Multi-Criteria Decision Model for Wild-fire Risk Assessment at the level of a House using the Analytical Hierarchy Process

(1)

HAL Id: hal-02596899

https://hal.inrae.fr/hal-02596899

Submitted on 15 May 2020

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Multi-Criteria Decision Model for Wild-fire Risk

Assessment at the level of a House using the Analytical

Hierarchy Process

S. Meher

To cite this version:

S. Meher. Multi-Criteria Decision Model for Wild-fire Risk Assessment at the level of a House using the Analytical Hierarchy Process. Environmental Sciences. 2010. �hal-02596899�

(2)

Multi-Criteria Decision Model for Wild-fire Risk Assessment at

the level of a House using the Analytical Hierarchy Process

Somnath Meher

Research rapport of

4th Year Under‐Graduate diploma (B.Arch)

Cemagref EMAX, Aix-en-Provence

France

(3)

22nd July 2010 Aix en Provence

Multi-Criteria Decision Model for Wild-fire Risk Assessment at the level of a House using the Analytical Hierarchy Process

EMAX Research Unit Cemagref

Somnath Meher

Research Intern (May – July 2010) Cemagref, Aix en Provence

(4)

(5)

Contents:

• Acknowledgement • Objective of the Project

• Overview of Wildfire Risk in South of France

Wild Land – Urban Interface Mapping • Modus Operandi – Steps and Procedure • Multi Criteria Decision Making Methods • MCDM - Analytical Hierarchy Process (AHP) • AHP Model for Wildfire Risk Assessment

Factors for consideration Hierarchy formed

Weights and Consistency Calculations Final Weights at Different Levels • Indicators for factors of Fire Risk

• Application of the Model: Prototype Case - Meyruil

Data Analysis and Output Comparative Measure • Conclusion and Inferences

(6)

Acknowledgement:

I would hereby, like to take this opportunity to pay my sincere thanks to my supervisor for this project Professor Corrine Lampin Malliet, without whom this project would not have reached to completion. Her guidance, support and advice have been invaluable during the entire course of forming this Multi Criteria Model, starting right from its inception to its final day of completion.

I would also like to thank Prof. Christopher Buillon for his tireless support and advice whenever I was in need of one, Prof. Marlene Long for her kind suggestions and remark over the work, and Denis Morge for all the pain to have gone through the data base and provide me with the relevant information I needed for the study of Meyruil case. Their support has played a pivotal role in helping me complete this project.

And finally I would like to thank all the members of the EMAX research unit and the staff at Cemagref for bearing with me whenever I needed help on petty matters. I am also obliged to my colleagues and fellow students at Cemagref, who were there as a constant source of inspiration and helped me generously to get along in the institute and outside, despite the small language barrier.

I sincerely hope that this model be of practical use to make a sound assessment of the risk of Wild-Fire in the Wild-land Urban Interface at the level of a house, and community at large; thereby helping mitigate the risk of damage and loss owing to the same. I also wish Cemagref and all its members all the very best with the works being carried out in this field.

Somnath Meher

Research Intern, Cemagref, Aix en Provence 4th Year Under-Graduate Student (B.Arch) Indian Institute of Technology, Kharagpur

(7)

Objective of the Project:

This project is in relevance to the study of Housing and Community Planning (HCP) as well as Building Construction Techniques and Materials (BCTM) as a part of my academic curriculum towards fulfillment of the major in Architecture, thereby helping me gain a thorough hands-on knowledge and experience of the same. The objective of this internship has been as follows:

To develop a multi criteria model to make the Wildfire risk Assessment in the Wild-land Urban Interface at the level of the house (and community)

Making a state of art for fire risk assessment considering criteria that increase/decrease fire risk levels in context of South France and make necessary recommendations

To elaborate and make assessment for a Test prototype case, in the region of Meyruil, Aix en Provence for which relevant data are already available

Overview of Wildfire Risk in Mediterranean and South-France:

Forest fires have been a huge problem in many countries across the globe as well as in the continent of Europe. The south of Europe which constitutes the Mediterranean Region is primarily affected by wild fires, along with a few other regions in the Central and Northern Europe. However, within this typical Mediterranean climate, the area burnt owing to forest fire has drastically come down since 1991. The credit for this could be given to the various fire prevention measures that have been made more efficient by improved meteorological prediction and innovative risk assessment methods.

(8)

Apart from Europe and, forest fires have been a prime source of concern for many parts of the world with a similar Mediterranean climatic conditions. Mediterranean forests are have their own specific complex and vulnerable characteristics. Their management with the objectives of protection, production and biodiversity calls for specialized and custom adapted techniques.

The fire activity in the given areas is attributed to the changing pattern of the usage of the land owing to change in agricultural and socio-economic factors. As such, in the recent years there has been major rise in the land abandonment as well as encroachment into the forest periphery over the Wildland Urban interface for various recreational as well as settlement purposes. In case of France, there has been a huge development on the industrial front with the growing needs over time. In fact, the country also sees a surplus of agricultural produce in the European Union. However, the farming land is far too often abandoned leading to its cover by regenerated forests which take place over the same land. The typical Conifers that grow in this Mediterranean region, happen to be the primary species that constitute this forest cover. Add to this, the phenomenon of climate change that has emerged into the picture in recent past leading to extreme conditions of elevated temperature, minimized rainfall and other factors that further attribute to the case of wild fire occurrence.

