Md.Asif Rahman, profile picture
Uploaded byMd.Asif Rahman
PPTX, PDF124 views

Community resilience modeling, field studies and implementation

The document discusses modeling community resilience through the development of a virtual test community called Centerville. It describes creating standardized models of Centerville's physical infrastructure like buildings, roads, and electrical grid. It also models Centerville's social and economic attributes like demographics, employment, and critical facilities. The goal is to integrate these physical and social models to allow testing how disruptions impact infrastructure interdependencies and community resilience metrics. Researchers from different disciplines can collaborate using this testbed to develop and validate modeling approaches before applying them to real communities.

Education◦

Embed presentation

Download to read offline
1 / 112
2 / 112
3 / 112
4 / 112
5 / 112
6 / 112
7 / 112
8 / 112
9 / 112
10 / 112
11 / 112
12 / 112
13 / 112
14 / 112
15 / 112
16 / 112
17 / 112
18 / 112
19 / 112
20 / 112
21 / 112
22 / 112
23 / 112
24 / 112
25 / 112
26 / 112
27 / 112
28 / 112
29 / 112
30 / 112
31 / 112
32 / 112
33 / 112
34 / 112
35 / 112
36 / 112
37 / 112
38 / 112
39 / 112
40 / 112
41 / 112
42 / 112
43 / 112
44 / 112
45 / 112
46 / 112
47 / 112
48 / 112
49 / 112
50 / 112
51 / 112
52 / 112
53 / 112
54 / 112
55 / 112
56 / 112
57 / 112
58 / 112
59 / 112
60 / 112
61 / 112
62 / 112
63 / 112
64 / 112
65 / 112
66 / 112
67 / 112
68 / 112
69 / 112
70 / 112
71 / 112
72 / 112
73 / 112
74 / 112
75 / 112
76 / 112
77 / 112
78 / 112
79 / 112
80 / 112
81 / 112
82 / 112
83 / 112
84 / 112
85 / 112
86 / 112
87 / 112
88 / 112
89 / 112
90 / 112
91 / 112
92 / 112
93 / 112
94 / 112
95 / 112
96 / 112
97 / 112
98 / 112
99 / 112
100 / 112
101 / 112
102 / 112
103 / 112
104 / 112
105 / 112
106 / 112
107 / 112
108 / 112
109 / 112
110 / 112
111 / 112
112 / 112
Community Resilience: Modeling, Field Studies, and Implementation Welcome! 1
2 Thrust 1: A Multidisciplinary computational environment with fully integrated supporting databases: “The Interconnected Networked - Community Resilience Modeling Environment”. Thrust 2: “Data management tools for community resilience systems”: A standardized data ontology, robust data architecture, and effective tools to support IN-CORE. Thrust 3: Resilience Data Architecture Validation Studies: Hindcasts and forecasts to test the data collection process and its integration into IN-CORE. Validate risk-informed intelligent search and decision algorithms; field studies. The Center Thrusts
3 • Quantitative tools and metrics for assessing community resilience either are in a rudimentary state of development or are not available. • Multiple hazards, interconnected and inter-dependent systems must be considered. • Uncertainties, both aleatory and epistemic, must be included for the resilience assessment and decision process to be risk- informed. • Community resilience plans and guidance are needed. • Existing building and infrastructure systems, as well as new construction, must be considered. • Codes, standards and regulatory documents must contain consistent performance goals. Center research Issues
4 How can community leaders know how resilient their community is? How can they know if their decisions and investments to improve resilience are making a significant difference? How can they deal with uncertainty and risk? Research and Implementation Questions
5 Thrust 1: The Development of IN-CORE
6 • General and Recovery • Infrastructure performance • Spatial and logical distribution of physical damage • Time to functionality • Availability of resources and leadership • Dependencies among built environment and social institutions Community Resilience Metrics • Economic and Social • Retaining businesses and employment • Revenue resources • Poverty and income distribution • Economic sustainability • Survival – safety; water/ food • Safety and security – personal safety, financial security, health • Growth and achievement opps
7 • User requirements/Stakeholder Alignment ‒ Support for new data types being developed ‒ Building taxonomies for all hazards • Standardized Data Ontology ‒ Defining new data types such as need for CGE development ‒ Review of schema and metadata • Development of Data management Tools ‒ GEM taxonomy, DDI Codebook, lifecycle Thrust 2: A Robust Data Architecture
8 • Systematic validation of the numerical models and data in Tasks 1 and 2. • 2011 Joplin, Missouri tornado • Lumberton, NC Field Study • Single sector, multi-sector, and full architectural validation Thrust 3: Resilience Architecture Validation Studies
9 • Year 1: Team building and Hazards Center Themes by Year
10 • Year 2: Dependencies and Testbeds Center Themes by Year
11 • Year 3: Recovery modeling with full interdependencies • Year 4: Risk-Informed Decision-Making • Year 5: Technical outreach including user workshops Center Themes by Year
Thank you
13 Special Issue of Sustainable and Resilient Infrastructure on the Centerville Testbed
14 Several testbeds have been initiated since the beginning of the Center. The testbeds have been designed to 1. allow Center research teams to initiate, test, and modify essential community resilience assessment models and algorithms before the IN-CORE platform becomes fully operational 2. stress these assessment models in a controlled manner 3. examine varying degrees of dependency between physical, social, and economic infrastructure systems 4. facilitate the interdisciplinary collaborations and approaches to community resilience assessment that will be essential for the Center to achieve its ultimate goal Background
15  A special issue of the journal Sustainable and Resilient Infrastructure (volume 1, 2016 – Issue 3-4) has been devoted to the first of these testbeds namely the Centerville Virtual Community  Sustainable and Resilient Infrastructure is and interdisciplinary journal published by Taylor and Francis that focuses specifically on the sustainable development of resilient communities Special issue on the Centerville Virtual Community
16  The Centerville Virtual Community: A fully integrated decision model of interacting physical and social infrastructure systems Bruce R. Ellingwood, Harvey Culter, Paolo gardoni, Walter gills Peacock, john W. van de Lindt & Naiyu Wang  Building portfolio fragility functions to support scalable community resilience assessment Peihui Lin & Naiyu Wang  A multi-objective optimization model for retrofit strategies to mitigate direct economic loss and population dislocation Weili Zhang & Charles Nicholson  Probabilistic framework for performance assessment of electrical power networks to tornadoes Vipin U. Unnikrishnan & John W. van de Lindt  Modeling the resilience of critical infrastructure: The role of network dependencies Roberto Guidotti, hana Chmielewski, Vipin U. Unnikrishnan , Paolo Gardoni, Therese McAllister & John van de Lindt  Integrating engineering outputs from natural disaster models into a dynamic spatial computable general equilibrium model of Centerville Harvey Cutler, Martin Shields, Daniele Tavani & Sammy Zahran In orange the names of the presenting authors The special issue includes the following contributions
