. 407–440. [CrossRef]
Guerra, M. A. , and Shealy, T. , 2018, “ Theoretically Comparing Design Thinking to Design Methods for Large-Scale Infrastructure Systems,” The Fifth International Conference on Design Creativity, Bath, UK, Jan. 31–Feb. 2, pp. 3–5.
Lennertz, B. , Lutzenhiser, A. , and Failor, T. , 2008, “ An Introduction to Charrettes,” Plan. Comm. J., 71, pp. 1–3. http://plannersweb.com/wp-content/uploads/2012/07/262.pdf
Gibson , G. E., Jr. , and Whittington, D. A. , 2009, “ Charrettes as a Method for Engaging Industry in Best Practices Research,” J. Constr. Eng. Manage., 136(1), pp. 66–75. [CrossRef]
Appleton, J. , 2013, Values in Sustainable Development, Routledge, Abingdon, UK.
Filmanowicz, S. , and Longwitz, W. , 2005, Mississippi Renewal Development Plan, Mississippi Renewal Forum, Gaithersburg, MD.
Saaty, T. L. , 2004, “ Decision Making—The Analytic Hierarchy and Network Processes (AHP/ANP),” J. Syst. Sci. Syst. Eng., 13(1), pp. 1–35. [CrossRef]
Zahedi, F. , 1986, “ The Analytic Hierarchy Process—A Survey of the Method and Its Applications,” Interfaces, 16(4), pp. 96–108. [CrossRef]
Veisi, H. , Liaghati, H. , and Alipour, A. , 2016, “ Developing an Ethics-Based Approach to Indicators of Sustainable Agriculture Using Analytic Hierarchy Process (AHP),” Ecol. Indic., 60, pp. 644–654. [CrossRef]
Ouma, Y. O. , and Tateishi, R. , 2014, “ Urban Flood Vulnerability and Risk Mapping Using Integrated Multi-Parametric AHP and GIS: Methodological Overview and Case Study Assessment,” Water, 6(6), pp. 1515–1545. [CrossRef]
Fraser, E. D. , Dougill, A. J. , Mabee, W. E. , Reed, M. , and McAlpine, P. , 2006, “ Bottom Up and Top Down: Analysis of Participatory Processes for Sustainability Indicator Identification as a Pathway to Community Empowerment and Sustainable Environmental Management,” J. Environ. Manage., 78(2), pp. 114–127. [CrossRef] [PubMed]
Davidson, C. H. , Johnson, C. , Lizarralde, G. , Dikmen, N. , and Sliwinski, A. , 2007, “ Truths and Myths About Community Participation in Post-Disaster Housing Projects,” Habitat Int., 31(1), pp. 100–115. [CrossRef]
Jerome, G. , 2017, “ Defining Community-Scale Green Infrastructure,” Landsc. Res., 42(2), pp. 223–229. [CrossRef]
El-Diraby, T. , 2013, “ Civil Infrastructure Decision Making as a Chaotic Sociotechnical System: Role of Information Systems in Engaging Stakeholders and Democratizing Innovation,” J. Infrastruct. Syst., 19(4), pp. 355–362. [CrossRef]
Roovers, G. J. , and van Buuren, M. W. , 2016, “ Stakeholder Participation in Long Term Planning of Water Infrastructure,” Infrastruct. Complex, 3, p. 1. [CrossRef]
Guerra, M. A. , and Shealy, T. , 2018, “ Teaching User-Centered Design for More Sustainable Infrastructure Through Role-Play and Experiential Learning,” J. Prof. Issues Eng. Educ. Pract., pp. 2–6.
Levy, J. K. , 2005, “ Multiple Criteria Decision Making and Decision Support Systems for Flood Risk Management,” Stoch. Environ. Res. Risk Assess, 19(6), pp. 438–447. [CrossRef]
Eichler, H.-G. , Kong, S. X. , Gerth, W. C. , Mavros, P. , and Jönsson, B. , 2004, “ Use of Cost-Effectiveness Analysis in Health-Care Resource Allocation Decision-Making: How are Cost-Effectiveness Thresholds Expected to Emerge?,” Value Health, 7(5), pp. 518–528. [CrossRef] [PubMed]
George, B. , Harris, A. , and Mitchell, A. , 2001, “ Cost-Effectiveness Analysis and the Consistency of Decision Making,” Pharmacoeconomics, 19(11), pp. 1103–1109. [CrossRef] [PubMed]
Copyright © 2019 by ASME
View article in PDF format.