Hence, the consciousness to study the various aspects of the wild fire risk and the factors attributing to it has been of prime interest in research centres across the globe. This demands the development of innovative new models to not just study the attributes of the wild fire occurrences but also means to help make an assessment of the conditions affecting it well in advance. There have been various efforts made to carry out classification of zones and regions in the Mediterranean forest based on their specific vegetation and soil characteristics, thereby determining the susceptibility to fire risk.

Carrying forward along the lines of the aforementioned discussion, the first step of this project has seen the EMAX research team of Cemagref work on the defining the Wild-land Urban Interface. This has been facilitated through a tool to carry out the WUI mapping, thereby defining the zone based on vegetation and housing structure.

- Structure of Vegetation: - virtually absent

- sparse and discontinuous - compact and continuous - Housing Configuration:

- Isolated - Scattered - Clustered

This innovative method aims at developing a measurable criterion that can be quantified, to offer a new reading of the municipal territory: by combining four types of residential housing and three types of vegetation structure, thereby defining 12 types of habitat-forest interfaces.

(9)

Modus Operandi – Steps and Procedure:

Step I. Understanding the Risk Scenario at the House Level:

The first step of the project has been to carry out an extensive study of the existing materials on study of wildfire and its consequences and other relevant attributes over the database. This has provided with an overview of the problems faced in France, and across the other parts of the world to help minimize the risk of wildfire. The study also includes a brief on different Multi-Criteria methods being used by researches for environmental risk assessment.

Step II. Assessment Model:

The second step of the project has been to develop a well defined Hierarchical structure in accordance to the Analytical Hierarchy Process (AHP) to be used for the Analysis of Fire Risk problem at hand. This structure includes the various factors affecting the wild fire risk, and a set of criteria thereby, defining these given factors. A coherent relation between these has then been established.

(10)

Step III. Defining the criterions:

The third step has been to take into consideration all the criteria to be taken into consideration for Wild fire risk assessment relevant to our specific case of study and further refine them to better suit the requirement. These criteria have then been assigned specific weights based on Saaty’s calculation method for AHP, based on Vector Analysis principles.

Step IV. Data Collection (Information on all the Indicators):

The next step has been the collection and compilation of the available data for a given community/region based on the case of study to test a prototype of the Multi-Criteria Analytical Model so developed. These refer to the data parcels received through satellite images as well as other stored documents on the topographic information of the region of study. The ground surveys of the site to gather specific information on definite topics and features by the Cemagref Team has also been taken into consideration.

(11)

Step V. Data Processing:

The next step has been to develop a hierarchy specifically on wildfire risk taking the factors that affect the same over space and time. These factors have then been sub-divided into various features, i.e. criteria that affect the same; and the calculation and an estimate for each of these criteria is carried out through a given set of indicators. Subsequently, based on expert judgment and available data analysis through AHP, these criteria have been assigned specific weights of importance at different levels in the hierarchy.

Step VI. Determining the relationship between the criteria (i.e. A mathematical model) :

The sixth step in the process has been to carry out the calculations defined for the Analytical Process based on pair wise comparison of the different set of attributes considered at each level of the

hierarchy. This is a reflection of the judgments based over a relative scale carried out, for which a certain consistency check is then performed.

(12)

Step VII. Quantifying the Output to judge on the final Goal (i.e. The assessment criteria) :

The final step of the project has been to carry out extensive data analysis and interpretation from the available records for the test Prototype Case (Meyruil). This is followed by a systematic representation of comparative risk value at the level of a House based on numerical outputs from the Model. Inferences have then be drawn to help experts make recommendations and suggestions to minimize the fire risk in the zone in future.

(13)

Multi-criteria decision Making (MCDM):

Multiple criteria decision making is an analytic method to evaluate the consequences and outcomes based on a given set of alternatives governed by multiple criteria. These include means to carry out complex decision making processes in the fields ranging from economics, to industrial research to environmental risk assessment.

MCDM problems can be broadly classified into two categories: Multiple objective programming and multiple criteria evaluation. The basic purpose in both, however, is to take explicit account of more than one criterion in supporting the decision process.

Properties of a specific multi-criteria evaluation method can be stated as:

- To assign a given set of multiple criteria that can be analysed for the problem at hand - To specify explicit weights for each criteria to assign preferential importance of each - To specify the acceptable metrics and their range, pertaining to the application at hand - To gather a wide range of judgmental opinions from varied sources and integrating them - To facilitate consideration of varied courses of action to be evaluated on a unique scale

MCDM methods have been widely used by researchers across the globe in recent times to carry out study on diverse topics. Each of the method for Multi Criteria Decision Making, as such provides a different approach to solving the given problem at hand through a process of choosing a particular among a given set of preselected alternatives, and rank them as desired.

The MCDM method that best suits our need to make a comprehensive Wild Fire Risk Assessment at the level of a house and a community happens to be the Analytical Hierarchy Process (AHP).

(14)

MCDM - Analytical Hierarchy Process (AHP):

The Analytic Hierarchy Process (AHP) is a structured technique for dealing with complex decision making requirements. The process presents a means to identify the best suited alternative, instead of picking one single outcome, through subjective understanding of the problem at hand. Hence, it comes rather handy when the decision making comes to listing down between a given set of choices and alternatives and the need to develop a rank among them, as in our case.

Based on mathematics and psychology, it was developed by Thomas L. Saaty in the 1970s and is often referred to as the Saaty method. But it has gone extensive refining since the very inception of its usage in varied fields of study. The crux of Analytical Hierarchy Process (AHP) lies in its effectiveness in establishing a coherent conclusion between conflicting alternatives. The judgment derived henceforth, are relative to each other based on their respective importance.