The Centerville Virtual Community: A Fully Integrated Decision Model of Interacting Physical and Social Infrastructure Systems Bruce R. Ellingwood, Harvey Cutler, Paolo Gardoni, walter Gills Peacock, John W. van de Lindt and Naiyu Wang 17 COE Semiannual Meeting 04/26/2017
Urban infrastructure 18
19 The Centerville Testbed Motivation • A simple community model with essential physical, social and economic infrastructure components and systems and interdependencies • Software developers at NCSA can begin coding IN-CORE modules while awaiting for more realistic and complex algorithms and datasets • The common community model requires engineers, economists and sociologists to begin working together immediately. Sustainable and Resilient Infrastructure, Vol. 1, Nos. 3-4, Sept-Dec, 2016.
Centerville Map 20
21 Population demographics Demographic Value Total Population 50,000 Total Households 19,684 Total male (proportion) 0.5 Total Female (proportion) 0.5 0 to 17 years (proportion) 0.19 18 to 64 years (proportion) 0.64 65 years and over (proportion) 0.17 Median Household Income $51,646 Total Owner Occupied (proportion) 0.5 Total renter Occupied (proportion) 0.5
Employment by sector 22 Sector Employees Employment (%) Retail 8087 25 Manufacturing 1923 6 Services 7921 25 Construction 2375 7 Health Care 3081 10 Education 589 2 P r o f e s s i o n a l Business Services 2795 9 Utilities 519 2 Government 4842 15 Total 32132 Health Care 9% Retail 25% Manufacturing 6% Services 25% Construction 7% Education 2% Professional Business Services 9% Utilities 2% Government 15%
23 Building Construction Occupancy ID Building Types Residential R1 SF, 1-story, Wood, 1,400 ft2, 1945-1970 R2 SF, 1-story, Wood, 2,400 ft2, 1985-2000 R3 SF, 2-story, Wood, 3,200 ft2, 1985-2000 R4 SF, 1-story with basement, Wood, 2,400 ft2, 1970-1985 R5 MF, 3-sto, 48-unit multi-family (12,000 ft2/floor), 1985 R6 SF, Mobile home Commercial/retail C1 1-story, Steel Braced frame, 50,000 ft2, 1980 C2 2-story, RC frame, 50,000 ft2, 1980 C3 2-story, Reinforced masonry, 25,000 ft2, 1960 C4 Tilt-up concrete 125,000 ft2, (similar to a Walmart or Home Depot), 1995 Industrial I1 2-story, steel braced frame, 100,000 ft2, 1975 I2 1-story, steel braced frame, 500,000 ft2, 1995 Special S1 4-story, Hospital, RC, 30,000 ft2/floor, 1980 S2 2-story, Fire Station, Reinf. masonry, 10,000 ft2, 1985 S3 3-story, School, RC, 100,000 ft2, 1990 S4 1-story, School, lightly reinforced masonry, 100,000 ft2
Centerville Roadway Network 24
Centerville Electrical Power Network 25
Centerville 26
Social and economic metrics 27 • Economic damages to physical infrastructure • Population dislocation • Loss of employment • Decline in household income/distribution • Change in domestic output
Peihui Lin Naiyu Wang, Ph.D. University of Oklahoma COE Semiannual Meeting 04/27/2017 Building Portfolio Analysis 28
Building Portfolio Analysis 29 Spatial and Temporal Hazzard Characterization At Community Scale Portfolio-level Damage Assessment Portfolio-level Functionality Loss Estimation Portfolio Recovery Time and Trajectory Modeling Risk-based decisions To Enhance Resilience of Community Building Portfolios inform Uncertainty And spatial correlation Functionality dependency on utility availabilities Engineering process and impact of social- economic systems Resilience objectives, gaps and mitigation strategies and policies Spatial and Temporal Hazzard Characterization At Community Scale Portfolio-level Damage Assessment Portfolio-level Functionality Loss Estimation Portfolio Recovery Time and Trajectory Modeling Challenges: 1. Uncertainty propagation and spatial correlation modeling 2. Building functionality assessment considering utility availabilities 3. Building portfolio recovery time and trajectory estimation 4. Ensuring scalability of analysis algorithms
Building Portfolio Performance Metrics 30 Portfolio Metrics Abbrev. Definition Immediately Occupancy Ratio IOR The percentage of a building portfolio that can provide safe occupancy immediately following a disaster Household Dislocation Ratio HDR The percentage of displaced households in the community due to loss of housing habitability Direct Loss Ratio DLR The ratio of total direct loss to total assessed value of a building portfolio Portfolio Recovery Index PRI The percentage of a buildings that are in any of the five predefined functionality states at any given time following a disaster Portfolio Recovery Time PRT The time takes for a target percent of buildings in the portfolio to regain their full functionality states Sustainable and Resilient Infrastructure, 1(3-4), 108 -122.
Testbed: Centerville 31 • 14,890 wood buildings in 7 residential zones • 151 steel and RC buildings in 2 retail and business zones • 70 steel buildings in 2 industrial zones • 19 RC and masonry buildings for schools, hospitals and fire stations Sustainable and Resilient Infrastructure, 1(3-4), 108 -122.
Damage and Functionality Loss 32 Damage to Functionality Mapping
Building Portfolio Recovery Model (Follow-up Work) 33 We developed a building portfolio recovery model (BPRM) which can output: • Functionality state probabilities at any time t following an event, for different building types; • Portfolio recovery time and recovery trajectory; • Uncertainties associated with the recovery trajectory. • Temporal evolution of the spatial variation in the portfolio recovery;
34 Special Issue of Sustainable and Resilient Infrastructure on the Centerville Testbed
Mitigation Resource Allocation 35 Sustainable and Resilient Infrastructure, 1(3-4), 123 -126. Allocate limited resources to retrofit building types in each zone, to specific code levels w.r.t. competing objectives: ‐ minimize economic loss (i.e., structure, non- structure, and content loss, provided by the Building team) ‐ minimize total dislocation (OLS model provided by the Social Science team) Requirement: ‐ Overall disparity in the dislocation rates by socio- economic status does not increase
Baseline 36 Sustainable and Resilient Infrastructure, 1(3-4), 123 -126. • Heterogeneous building types, code levels • Ideal budget ($346M) is sufficient to improve all buildings to highest code level • Magnitude 7.8 earthquake, 35km to southwest ‐ Total direct economic loss: $856M ‐ Total dislocation: 3,203 HH Evaluation of optimization results under three budget levels: ‐ Low: 15% of ideal budget ‐ Medium: 30% of ideal budget ‐ High: 60% of the ideal budget
Pareto Optimality 37 Sustainable and Resilient Infrastructure, 1(3-4), 123 -126. Budget Level: 15% ($52M) Policy I reduces expected loss by $80M and reduces dislocation by about 700 HH from baseline Policy II reduces expected loss by $125M and reduces dislocation by about 480 HH from baseline
38 Investment Decisions: Residential vs. Non-Residential Sustainable and Resilient Infrastructure, 1(3-4), 123 -126.