References

Jalayer, F. , De Risi, R. , De Paola, F. , Giugni, M. , Manfredi, G. , Gasparini, P. , Topa, M. E. , Yonas, N. , Yeshitela, K. , Nebebe, A. , Cavan, G. , Lindley, S. , Printz, A. , and Renner, F. , 2014, “ Probabilistic GIS-Based Method for Delineation of Urban Flooding Risk Hotspots,” Nat. Hazards, 73(2), pp. 975–1001.
Kasvi, E. , Alho, P. , Lotsari, E. , Wang, Y. , Kukko, A. , Hyyppä, H. , and Hyyppä, J. , 2015, “ Two-Dimensional and Three-Dimensional Computational Models in Hydrodynamic and Morphodynamic Reconstructions of a River Bend: Sensitivity and Functionality,” Hydrol. Process, 29(6), pp. 1604–1629. [CrossRef]
Lee, J.-W. , Hong, S.-Y. , Kim, J.-E. E. , Yoshimura, K. , Ham, S. , and Joh, M. , 2015, “ Development and Implementation of River-Routing Process Module in a Regional Climate Model and Its Evaluation in Korean River Basins,” J. Geophys. Res. Atmos., 120(10), pp. 4613–4629. [CrossRef]
Lennertz, W. R. , and Lutzenhiser, A. , 2014, The Charrette Handbook: The Essential Guide for Accelerated, Collaborative Community Planning, American Planning Association, Chicago, IL.
Hackett, E. J. , and Rhoten, D. R. , 2009, “ The Snowbird Charrette: Integrative Interdisciplinary Collaboration in Environmental Research Design,” Minerva, 47(4), pp. 407–440. [CrossRef]
Guerra, M. A. , and Shealy, T. , 2018, “ Theoretically Comparing Design Thinking to Design Methods for Large-Scale Infrastructure Systems,” The Fifth International Conference on Design Creativity, Bath, UK, Jan. 31–Feb. 2, pp. 3–5.
Lennertz, B. , Lutzenhiser, A. , and Failor, T. , 2008, “ An Introduction to Charrettes,” Plan. Comm. J., 71, pp. 1–3. http://plannersweb.com/wp-content/uploads/2012/07/262.pdf
Gibson , G. E., Jr. , and Whittington, D. A. , 2009, “ Charrettes as a Method for Engaging Industry in Best Practices Research,” J. Constr. Eng. Manage., 136(1), pp. 66–75. [CrossRef]
Appleton, J. , 2013, Values in Sustainable Development, Routledge, Abingdon, UK.
Filmanowicz, S. , and Longwitz, W. , 2005, Mississippi Renewal Development Plan, Mississippi Renewal Forum, Gaithersburg, MD.
Saaty, T. L. , 2004, “ Decision Making—The Analytic Hierarchy and Network Processes (AHP/ANP),” J. Syst. Sci. Syst. Eng., 13(1), pp. 1–35. [CrossRef]
Zahedi, F. , 1986, “ The Analytic Hierarchy Process—A Survey of the Method and Its Applications,” Interfaces, 16(4), pp. 96–108. [CrossRef]
Veisi, H. , Liaghati, H. , and Alipour, A. , 2016, “ Developing an Ethics-Based Approach to Indicators of Sustainable Agriculture Using Analytic Hierarchy Process (AHP),” Ecol. Indic., 60, pp. 644–654. [CrossRef]
Ouma, Y. O. , and Tateishi, R. , 2014, “ Urban Flood Vulnerability and Risk Mapping Using Integrated Multi-Parametric AHP and GIS: Methodological Overview and Case Study Assessment,” Water, 6(6), pp. 1515–1545. [CrossRef]
Fraser, E. D. , Dougill, A. J. , Mabee, W. E. , Reed, M. , and McAlpine, P. , 2006, “ Bottom Up and Top Down: Analysis of Participatory Processes for Sustainability Indicator Identification as a Pathway to Community Empowerment and Sustainable Environmental Management,” J. Environ. Manage., 78(2), pp. 114–127. [CrossRef] [PubMed]
Davidson, C. H. , Johnson, C. , Lizarralde, G. , Dikmen, N. , and Sliwinski, A. , 2007, “ Truths and Myths About Community Participation in Post-Disaster Housing Projects,” Habitat Int., 31(1), pp. 100–115. [CrossRef]
Jerome, G. , 2017, “ Defining Community-Scale Green Infrastructure,” Landsc. Res., 42(2), pp. 223–229. [CrossRef]
El-Diraby, T. , 2013, “ Civil Infrastructure Decision Making as a Chaotic Sociotechnical System: Role of Information Systems in Engaging Stakeholders and Democratizing Innovation,” J. Infrastruct. Syst., 19(4), pp. 355–362. [CrossRef]
Roovers, G. J. , and van Buuren, M. W. , 2016, “ Stakeholder Participation in Long Term Planning of Water Infrastructure,” Infrastruct. Complex, 3, p. 1. [CrossRef]
Guerra, M. A. , and Shealy, T. , 2018, “ Teaching User-Centered Design for More Sustainable Infrastructure Through Role-Play and Experiential Learning,” J. Prof. Issues Eng. Educ. Pract., pp. 2–6.
Levy, J. K. , 2005, “ Multiple Criteria Decision Making and Decision Support Systems for Flood Risk Management,” Stoch. Environ. Res. Risk Assess, 19(6), pp. 438–447. [CrossRef]
Eichler, H.-G. , Kong, S. X. , Gerth, W. C. , Mavros, P. , and Jönsson, B. , 2004, “ Use of Cost-Effectiveness Analysis in Health-Care Resource Allocation Decision-Making: How are Cost-Effectiveness Thresholds Expected to Emerge?,” Value Health, 7(5), pp. 518–528. [CrossRef] [PubMed]
George, B. , Harris, A. , and Mitchell, A. , 2001, “ Cost-Effectiveness Analysis and the Consistency of Decision Making,” Pharmacoeconomics, 19(11), pp. 1103–1109. [CrossRef] [PubMed]

Figures

Grahic Jump Location
Fig. 1

Urban flood risk contributing factors

Grahic Jump Location
Fig. 2

Addis Ababa flood risk map

Tables

Table Grahic Jump Location
Table 1 Pairwise comparision table
Table Grahic Jump Location
Table 2 Portfolio of infrastructure projects and attributes with their relative weights
Table Grahic Jump Location
Table 3 Alternatives ranked under weighted attributes
Table Footer NoteNote: The boldface values represent the highest and lowest values of each attribute.
Table Grahic Jump Location
Table 4 Cost-effectiveness analysis - ranked alternatives
Table Grahic Jump Location
Table 5 Final selection of alternatives for a limited budget
Table Footer NoteNote: The boldface values represent the cumulative cost values that are limiting the proposed budget.

Errata

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Articles from Part A: Civil Engineering
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In