The basic steps in the AHP process for making a decision on the given problem could be listed down as: I. To simplify the decision problem into a hierarchy of more easily comprehended sub-problems,

each of which can be analyzed independently.

II. To decide which criteria are relevant for the given particular problem, and what metrics should be used to assess these set of criteria.

III. To evaluate its various elements (criteria) by comparing them to one another two at a time so as to make comparative analysis with the help of given data

IV. To check the consistency of the judgments and come to a final decision based on the results of this process.

As mentioned above, the method includes structuring the problem into a hierarchical scheme by distributing it into the various factors affecting the problem. This is followed by a set of criteria that go on to affect these factors, thereby contributing the main issue at the top of hierarchy. The next phase lies in deciding upon the relative importance among these set of factors and criteria based on human judgments.

(15)

Hence, the method helps cover not just a wide range of alternatives that affect the problem at hand, but also helps include a wide spectrum of ideas from various sources and decision makers, and correlate them to get a final conclusion. Further, the strength of the process lies in the ability to make a consistency check over the judgments made over the problem in consideration.

An analysis of the data available for the elements under consideration can be carried out to establish their impact and overall taking on the goal to address the problem. The analysis is threaded up the hierarchy, there by finally deriving a numerical value out of the entire process, to have a generic perspective of the analyzed result. The common scale that is used for the comparison is as given:

The process of judgment is carried out through a Saaty defined scale of comparison. In this scale, every alternative is compared to every other at the same level in hierarchy in a set of two; thereby establishing a comprehensive comparison among all the alternatives with respect to each other.

[The mathematical basis of the AHP lies in fundamental calculation over simple matrix and vector outlining the set of alternatives.

In case, we have n elements to be compared, where W1 … Wn and denote the relative ‘weight’

of Wi with respect to Wj by aij and form a square matrix A=(aij) of order n with the constraints that aij = 1/aji, for i ≠ j, and aii = 1, all i. Such a matrix is said to be a reciprocal matrix.

The weights are consistent if they are transitive, that is aik = aijajk for all i, j, and k. Such a matrix might exist if the aij are calculated from exactly measured data. Then find a vector ω of order n such that Aω = λω . For such a matrix, ω is said to be an eigenvector (of order n) and λ is an eigenvalue. For a consistent matrix, λ = n . For matrices involving human judgement, the condition aik = aijajk does not hold as human judgements are inconsistent to a greater or lesser degree. In such a case the ω vector satisfies the equation Aω= λmaxω and λmax ≥ n. The difference, if any, between λmax and n is an indication of the inconsistency of the judgments. If λmax = n then the judgements have turned out to be consistent.

(16)

Finally, a Consistency Index can be calculated from (λmax-n)/(n-1). That needs to be assessed against judgments made completely at random and Saaty has calculated large samples of random matrices of increasing order and the Consistency Indices of those matrices. A true Consistency Ratio is calculated by dividing the Consistency Index for the set of judgments by the Index for the corresponding random matrix. Saaty suggests that if that ratio exceeds 0.1 the set of judgments may be too inconsistent to be reliable. In practice, CRs of more than 0.1 sometimes have to be accepted. A CR of 0 would mean that the judgments are perfectly consistent.]*

Hence, the various advantages of AHP in a group setting can be noted as follows:

• A comprehensive understanding and simplification of the given problem into a structured system of the desired goal/objective and the factors affecting it.

• The discussion in a group can be focused on objectives rather than alternatives by Listing down the elements that attribute to the decision making process, both quantifiable and

non-quantifiable.

• Establishing the relation between the elements and the overall goal, thereby helping evaluate the options to come up with a desirable solution.

• In a structured analysis, the discussion continues until all relevant information from each individual member in a group has been considered and a consensus choice of the decision alternative is achieved.

(17)

AHP Model for Wildfire Risk Assessment:

Factors for Consideration:

Based on an extensive study of the different aspects of Wild Fire risk, the causes, properties and the consequences across the globe as well as a specific reference to the case of Mediterranean Europe, a set of exhaustive list of factors attributing to the same has been prepared. These factors directly or indirectly affect the risk of wild-fire in the given wild-land Urban interface. The list comprises a host of features ranging from the natural environmental system to the man controlled systems and activities. As such, the various factors have been broadly classified into:

- Spatial Factors - Temporal factors

The spatial factors include all the considerations over the specific region of study based on its topographical and demographic aspects. On the other hand the Temporal Factors are time specific and tend to vary over a given period of time within the day as well as over a given year or decades. These both factors combine to give a generic presumption of the net wildfire risk. The basic fundamental of it being inclusion of all the possibilities that affect:

- The very probability of Occurrence of a wildfire - The damage incurred in case of a wildfire

Hence, the various critical aspects based on the above mentioned observation can be categorized as follows:

- Bio-Climatic: Dealing with the various weather phenomenon and features over the given location of study.

- Topography: Affecting the overall landscape through which the fire propagates to reach the subject spot/location.

- Vegetation: The type of vegetation being the very fundamental aspect serving as the major fuel for wildfire.

- Planning Features: Including the community planning and housing configuration as well as other infrastructure and their maintenance.

- Building Construction Techniques and Materials: Referring to the relevant means and methods of construction incorporated in the specific region along with the combustibility of the materials used.

- Socio-Economic Factors: Referring to the human activities and their over all impact and contribution to the wildfire risk as well as the consequential damage incurred.