39 Pareto Optimal Frontier: Capital Stock Loss and Dislocation Sustainable and Resilient Infrastructure, 1(3-4), 123 -126. Zone1: W2: 1,072 buildings from level 2 to 3 W4: 2,196 buildings from level 1 to 2 Zone2: W1: 767 buildings from level 1 to 3 Zone4: W1: 2,567 buildings from level 1 to 3 W5: 25 buildings from level 2 to 3 Zone6: W5: 59 buildings from level 2 to 3
40 Special Issue of Sustainable and Resilient Infrastructure on the Centerville Testbed
41  It proposes a unified methodology for modeling dependent infrastructure networks and uses a 6-step probabilistic procedure to assess system resilience.  To estimate system recovery as a function of time at a community scale, the proposed methodology simulates direct physical damage, cascading effects due to dependencies, and recovery levels of network functionality.  The methodology for infrastructure resilience assessment is general, can be applied to other infrastructure networks and hazards, and can model different types of dependencies between networks.  As an illustration, the paper presents the direct physical effects of seismic events on the functionality of a potable water distribution network (WN) and the cascading effects of a damaged electric power network (EPN) on the WN. The paper makes the following contributions
42  The main diagonal of A has the symmetric adjacency tables of each network, that represent the pairwise connections within each network; the out-of-diagonal, generally rectangular, tables are used to represent pairwise connections between nodes of different networks  The likelihood table L describes the likelihood of failure of a node, given the failure of a different node  The dependency table P captures the failure propagation, combining tables A and L The dependency of networks is modeled through an augmented adjacency table A, a likelihood table L and a dependency table P
43 The 6-step probabilistic procedure to assess infrastructure resilience is as follows: Generate network modelStep 1 Generate hazard for the network areaStep 2 Assess direct physical damage to network componentsStep 3 Propagate cascading effects due to dependenciesStep 4 Assess functionality lossStep 5 Predict recovery time for network functionalityStep 6 Iterate to describe level of functionality within specified Tolerance
44  WN system is dependent on the EPN system  Insets A, B, and C show the WN and EPN node dependencies  The dependencies between WN and EPN are modeled with the out-of- diagonal adjacency tables which have values equal to 1 for the node pairs just described and zeroes for all of the other terms Centerville’s WN and EPN are used as testbed of the proposed procedure
45  Two cases are used to describe the one- way dependency of the WN on the EPN:  Perfect independence: the WN is decoupled from the EPN (WN)  Perfect dependence: failure at any EPN nodes fails the dependent WN node (WN + EPN)  The minimum values of the functionality metrics are lower for the WN + EPN than for the WN  Although recovery begins around the same time, it is slower for the dependent system (WN + EPN) than for the independent system (WN) The case study results show the role of dependencies in assessing network resilience
46  The loss of functionality and delay in the recovery process for network dependency can be quantified.  Simulating the dependency between networks can be important in estimating the resilience of critical infrastructure.  The cascading failures in the WN due to EPN failures can lead to increased recovery times or increased extent of functionality loss.  The modularity of the proposed methodology provides flexibility and allows the user to update fragilities, component and network models, hazards, and repair rates. Concluding Remarks
47 Figure 1: Flow Diagram of a Computable General equilibrium Model
48 East and West Centerville
49 Restrict the earthquake shock to hit East and West Centerville separately • Damages commercial and residential buildings • Understand the value of the spatial CGE model Dynamic outcomes • Allow the earthquake to hit both sides of Centerville simultaneously • Damages to ‐ Buildings and transportation network ‐ Workers and shoppers have to use alternative routes Motivation
50 Introduction and Outline
51 Dynamic Outcomes
52 • Spatial characteristics are important • Commuting in and out (Joplin) • Dynamic estimates will be used to model recovery
53 • Building portfolio analysis (Earthquake) • Electric Power Network (EPN) damage analysis (Tornado) • Water network recovery analysis with EPN dependency (Earthquake) • Mitigation Resources Allocation (MRA) analysis • Computable General Equilibrium (CGE) analysis IN-CORE v1 Demo for Centerville Analyses
54 • Uses IN-CORE scenario earthquake • Computes these for a designated zone in Centerville Building Portfolio Analysis • Immediately Occupancy Ratio • Direct Loss Ratio • Household Dislocation Ratio
55 Demo of Software!!!
56 • Uses IN-CORE scenario tornado • Computes damages of Electric power network EPN Damage Analysis
57 Demo of Software!!!
58 • Uses IN-CORE scenario earthquake • Water network recovery analysis with EPN dependency • Computes changes of avg. of pressure, demand, and contamination Water Network Recovery Analysis
59 Demo of Software!!!
60 • Multi-objective optimization • Computes solutions of resource allocation Mitigation Resource Allocation Analysis
61 Demo of Software!!!
62 • Uses IN-CORE scenario earthquake • Computes the impact of the earthquake through the building infrastructure and total factor productivity transportation channels Computable General Equilibrium Analysis
63 Demo of Software!!!
64 Contributors
65 Outline • 2011 Joplin MO Tornado • Field Study Data Collection & Mapping • Single Sector Validation: Electric Power Network • Single Sector Validation: Buildings • Multisector Validation: Buildings-EPN • Post-Disaster Functionality Methodology • Business Interruption • Socially Vulnerable Populations • Population Dislocation • Future Plans
66 Hindcasting Hindcasting is a systematic process by which we check the accuracy of mathematical (and other) models using a past event. For Joplin, this will be accomplished through a robust series of checks and hindcasts ranging from the single building or infrastructure component level, economic component, to single infrastructure sector validation – this will be done at points in time to validate damage modeling, loss estimation, population dislocation, and functionality and recovery models.
67 City of Joplin, Missouri Hindcast is a way of testing mathematical models; known or closely estimated inputs for past events, are entered into the model to see how well the output matches the known results. • Land area in square miles: 35.56 • Population: 51,316 (2014 estimate) • Housing units: 23,322 • Median value of housing: $103,300 Joplin tornado – May 22, 2011 • EF5 multiple-vortex • Fatalities: 161 • Injured: 1150 • Total loss: $2.8 billion • Costliest single tornado in US history