- Housing Type: This specifically refers to the housing configuration in the Wild Land – Urban interface based on the definite mapping scheme carried out at Cemagref, thereby categorizing the Wild-Fire risk zones.

(18)

Following is a set of exhaustive list of Criteria taken into consideration for the Wild-fire risk assessment in our case based on the above mentioned categorization of factors:

I. Temporal Factors: A. Bio Climatic

1. Temperature: It is the measurement of warmth or coldness thereby concluding that a high temperature would aid the heat for burning.

2. Wind: The horizontal movement of air relative to the surface of the earth which would essentially act as a carrier of heat and firebrands helping the fire to propagate.

3. Relative humidity (RH): It is the ratio of the amount of moisture in the air to the amount which the air could hold at a given temperature and pressure, giving a measure of the fuel dryness and susceptibility to catch fire in case of low moisture content.

II. Spatial Factors:

A. Topographic /Geographical

1. Slope: It refers to the change in elevation divided by horizontal distance, thereby defining the gradient of the land form to affect the travelling of wild-fire.

2. Aspect (Sun): It is the facing of slope with respect to the sun helping the one getting more sunlight to be dry, which is good for ignition and burning.

3. Country shape: It is an indication of the type of land form in the given specific region of study as they tend to affect the channelling of wild fire and the overall propagation.

B. Vegetation

1. Horizontal: This refers to the vegetation type (grass, shrubs, forest stands) and its specific characteristics that act as fuel for fire.

2. Vertical: This takes into consideration the specific distribution and concentration of vegetation within the given zone thereby affecting the overall fire risk.

C. Planning Features

1. Protection Zone: It refers to the area between the given property and nearest fuel from the potential wildfire source, often cleared so as to help protect life and property, spreading of fire and also provide space for fire fighters to work to protect life and property.

2. Accessibility (Road Type): It is considered at both the levels of a house and the community defining the Road width, Access Type, Ingress/Egress type, Surface Maintenance, Slope as well as other infrastructure for easy reach.

(19)

3. Electrical Utilities: It takes into consideration the various power lines which are either underground or over the ground power at a given Distance from the Structure, acting as potential accident sources in case of a wildfire as well as for ignition of a wildfire.

4. Gas/Fuel Utilities: It refers to the storage of fuel in different forms within the vicinity of a given structure in a storage vessel/barrel.

5. Water Supply: It refers to the presence of Hydrants and their proximity to the structure as well as the availability of a draft source to help mitigate fire in case of its occurrence.

D. Building Construction Materials and techniques

1. Roofing Material: This gives a representation of the combustibility and the maintenance of the roof thereby affecting the damage level caused in case of a wild fire.

2. Envelope/Cover: It is a representation of the construction material used for the basic envelope of the housing unit thereby giving an indication of its combustibility.

4. Siding/fence: It represents the outer fencing around the house, often in contact to the given building

5. Window Openings Shutter: This forms a very important feature of the building construction unit thereby helping facilitate channel for air flow and aid in burning in case of wild fire. The shutter material in itself further attributes to the case of combustion.

6. Machine Storage: It refers to the presence of any given machine equipment or vehicle with potential flammability through the fuel to power the machines, and is characteristic to its proximity from the structure

7. Firewood Storage: It refers to the presence of firewood piles or woodsheds in proximity to the structure acting as fuel source in case of a fire occurrence.

8. Barbecue/Chimney/Fireplace: It refers to the presence of any other potential fire ignition point in the form of barbecue or a chimney/fireplace within the living entity, based on its being screened and its maintenance.

E. Socio-Economic (Human) Factors

1. Type of Inhabitancy/Usage: The type of inhabitancy and the activities carried out within the built structure establishes the relative importance/value of the structure and hence, the damage quantification in case of fire.

2. Spot Identification/Demarcation: It refers to the official registration and demarcation of houses in a definite planned plotting helping easy access and identification for fire risk mitigation.

(20)

3. Viewshed: It refers to the ease of view and location from the major access and a given vantage point for spotting fire ignition as well as propagation.

5. Fire History: A previous record of wildfire occurrence in the vicinity gives an idea of the zone’s susceptibility and consequences to be incurred in case of further such occurrences.

6. Landscape Maintenance: It gives an indication of the frequency of clearing of fuel for the wild fire around the given structure under study based on the social and economic status of the inhabitants.

F. Housing Type

This is in relation to the mapping of the Wild-Land Interface using the WUI Tool developed by the research unit at Cemagref. By the help of this facility, it is easier to characterize the different dwelling units to have a good indication of the exact zoning of the given house based on its planning pattern. This is further in direct relation to the risk to the given structure in case of a Wild-Fire.

The various Housing Types categorized as such are as follows: - WUI Isolated

- WUI Scattered - WUI – Very Dense - WUI - Dense Clustered - Other than WUI

(21)

The Hierarchy Formed:

The factors so developed in the hierarchy are governed by a set of criteria each. These criteria hence form the subordinate level of the hierarchy in the AHP. Some of the criteria are quantifiable i.e. can be measured on mathematical terms while the rest cannot be quantified. Yet a comparative analysis gives an idea structure for deciding a relative importance between them and assessment with respect to the specific house being studied.

Following is a tabular representation of the entire hierarchy distributing the factors and the respective criteria into different levels of consideration. These criteria are further governed by a given set of indicators which give a parametric indication over the criteria of study.

(22)

Tabular Representation along with the Criteria:

Goal Factors Criteria Sub-Criteria

Temperature Wind

Temporal Bio-Climatic Relative Humidity

Elevation Slope Aspect (Sun) Aspect (Wind) Topography Country shape Vertical Vegetation Horizontal Protection Zone Accessibility Response Resources Building Density Electrical Utilities Gas/Fuel Utilities Water Supply Planning Features Obstacles Roof/Deck Material Foundation & Stilts Siding/ Fence Eaves &Balconies Window Openings Machine/Vehicles Firewood Storage Building Construction

Materials and Techniques

Chimney Land value Type of Inhabitancy/Usage Spot Identification View shed Fire History Socio-Economic (Human) Landscape Maintenance

Fire Risk Assessment

Spatial

(23)

Weights and Consistency Calculations:

Based on the comparative analysis scheme according to Saaty’s Analytical Hierarchy Process discussed earlier in the report, following calculations and comparisons have been carried out based on expert opinions and analytical studies. These comparisons yield a given set of weights at each level of the given hierarchy thereby defining the relative importance of the factors and the subsequent criteria with respect to each other.

Following the weight calculations, a given consistency check has been carried out for each of them to establish coherency in the decision making and eradicate human error in judgments.

The calculations along with the Maximum Eigen value and Consistency Index (which needs to be below 0.1) have been mentioned below:

LEVEL: 0 Factors for Basic Classification of Wild-Fire Risk

Calculation of Weights – Level o :

Criteria More Important Intensity

A B

Temporal Spatial A 5

Weghts and C.I.

Maximum Eigen Value = 2 C.I.= 0

Weights (Eigen Vector)

Temporal 0.833333

Spatial

0.166667

LEVEL: 1 CRITERIA for Fire Risk Assessment

Calculation of Weights – Level I :

A B

Vegetation Topography A 5

Vegetation BCTM Both 1

Vegetation Planning A 2

(24)

Topography BCTM B 7 Topography Planning B 6 Topography Socio-economic B 5 BCTM Planning A 2 BCTM Socio-economic A 3 Planning Socio-economic B 2

Housing Type Vegetation A 7

Housing Type Topography A 7

Housing Type BCTM Both 1

Housing Type Planning B 7

Housing Type Socio-economic A 3

Weghts and C.I.

Maximum Eigen Value = .58157 C.I.=0.095

Vegetation 0.114616 Topography 0.0280336 BCTM 0.208096 Planning Features 0.0910252 Housing Type 0.402435 Socio-Economic 0.155794

(25)

LEVEL: 2 SUB-CRITERIA for Bio Climatic Consideration

Calculation of Weights – Level II (A):

Criteria More

Important Intensity

A B

Temperature Wind Both 1

Temperature Relative Humidity A 3

Wind Relative Humidity A 5

Weghts and C.I.

Maximum Eigen Value =3.02906 C.I.=0.0145319

Temperature 0.405388

Wind 0.48064

Relative Humidity 0.113972

LEVEL: 2 SUB-CRITERIA for Topographic Consideration

Calculation of Weights – Level II (B):

A B

Slope Aspect (Sun) B 8

Slope Aspect (Wind) B 8

Slope Country Shape B 8

Aspect (Sun) Aspect (Wind) Both 1

Aspect (Sun) Country Shape A 5

(26)

Weghts and C.I.

Slope 0.036077

Aspect (Sun) 0.413963

Aspect (Wind) 0.413963

Country Shape 0.135996

LEVEL: 2 SUB-CRITERIA for Planning Consideration

Calculation of Weights – Level II (C):

A B

Protection Zone Accessibility A 7

Protection Zone Electrical Utilities A 9

Protection Zone Gas/Fuel Utilities A 5

Protection Zone Water Supply A 5

Accessibility Electrical Utilities A 7

Accessibility Gas/Fuel Utilities B 1

Accessibility Water Supply A 3

Electrical Utilities Gas/Fuel Utilities B 8

Electrical Utilities Water Supply B 6

Gas/Fuel Utilities Water Supply A 3

Weghts and C.I.

(27)

Accessibility 0.154496

Electrical Utilities 0.0274393

Gas/Fuel Utilities 0.160463

Water Supply 0.0857818

LEVEL: 2 SUB-CRITERIA for Building Construction Technique and Materials (BCTM)

Calculation of Weights – Level II (D):

A B

Roof Stilts A 3

Roof Siding/ Fence B 8

Roof Window Openings/Shutter B 5

Roof Machine/Vehicles B 5

Roof Firewood Storage B 7

Roof Chimney A 3

Stilts Siding/ Fence B 8

Stilts Window Openings/Shutter B 7

Stilts Machine/Vehicles B 5

Stilts Firewood Storage B 5

Stilts Chimney A 3

Siding/ Fence Window Openings/Shutter Both 1

Siding/ Fence Machine/Vehicles A 5

Siding/ Fence Firewood Storage A 3

Siding/ Fence Chimney A 7

(28)

Window Openings/Shutter Firewood Storage A 5

Window Openings/Shutter Chimney A 7

Machine/Vehicles Firewood Storage B 2

Machine/Vehicles Chimney A 3

Firewood Storage Chimney A 5

Weghts and C.I.