68 2011 Joplin Tornado Overview • Touched down at 5:34 PM CDT, Sunday, May 22, 2011.1 Stayed on ground for about 22 miles (6 miles in City of Joplin) and 15 minutes • Enhanced Fujita Scale EF-5 tornado; estimated maximum wind speeds: 200+ mph • Damaged/destroyed ~ 8,000 buildings. • Affected ~41% of City’s population (20,820 of 50,173). Costliest tornado on record (~$1.8 billion insured loss). • 161 fatalities and over 1,000 injuries. Deadliest single tornado on record. Exceeded U.S. average deaths/year for all tornados (91.6), hurricanes (50.8), & earthquakes (7.5). • Official warning time of 17 minutes (national average is 14 minutes).
69 NIST Objectives • Determine the tornado hazard characteristics and associated wind fields • Determine the response of residential, commercial, and critical buildings, including the performance of designated safe areas • Determine the performance of lifelines relative to the community of operations of residential, commercial, and critical buildings • Determine the pattern, location, and cause of fatalities and injuries, and associated emergency communications and public response • Identify areas in current building, fire, and emergency communications codes, standards, and practices that warrant revision
70 NIST Findings • 40% of the fatalities and as much as 90 % of the tornado area were associated with EF-3 or lower wind speeds. • Of the buildings damaged, 7,411 were residential and 553 were non-residential. • The failure of building envelopes at St. John’s Regional Medical Center led to its loss of functionality; the structural frame withstood the tornado without structural collapse. • All utilities (water, gas, power) were lost in the areas most damaged by the May 22. The utility providers restored service to critical buildings (SJRMC, water treatment plant) within 24 hours.
71 Field Study Data Collection & Mapping
72 Single Sector Validation: Electric Power Network
73 Estimating Service Area of Each Substation Based on Weighted Cellular Automata Combined with Distribution Lines
74 EPN Damage Analysis Results
75 Categorizing Buildings in Joplin Single Sector Validation: Buildings Developing Damage Fragilities Bldg. ID Building Description T1 Res Wood Bldg. – Small Rectangular Plan – Gable Roof - 1 Story T2 Res Wood Bldg. – Small Square Plan – Gable Roof - 2 Stories T3 Res Wood Bldg. – Medium Rectangular Plan – Gable Roof - 1 Story T4 Res Wood Bldg. – Medium Rectangular Plan – Hip Roof - 2 Stories T5 Res Wood Bldg. – Large Rectangular Plan – Gable Roof - 2 Stories T6 Business and Retail Buildings T7 Light Industrial Buildings T8 Heavy Industrial Buildings T9 Elementary/Middle School – Unreinforced Masonry T10 High School – Reinforced Masonry T11 Fire/Police Station T12 Hospital T13 Community Center/Church T14 Government Buildings T15 Large Big Box T16 Small Big Box T17 Mobile Homes T18 Shopping Mall T19 Office Buildings
76 Building Taxonomy for Tornado Hazard Derivation Resources: a) Building Attributes Affecting Structural Response [GEM Building Taxonomy (2013)] a) Building Attributes Affecting Wind Load Intensity [ASCE7-10 chapters 26 to 31]
77 Building Damage Analysis Results
78 Building Damage Analysis Results
79 Multisector Validation: Buildings-EPN Combining Building and Electric Power Network
80 Functionality of building Based on Damage to Buildings and Damage to Electric Power Network
81 Post-Disaster Functionality Methodology • Developed for: • Any type of building (i.e. residential buildings, health care facilities, industrial, commercial etc. • Any hazard (hazard agnostic) • Integrates Performance- Based Engineering (PBE) into Community Resilience Assessment studies • Four-step probabilistic functionality methodology
82 Post-Disaster Functionality Methodology • Four damage states (Slight, moderate, extensive & complete) • Three different building types: school, heavy industrial & strip mall • Output: • Intrinsic Functionality Fragilities (IFF) accounts only for construction and clean-up time • Semi-intrinsic Functionality Fragilities (SIFF) accounts for construction and clean-up time & time to obtain financing, permits and complete design • Results presented for the assumption of being in Damage State “Extensive” • Results format: Empirical and log-normal fitted cumulative distribution curves
83 Business Interruption Model
84 Estimating Probability of Business Closure After 3 Months
85 Estimating Business Interruption
86 Influences of the Event on Economic Factors • A social accounting matrix (SAM) summarizes the data that reflects the intersection between households, firms and the government. • SAMs will be constructed for 2010, 2011 and 2012 for Joplin • CGE model for 2010 and then use it to forecast future SAM values and compare these forecasts to the values in the 2011 and 2012 SAMs. • The forecast errors (Actual – Forecast) will determine initial ability to forecast the impacts of a tornado.
87 Socially Vulnerable Populations ‐ Census Tract Level example for the city ‐ Population Data Geocoded within Census Tracts ‐ Census Block Group Level example for Joplin Missouri ‐ Population Data Geocoded within Census Block Groups
88 Socially Vulnerable Populations ‐ 90% Confidence Interval Population Estimates for population data within the EF0-EF4 2011 Joplin Tornado Path ‐ Block Group Level provides larger estimates due to variation in population density ‐ Census Tract Level estimates underestimate vulnerable populations Factors Census Tract Estimate Block Groups Estimate Percent Difference Total Population 18,325±1,001 20,168±1,535 7.6% - 12.3% 65 years and over 2,698±326 3,130±365 15.6 - 16.5% Single-parent with own children 900±234 1,069±270 18.0% - 20.0% Occupied Housing Units without a vehicle 502±144 628±182 24.8% - 25.5% Total Housing Units with 10 or more units 540±213 661±230 18.3% - 31.8%
89 Future Plans • One-way dependencies beyond EPN-buildings • Recovery functions; functionality, permitting • Systematically test each of the algorithms presented in the SRI Special Issue portion of the webinar • Apply the CGE model in one year time steps • Predict population dislocation • Systematically apply decisions made by Joplin leadership over time
90
91 Field Studies Thrust in the CoE • In order to measure a community’s resilience, the CoE will collect practical metrics that can be divided into two categories: 1) Metrics that require field study data: a) Population dislocation: Households displaced b) Business interruption: Business closed c) Employee dislocation: Employees failing to report to work d) Critical facilities impact: Critical facilities closed e) Housing loss: Units lost 2) Metrics that support (but do not require) field study data: a) Physical and mental morbidity b) Mortality c) Fiscal impact: Loss of property tax and sales tax • The community dimensions that are described in NIST GCR 16-001 are embedded into these metrics ‐ Dimensions include: sustenance, health, housing and shelter, education and personal development, security and safety, culture and identity, and belonging and relationships
92 Disaster and Failure Studies (DFS) at NIST Statutory Arm • Evaluate hazard events against deployment criteria – new knowledge or national impact • Federal Advisory Committee (NCSTAC) • Conduct reconnaissance, field studies, or investigations • MOUs with other agencies Standard Operating Procedures • Study objectives • Field and safety protocols • Human subjects protocols (IRB, PRA) • Equipment for data collection and personnel safety • Data preservation and management Research Arm • Develop research program focused on disaster metrology • Coordination with NIST groups • Coordination with NIST’s CoE • Coordination with other federal agencies and research centers • Outreach and committee work