Roof Cover/Maintenance 0.0482384 Evelope/Cover 0.0340961 Siding/Fence 0.312417 Window Openings/Shutter 0.327649 Machines/Vehicles 0.101868 Firewood Storage 0.148946 Barbecue/Chimney/Fireplace 0.0267868

LEVEL: 2 SUB-CRITERIA for Socio-Economic Consideration

Calculation of Weights – Level II (E):

A B

Land Value Usage Type B 7

Land Value Spot Identification B 3

Land Value View Shed B 5

Land Value Fire History A 2

Land Value Landscape Maintenance B 8

(29)

Usage Type View Shed A 2

Usage Type Fire History A 7

Usage Type Landscape Maintenance Both 1

Spot Identification View Shed A 2

Spot Identification Fire History A 2

Spot Identification Landscape Maintenance B 7

View Shed Fire History A 3

View Shed Landscape Maintenance B 5

Fire History Landscape Maintenance B 7

Weghts and C.I.

Maximum Eigen Value =6.4158 C.I.= 0.0831605

Land Value 0.0421083 Usage Type 0.297633 Spot Identification 0.103651 View Shed 0.112446 Fire History 0.0389181 Landscape Maintenance 0.405243

LEVEL: 2 SUB-CRITERIA for Vegetation

A B

Horizontal Vertical A 3

Weghts and C.I.

(30)

C.I.=0

Horizontal 0.75

Vertical 0.25

LEVEL: 1 CRITERIA for Housing Type

Calculation of Weights – Level I (B):

Criteria More Important Intensity A B OWUI WUI-IS B 7 OWUI WUI-SC B 9 OWUI WUI-DC B 5 OWUI WUI-VDC B 5 WUI-IS WUI-SC A 7 WUI-IS WUI-DC A 1 WUI-IS WUI-VDC A 3 WUI-SC WUI-DC A 8 WUI-SC WUI-VDC A 6 WUI-DC WUI-VDC B 3

Weghts and C.I.

OWUI 0.0304456 WUI-IS 0.478786 WUI-SC 0.244394 WUI-DC 0.0967634 WUI-VDC 0.14961

(31)

Weights and Consistency Calculations:

Based on the comparative analysis scheme according to Saaty’s Analytical Hierarchy Process discussed earlier in the report, following calculations and comparisons have been carried out based on expert opinions and analytical studies. These comparisons yield a given set of weights at each level of the given hierarchy thereby defining the relative importance of the factors and the subsequent criteria with respect to each other.

Following the weight calculations, a given consistency check has been carried out for each of them to establish coherency in the decision making and eradicate human error in judgments.

Calculation of Weights – Level o :

A B

Temporal Spatial A 5

Weghts and C.I.

Maximum Eigen Value =2 C.I.=0

Temporal 0.833333

Spatial

0.166667

Calculation of Weights – Level I :

A B Vegetation Topography A 5 Vegetation BCTM Both 1 Vegetation Planning A 2 Vegetation Socio-economic B 3 Topography BCTM B 7

(32)

Topography Planning B 6

Topography Socio-economic B 5

BCTM Planning A 2

BCTM Socio-economic A 3

Planning Socio-economic B 2

Housing Type Vegetation A 7

Housing Type Topography A 7

Housing Type BCTM Both 1

Housing Type Planning B 7

Housing Type Socio-economic A 3

Weghts and C.I.

Maximum Eigen Value = .58157 C.I.=0.095

Vegetation 0.114616 Topography 0.0280336 BCTM 0.208096 Planning Features 0.0910252 Housing Type 0.402435 Socio-Economic 0.155794

(33)

Calculation of Weights – Level II (A):

A B

Temperature Wind Both 1

Temperature Relative Humidity A 3

Wind Relative Humidity A 5

Weghts and C.I.

Temperature 0.405388

Wind 0.48064

Relative Humidity 0.113972

Calculation of Weights – Level II (B):

A B

Slope Aspect (Sun) B 8

Slope Aspect (Wind) B 8

Slope Country Shape B 8

Aspect (Sun) Aspect (Wind) Both 1

(34)

Aspect (Wind) Country Shape A 5

Weghts and C.I.

Slope 0.036077

Aspect (Sun) 0.413963

Aspect (Wind) 0.413963

Country Shape 0.135996

Calculation of Weights – Level II (C):

A B

Protection Zone Accessibility A 7

Protection Zone Electrical Utilities A 9

Protection Zone Gas/Fuel Utilities A 5

Protection Zone Water Supply A 5

Accessibility Electrical Utilities A 7

Accessibility Gas/Fuel Utilities B 1

Accessibility Water Supply A 3

Electrical Utilities Gas/Fuel Utilities B 8

Electrical Utilities Water Supply B 6

Gas/Fuel Utilities Water Supply A 3

Weghts and C.I.

(35)

Protection Zone 0.571821

Accessibility 0.154496

Electrical Utilities 0.0274393

Gas/Fuel Utilities 0.160463

Water Supply 0.0857818

Calculation of Weights – Level II (D):

A B

Roof Stilts A 3

Roof Siding/ Fence B 8

Roof Window Openings/Shutter B 5

Roof Machine/Vehicles B 5

Roof Firewood Storage B 7

Roof Chimney A 3

Stilts Siding/ Fence B 8

Stilts Window Openings/Shutter B 7

Stilts Machine/Vehicles B 5

Stilts Firewood Storage B 5

Stilts Chimney A 3

Siding/ Fence Window Openings/Shutter Both 1

Siding/ Fence Machine/Vehicles A 5

Siding/ Fence Firewood Storage A 3

(36)

Window Openings/Shutter Machine/Vehicles A 5

Window Openings/Shutter Firewood Storage A 5

Window Openings/Shutter Chimney A 7

Machine/Vehicles Firewood Storage B 2

Machine/Vehicles Chimney A 3

Firewood Storage Chimney A 5

Weghts and C.I.