93 CoE/NIST Collaborative Field Studies • Data collection ‐ collect data and establish likely technical factors responsible for poor/successful performance of buildings and infrastructure in the aftermath of disasters. ‐ collect data related to community impact and recovery. ‐ collect data to validate models under development. • Field methods ‐ test new field sampling protocols. ‐ assess new field equipment. • Findings ‐ recommend frequency of data collection to capture community functioning over time. ‐ recommend, as necessary, specific improvements to standards, codes, and practices based on field studies.
94 Hurricane Matthew Overview • Made landfall Oct 8, 2016 as a Cat 1 (75 mph sustained winds), in South Carolina • 27 deaths in NC; more than 2,300 people saved in over 600 rescue operations • Thousands displaced; 42 schools closed for 3 weeks • $1.5 B in damage losses; 900k without power during peak loss • Severe transportation network disruptions (including I-95 & I-40)
95 Robeson County (Lumberton) Impacts • Demographics: 40% white; 37 percent black; 13% Native American (Lumbee Indian) • Per capita income was $15,321 • 1/3 population in poverty • 5,000 displaced • 7,057 structures affected • >5,000 power customer outages • 133 roads impacted or damaged • 4 public housing developments damaged, totaling 545 units
96 Field Deployment Research Goals 1) Assess damage to school buildings and impact on educational system (e.g., school closures, food programs, other services); 2) Study how the floods impacted housing stock (collect empirical data to compare to existing fragility curves) and population displacement; 3) Investigate the dependency of these important community functions on distributed infrastructure networks (e.g., power, water, wastewater, etc.).
97 Field Study Team-Lumberton, NC
98 Field Study Team-Lumberton, NC
99 Implementing Field Research Tools • Engineering ‐ Damage assessment of a random sample of housing units (HUs) to inform fragility curves • Social Science ‐ Household interviews of a random sample of Hus/ households to inform general population models • Engineering + Social Science ‐ Qualitative survey of community leaders and stakeholders to understand impact, response, and recovery planning and activities
100 IRB Field Study Protocol • Institutional Review Board (IRB) approval from all participating Universities ‐ Develop protocols for semi-structured and structured interviews, damage assessment, and photography • Recruitment of interviewees • Informed consent • Data collection methodologies and instruments • Types of data collected • Photography: publically viewable versus internal photos and release statements
101 Random sample of Housing Units/Households Target area was the largest school attendance zone (Lumberton Jr. High) Developed sampling scheme to capture: 1. Both heavily and lightly/ none flooded areas 2. Obtain random representative sample of residential housing units and household 3. Stratified two stage random sampling procedure based on census blocks and housing units.
102 Used Google Mapping tools to guide teams To facilitate teams finding the sampled Hus we employed Google Maps (Google My Maps) that could be view on smart phones.
103 Used Google Mapping tools to guide teams The sample linked to a Google Doc Spreadsheet – providing information on each HU and updated each evening on status of damage assessment and HH interview.
104 Daily Activities • Briefing on the day’s activities and goals • Division into qualitative and survey sampling teams • Assignments of areas or stakeholders/meetings to be attended • Lunch time check in, update, and assessment • Evening check in: • Debriefing • update of sample/survey activities • initial development of next day’s activities and assignments • Set-up for next day
105 Damage Assessment Data • Most of the damaged homes are wood light-frame structures (many with brick veneer). • Average quality and typically one to two stories. • Almost two-thirds have crawlspaces. • Generally single floor • Majority single family: multi-family often duplexes or triplexes. Foundation Type (with Number of Buildings) Building Type Distribution of Buildings (%) Distribution of Number of Stories (%) Distribution of Construction Type (%) One Two Wood Masonry Both Crawlspace (273) Single 96.3 90.1 8.4 60.0 28.1 8.0 Multi 1.8 100 0.0 80.0 0.0 20.0 Slab (116) Single 33.6 89.7 10.3 56.4 25.6 15.4 Multi 64.7 86.7 12.0 41.3 29.3 18.7
106 Damage Assessment Data DS Level Description 0 No damage although water enters crawlspace or touches foundation (crawlspace or slab on grade). No contact to electrical or plumbing, etc. in crawlspace. No contact with floor joists. No sewer backup into living area. 1 Minor water enters house; damage to carpets, pads, baseboards, flooring. Approximately 1”, but no drywall damage. Touches joists. Could have some mold on subfloor above crawlspace. Could have minor sewer backup and/or minor mold issues. 2 Drywall damage up to approximately 2 feet and electrical damage, heater and furnace and other major equipment on floor damaged. Lower bathroom and kitchen cabinets damaged. Doors or windows need replacement. Could have major sewer backup and/or major mold issues. 3 Substantial drywall damage, electrical panel destroyed, bathroom/kitchen cabinets and appliances damaged; lighting fixtures on walls destroyed; ceiling lighting may be ok. Studs reusable; some may be damaged. Could have major sewer backup and/or major mold issues. 4 Significant structural damage present; all drywall, appliances, cabinets etc. destroyed. Could be floated off foundation. Building must be demolished or potentially replaced.
107 Empirical Flood-Damage Fragilities Next step is to include social factors into engineering fragilities!
108 Household Dislocation, Damage, and Socio-Economic Data • There was considerable dislocation and that dislocation was clearly a function of damage • We also see variations with respect to tenure, race/ethnicity, and income Household Dislocation Total Sample Damage State DS0 DS1 DS2 DS3+ Yes 48.6% 25.0% 71.5% 90.4% 94.2% No 51.4% 75.0% 28.5% 9.6% 5.8% Household Dislocation Tenure Household Racial Status Renter Owner White Black Native Am. Yes 68.5% 48.3% 26.5% 74.2% 84.1% No 31.5% 51.7% 73.5% 25.8% 15.9%
109 Consideration of Socio-Economic Factors
110 Considering Damage and Socio-Economic Factors can be Important: two examples • Here we see that the probabilities of household dislocation vary not only by damage, but also by tenure and race.
111 Where are we Headed • Continue analyses ‒ linking Engineering and social science.. • Consequences of infrastructure disruption • Dislocation time • Aid and initial recovery efforts ‒ Undertake additional data collection activities to assess resiliency
Thank You