Roof Cover/Maintenance 0.0482384 Evelope/Cover 0.0340961 Siding/Fence 0.312417 Window Openings/Shutter 0.327649 Machines/Vehicles 0.101868 Firewood Storage 0.148946 Barbecue/Chimney/Fireplace 0.0267868

A B

Land Value Usage Type B 7

Land Value Spot Identification B 3

Land Value View Shed B 5

Land Value Fire History A 2

Land Value Landscape Maintenance B 8

(37)

Usage Type View Shed A 2

Usage Type Fire History A 7

Usage Type Landscape Maintenance Both 1

Spot Identification View Shed A 2

Spot Identification Fire History A 2

Spot Identification Landscape Maintenance B 7

View Shed Fire History A 3

View Shed Landscape Maintenance B 5

Fire History Landscape Maintenance B 7

Weghts and C.I.

Maximum Eigen Value =6.4158 C.I.= 0.0831605

Land Value 0.0421083 Usage Type 0.297633 Spot Identification 0.103651 View Shed 0.112446 Fire History 0.0389181 Landscape Maintenance 0.405243

A B

Horizontal Vertical A 3

Weghts and C.I.

Maximum Eigen Value =2 C.I.=0

(38)

Weights (Eigen Vector) Horizontal 0.75 Vertical 0.25 Criteria More Important Intensity A B OWUI WUI-IS B 7 OWUI WUI-SC B 9 OWUI WUI-DC B 5 OWUI WUI-VDC B 5 WUI-IS WUI-SC A 7 WUI-IS WUI-DC A 1 WUI-IS WUI-VDC A 3 WUI-SC WUI-DC A 8 WUI-SC WUI-VDC A 6 WUI-DC WUI-VDC B 3

Weghts and C.I.

OWUI 0.0304456 WUI-IS 0.478786 WUI-SC 0.244394 WUI-DC 0.0967634 WUI-VDC 0.14961

(39)

(40)

Indicators for factors of Fire Risk:

Based on the hierarchy and set of various factors and criteria that has been established till now, following are a set of various indicators taken into consideration. These would help to carry out the calculation for different conditions of the set of houses for which risk assessment is to be carried out based on respective criteria each.

These indicators are so formulated to take into consideration both the quantifiable as well as the non-quantifiable parameters to affect the over all risk value. The data for some of these have been readily available with the concerned sources, while some others have been carried out through a physical survey. The rest are a conclusion of thorough analysis and study of relevant sources.

Criteria Sub-Criteria Indicators Wt

Temperature > 30⁰ C 20⁰ - 30⁰ C < 20⁰ C 0.3 0.2 0.1

Wind General Direction and Speed:

> 30 kmph & North East > 30 kmph & South East 0 kmph < speed < 30 kmph No wind 0.3 0.2 0.1 0 Bio-Climatic

Relative Humidity in Air Minm RH in Air: < 30 % > 30 % 0.5 0.2 Slope >30% 15-30% <15% 0.3 0.2 0.1 Aspect (Sun) Direction of Slope:

Between South-West and South-East Between South-West and North West Between South-East and North-East Between North-West and North-East

0.4 0.3 0.2 0.1 Aspect (Wind) South East

North West (Mistral)

0.4 0.2 Topography

Country shape Ridge

Drew end side Flat area

0.3 0.2 0.1

Vertical

Mixed: Shrubs and Stands Forest/stands Scrubland 0.3 0.2 0.1 Vegetation Horizontal

Part of High AI > 33% of total area Part of High AI < 33% & Low AI > 33% Part of High AI < 33% & Low AI < 33%

0.3 0.2 0.1

(41)

20mts clearing 50mts clearing 100mts clearing 0.3 0.2 0.1 Accessibility Limited Access/ Unmaintained

Limited Access/ Maintained Complete Access/ Unmaintained Complete Access/ Maintained

0.4 0.3 0.2 0.1 Electrical Utilities In contact with vegetative fuel

Underground

0.5 0 Gas/Fuel Utilities Stored 10 mts from house

Stored 20 mts from house Underground Storage

0.3 0.2 0 Planning Features

Water Supply No hydrants, no draft source Hydrants > 25mts, No draft source Hydrants > 25mts & Draft sources Hydrants < 25mts

0.4 0.3 0.2 0.1 Roof Material/Maintenance Combustible & Fuel cover

Combustible but no fuel cover Non combustible & Fuel Cover Non combustible

0.4 0.3 0.2 0.1 Envelope/ Outer Cover Combustible

Non-combustible

0.5 0.1 Siding/ Fence Flammable fence, Adjacent to structure

Flammable fence, Separated from structure Non-flammable fence

0.2 0.1 0 Window Openings/Shutter Opening > 1.2 sqm, combustible shutter

Opening > 1.2 sqm, non-combustible shutter Opening < 1.2 sqm, combustible shutter Opening < 1.2 sqm, non-combustible shutter

0.4 0.3 0.2 0.1 Machine/ Vehicles Combustible, < 10mts from structure

Combustible, > 10mts from structure Non- Combustible, < 10mts from structure No Machines/ Vehicles

0.3 0.2 0.1 0 Firewood Storage Flammable, < 10mts from structure

Flammable, > 10mts from structure No Firewood Storage 0.2 0.1 0 Building Construction Materials and Techniques

Chimney/Fireplace/Barbecue Barbecue/Chimney Fixe (close to structure) Barbecue/Chimney Fixe (far from structure) No Barbecue/Chimney