More Related Content

PPTX
Community Resilience Planning, Stephen Cauffman
byGlobal Risk Forum GRFDavos
 
PPTX
Supporting community resilience (public)
byMd.Asif Rahman
 
PDF
How future cities will succeed
byBob Prieto
 
PDF
Christopher Rogers
byRuthMcIlmoyleUKCRIC
 
PDF
Strengthening the resilience of critical infrastructure - Michael Schack, ENGIE
byOECD Governance
 
PDF
Professor Brian Collins
byRoyal United Services Institute for Defence and Security Studies
 
PPTX
Resilience ortenzi fp_fb_lg
byStroNGER2012
 
PPTX
Bolsterning National & Global Resilience in the Face of 21st Century Mayhem
byNU_Seattle
 
Community Resilience Planning, Stephen Cauffman
byGlobal Risk Forum GRFDavos
 
Supporting community resilience (public)
byMd.Asif Rahman
 
How future cities will succeed
byBob Prieto
 
Christopher Rogers
byRuthMcIlmoyleUKCRIC
 
Strengthening the resilience of critical infrastructure - Michael Schack, ENGIE
byOECD Governance
 
Professor Brian Collins
byRoyal United Services Institute for Defence and Security Studies
 
Resilience ortenzi fp_fb_lg
byStroNGER2012
 
Bolsterning National & Global Resilience in the Face of 21st Century Mayhem
byNU_Seattle
 

Similar to Community resilience modeling, field studies and implementation

PPTX
Resilience Metrics and Approaches for Quantification, Igor LINKOV
byGlobal Risk Forum GRFDavos
 
PDF
Routledge Handbook of Sustainable and Resilient Infrastructure Paolo Gardoni
byrmzljlxvg8087
 
PPTX
South Norwalk Lexington & South Main Community Development Plan Public Meetin...
byNorwalkRedevelopment
 
PDF
Resilient Parks & Open Space in the Southwest
byDekker/Perich/Sabatini
 
PPT
Understanding and Measuring Urban Resilience: A new UN-Habitat's initiative
byGlobal Risk Forum GRFDavos
 
PDF
BIM Technology_RESTORE WG4 Meeting Riga_F.Bejtullahu (2019_06)
byRESTORE
 
PPTX
ICCAUA2025EN0407_CONFERENCE POWER POINT PRESENTATION_.pptx
byopagboola
 
PDF
Invest in Infrastructure throughout its Life Cycle
byBob Prieto
 
PDF
Strengthening the resilience of critical infrastructure - Matt Crossman, Unit...
byOECD Governance
 