0.2 0.1 0

Land value High

Moderate Low

0.3 0.2 0.1 Type of Inhabitancy/Usage Public building/ Permanent Usage

Residential Building/ Permanent Usage Public Building/ Seasonal Usage Residential Building/ Seasonal Usage

0.4 0.3 0.2 0.1 Spot Identification Unregistered plot/ No direction signage

Unregistered plot with direction signage Registered plot/ No direction signage Registered plot with direction signage

0.4 0.3 0.2 0.1 Socio-Economic (Human)

View shed Not Visible from main access & a vantage point Not visible from access, visible from vantage point

0.4 0.3

(42)

Visible from main access, not from a vantage point Visible from both main access & a vantage point

0.2 0.1

Fire History HFID and HWFD

HFID and LWFD LFID and HWFD LFID and LWFD None 0.4 0.3 0.2 0.1 0 Landscape Maintenance Low income/unemployed; frequency > 1 year

Employed; frequency > 1 year

Low income/unemployed; frequency < 1 year Employed; frequency < 1 year

0.4 0.3 0.2 0.1 Housing Type WUI – Isolated WUI – Scattered

WUI – Very Dense / Dense Clustered Other than WUI

0.3 0.2 0.1 0.0

(43)

Application of the Model: Meyreuil Case

Data Analysis an Output: [calculation files attached]

Based on the Data acquired over the region through varied sources, followed by application of the Saaty’s AHP procedure and comparative analysis of the criteria of consideration, a final set of risk value has been calculated. This value has further been normalized to draw a definite percentage of the same to better facilitate a generic overview of the amount of risk.

House Number Risk Value Normalized Value Risk Percentage

TYPEBAT IDENBAT 1 172 0,03421866 0,034077879 3,407787864 2 207 0,03383317 0,033693973 3,369397262 3 209 0,03461126 0,034468859 3,446885943 4 225 0,03402558 0,033885596 3,388559599 5 229 0,03392239 0,033782828 3,378282755 6 231 0,03379059 0,033651573 3,365157278 7 253 0,03411501 0,033974652 3,397465209 8 261 0,03413901 0,033998559 3,399855932 9 279 0,03437597 0,034234544 3,423454443 10 285 0,03426194 0,034120978 3,412097759 11 293 0,03485931 0,034715896 3,471589588 12 309 0,03476166 0,034618644 3,461864365 13 371 0,03418641 0,034045759 3,404575933 14 376 0,03991618 0,039751961 3,975196102 15 384 0,03448929 0,034347393 3,434739323 16 397 0,03350751 0,033369656 3,336965643 17 411 0,03417498 0,034034376 3,403437636 19 421 0,03879785 0,038638225 3,863822506 20 425 0,03393958 0,03379995 3,379994981 21 430 0,03283922 0,032704118 3,270411788 22 444 0,03443667 0,034294991 3,429499071 23 447 0,03394498 0,033805324 3,380532361 24 1399 0,03326954 0,033132664 3,313266428 25 1403 0,0390244 0,038863845 3,886384498 26 1445 0,03395635 0,033816651 3,381665082 27 1592 0,03509062 0,034946253 3,494625323 28 2000 0,03363896 0,033500565 3,350056462 29 2001 0,03429952 0,034158408 3,415840796 30 148155 0,03370455 0,033565881 3,356588079 1,00413117 1 100

(44)

Comparative Measure:

Hereby, given is a comparative graph of the final risk values calculated of the 29 houses. These happen to be an outcome of individual comparative study and analysis with respect to each other based on all the criteria of consideration.

The outcome though, is purely governed by the weights so assigned through the process and the available data within the limited resources. Following are two schematic representations of the values so obtained. Diagram I 0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Série1 Diagram II 0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Série1

(45)

Conclusion and Inferences:

Based on the output of the model, the houses taken into consideration of the study can be characterized into specific zones of wild-fire risk at the given Level of a House study. All the criteria along with other basic parameters can be developed into an algorithm or software, where the indicators define each of the criteria. The value of each of these indicators could be answered by the house owner himself by choosing among the set of options. The calculation then carried out, can draw a comparative study with all the other houses in the given zone (vicinity), and draw a mapping of the set of houses under different levels of risk to wild-fire.

Schematic Zoning: 0 0,5 1 1,5 2 2,5 3 3,5 4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Série1

The output could be diagrammatically modeled as represented, and necessary recommendations could be made specific to each house, based on its risk level. This would go a long way in creating an individual level awareness among the residents to the risk of wildfire as well as helping the researchers carry out the study in a more definite and refined manner.

(46)

Bibliography:

• Vulnerability to Natural Hazards: Study of Subdivisions Burned by the Hayman, Fire Using IKONOS Imagery and High Resolution (Author: Uddhab Bhandary, University of Colorado)

• Wildfire Hazard Assessment and the Wildland-Urban Interface of the North Olympic Peninsula, Washington (by Chris DeSisto, Dwight Barry, Tiffany Nabors, and Erin Drake, Peninsula College and Western Washington University Huxley College of the Environment)

• Fire Risk Rating for Homes in the Wildland/Urban Interface (created by the Alaska Wildland Fire Coordinating Group (AWFCG))Maryland Wildland/Urban Interface Home Risk Assessment

• Wildfire Risk Assessment Guide for Homeowners in the Southern United States (by Martha Monroe, faculty member; Ludie Ehlers and Anna Behm, graduate students,in the School of Forest Resources and Conservation, University of Florida)

• WILDFIRE HAZARD RISK ASSESSMENT Access Based Subdivision Survey (by Texas Forest Service and Texas State University