PPTX
City_politics_and_the_environment_g2[1].pptx
byRoxanneMendez2
 
PPTX
Green Infrastructure for Regional Resilience
byArbor Day Foundation
 
PDF
The Resilience Shift - A short Introduction
byThe Resilience Shift
 
PDF
Assessing the resilience of a city in relation to its healthy urban systems: ...
byDr.Hayam alsa'atee
 
PDF
Assessing the resilience of a city in relation to its healthy urban systems: ...
byDr.Hayam alsa'atee
 
PDF
Designing for Resilience as a New Nuclear Safety Construct
byBAyyub
 
PDF
Resilience in Spatial and Urban Systems
byRegional Science Academy
 
PDF
Asce tdi-2016-01-10-public infrastructure asset management system m-johnson
byShawn Ghadimi
 
PPTX
Data 2.0: North Port - Section One - Introduction
byAllisonPinto
 
PDF
SMART International Symposium for Next Generation Infrastructure: Engineering...
bySMART Infrastructure Facility
 
PPTX
Major Projects Association - Shifting major project practices to safeguard cr...
byThe Resilience Shift
 
Resilience Metrics and Approaches for Quantification, Igor LINKOV
byGlobal Risk Forum GRFDavos
 
Routledge Handbook of Sustainable and Resilient Infrastructure Paolo Gardoni
byrmzljlxvg8087
 
South Norwalk Lexington & South Main Community Development Plan Public Meetin...
byNorwalkRedevelopment
 
Resilient Parks & Open Space in the Southwest
byDekker/Perich/Sabatini
 
Understanding and Measuring Urban Resilience: A new UN-Habitat's initiative
byGlobal Risk Forum GRFDavos
 
BIM Technology_RESTORE WG4 Meeting Riga_F.Bejtullahu (2019_06)
byRESTORE
 
ICCAUA2025EN0407_CONFERENCE POWER POINT PRESENTATION_.pptx
byopagboola
 
Invest in Infrastructure throughout its Life Cycle
byBob Prieto
 
Strengthening the resilience of critical infrastructure - Matt Crossman, Unit...
byOECD Governance
 
City_politics_and_the_environment_g2[1].pptx
byRoxanneMendez2
 
Green Infrastructure for Regional Resilience
byArbor Day Foundation
 
The Resilience Shift - A short Introduction
byThe Resilience Shift
 
Assessing the resilience of a city in relation to its healthy urban systems: ...
byDr.Hayam alsa'atee
 
Assessing the resilience of a city in relation to its healthy urban systems: ...
byDr.Hayam alsa'atee
 
Designing for Resilience as a New Nuclear Safety Construct
byBAyyub
 
Resilience in Spatial and Urban Systems
byRegional Science Academy
 
Asce tdi-2016-01-10-public infrastructure asset management system m-johnson
byShawn Ghadimi
 
Data 2.0: North Port - Section One - Introduction
byAllisonPinto
 
SMART International Symposium for Next Generation Infrastructure: Engineering...
bySMART Infrastructure Facility
 
Major Projects Association - Shifting major project practices to safeguard cr...
byThe Resilience Shift
 

More from Md.Asif Rahman

PPT
Graduate Showcase
byMd.Asif Rahman
 
PPTX
ME 570 Finite Element Methods
byMd.Asif Rahman
 
PPTX
ME 510 Continuum Mechanics
byMd.Asif Rahman
 
PPTX
CE 502 Computational Techniques
byMd.Asif Rahman
 
PDF
RETROFITTING
byMd.Asif Rahman
 
PPTX
ONE WAY SLAB DESIGN
byMd.Asif Rahman
 
Graduate Showcase
byMd.Asif Rahman
 
ME 570 Finite Element Methods
byMd.Asif Rahman
 
ME 510 Continuum Mechanics
byMd.Asif Rahman
 
CE 502 Computational Techniques
byMd.Asif Rahman
 
RETROFITTING
byMd.Asif Rahman
 
ONE WAY SLAB DESIGN
byMd.Asif Rahman
 

Recently uploaded

PPTX
literary theory and criticism by Vivek p
byvivekrathodborda
 
PDF
Schrodinger's Capital Finals (SciBiz Quiz).pdf
byConquiztadors- the Quiz Society of Sri Venkateswara College
 
PDF
Chapter 05 Drug Acting on CNS Sedatives & Hypnotics | Antipsychotics.pdf
bySandeshSul
 
PPT
West Hatch High School - GCSE History Option
byWestHatch
 
PPTX
West Hatch High School - GCSE Spanish Presentation
byWestHatch
 
PPTX
Metabolism ( BIOCHEMISTRY & CLINICAL PATHOLOGY )
byBushraDeshpande
 
PPTX
Drug Distribution in Pharmacology: Mechanisms, Barriers, Factors & Volume of ...
byamrin3379
 
PDF
Q4_LE_English 5_Lesson 1_Week 1.pdf learning instruction
byMichelleCajefe2
 
PPTX
Mohan - "A man of silence and authority"
bynidhibachauhan46
 
PDF
Information about the author Shashi Deshpande
bydarshanaparmar6522
 
PPTX
Toba Tek Singh - Visualising Partition through Manto's Lens
bymiraljoshi48
 
PPTX
How to Create & Configure Rewards in Odoo 18 Referrals
byCeline George
 
PDF
APPSC APPSC AEE-AE GENERAL STUDIES QUESTION PAPER.pdf
byssuser9d8e4e
 
PPTX
Biological source, chemical constituents, and therapeutic efficacy of the fol...
bySai Meer College of Pharmacy
 
PDF
West Hatch High School - GCSE Media Specification
byWestHatch
 
PDF
West Hatch High School - GCSE French Specification
byWestHatch
 
PDF
Sharad Bisen_Soil Sterilization_Objectives_ICAR Course, Sixth Dean Committee.pdf
bybisensharad
 
PPTX
Appreciations - Jan 26 01.pptxkkkmmkmkmkmkm
bypreetheshparmar
 
PDF
CAM, DHT11, GAS, ESP32, FLAME, IR, TEMP, LCD, PIR, SOIL, SOUND, RELAY, TURBID...
byPravinDhanrao3
 
PPTX
"Aristotle : Father Of Western Philosophy"
bymitalbarathod28
 
literary theory and criticism by Vivek p
byvivekrathodborda
 
Schrodinger's Capital Finals (SciBiz Quiz).pdf
byConquiztadors- the Quiz Society of Sri Venkateswara College
 
Chapter 05 Drug Acting on CNS Sedatives & Hypnotics | Antipsychotics.pdf
bySandeshSul
 
West Hatch High School - GCSE History Option
byWestHatch
 
West Hatch High School - GCSE Spanish Presentation
byWestHatch
 
Metabolism ( BIOCHEMISTRY & CLINICAL PATHOLOGY )
byBushraDeshpande
 
Drug Distribution in Pharmacology: Mechanisms, Barriers, Factors & Volume of ...
byamrin3379
 
Q4_LE_English 5_Lesson 1_Week 1.pdf learning instruction
byMichelleCajefe2
 
Mohan - "A man of silence and authority"
bynidhibachauhan46
 
Information about the author Shashi Deshpande
bydarshanaparmar6522
 
Toba Tek Singh - Visualising Partition through Manto's Lens
bymiraljoshi48
 
How to Create & Configure Rewards in Odoo 18 Referrals
byCeline George
 
APPSC APPSC AEE-AE GENERAL STUDIES QUESTION PAPER.pdf
byssuser9d8e4e
 
Biological source, chemical constituents, and therapeutic efficacy of the fol...
bySai Meer College of Pharmacy
 
West Hatch High School - GCSE Media Specification
byWestHatch
 
West Hatch High School - GCSE French Specification
byWestHatch
 
Sharad Bisen_Soil Sterilization_Objectives_ICAR Course, Sixth Dean Committee.pdf
bybisensharad
 
Appreciations - Jan 26 01.pptxkkkmmkmkmkmkm
bypreetheshparmar
 
CAM, DHT11, GAS, ESP32, FLAME, IR, TEMP, LCD, PIR, SOIL, SOUND, RELAY, TURBID...
byPravinDhanrao3
 
"Aristotle : Father Of Western Philosophy"
bymitalbarathod28
 

Community resilience modeling, field studies and implementation

  • 1.
    Community Resilience: Modeling,Field Studies, and Implementation Welcome! 1
  • 2.
    2 Thrust 1: AMultidisciplinary computational environment with fully integrated supporting databases: “The Interconnected Networked - Community Resilience Modeling Environment”. Thrust 2: “Data management tools for community resilience systems”: A standardized data ontology, robust data architecture, and effective tools to support IN-CORE. Thrust 3: Resilience Data Architecture Validation Studies: Hindcasts and forecasts to test the data collection process and its integration into IN-CORE. Validate risk-informed intelligent search and decision algorithms; field studies. The Center Thrusts
  • 3.
    3 • Quantitative toolsand metrics for assessing community resilience either are in a rudimentary state of development or are not available. • Multiple hazards, interconnected and inter-dependent systems must be considered. • Uncertainties, both aleatory and epistemic, must be included for the resilience assessment and decision process to be risk- informed. • Community resilience plans and guidance are needed. • Existing building and infrastructure systems, as well as new construction, must be considered. • Codes, standards and regulatory documents must contain consistent performance goals. Center research Issues
  • 4.
    4 How can communityleaders know how resilient their community is? How can they know if their decisions and investments to improve resilience are making a significant difference? How can they deal with uncertainty and risk? Research and Implementation Questions
  • 5.
    5 Thrust 1: TheDevelopment of IN-CORE
  • 6.
    6 • General andRecovery • Infrastructure performance • Spatial and logical distribution of physical damage • Time to functionality • Availability of resources and leadership • Dependencies among built environment and social institutions Community Resilience Metrics • Economic and Social • Retaining businesses and employment • Revenue resources • Poverty and income distribution • Economic sustainability • Survival – safety; water/ food • Safety and security – personal safety, financial security, health • Growth and achievement opps
  • 7.
    7 • User requirements/StakeholderAlignment ‒ Support for new data types being developed ‒ Building taxonomies for all hazards • Standardized Data Ontology ‒ Defining new data types such as need for CGE development ‒ Review of schema and metadata • Development of Data management Tools ‒ GEM taxonomy, DDI Codebook, lifecycle Thrust 2: A Robust Data Architecture
  • 8.
    8 • Systematic validationof the numerical models and data in Tasks 1 and 2. • 2011 Joplin, Missouri tornado • Lumberton, NC Field Study • Single sector, multi-sector, and full architectural validation Thrust 3: Resilience Architecture Validation Studies
  • 9.
    9 • Year 1:Team building and Hazards Center Themes by Year
  • 10.
    10 • Year 2:Dependencies and Testbeds Center Themes by Year
  • 11.
    11 • Year 3:Recovery modeling with full interdependencies • Year 4: Risk-Informed Decision-Making • Year 5: Technical outreach including user workshops Center Themes by Year
  • 12.
  • 13.
    13 Special Issue ofSustainable and Resilient Infrastructure on the Centerville Testbed
  • 14.
    14 Several testbeds havebeen initiated since the beginning of the Center. The testbeds have been designed to 1. allow Center research teams to initiate, test, and modify essential community resilience assessment models and algorithms before the IN-CORE platform becomes fully operational 2. stress these assessment models in a controlled manner 3. examine varying degrees of dependency between physical, social, and economic infrastructure systems 4. facilitate the interdisciplinary collaborations and approaches to community resilience assessment that will be essential for the Center to achieve its ultimate goal Background
  • 15.
    15  A specialissue of the journal Sustainable and Resilient Infrastructure (volume 1, 2016 – Issue 3-4) has been devoted to the first of these testbeds namely the Centerville Virtual Community  Sustainable and Resilient Infrastructure is and interdisciplinary journal published by Taylor and Francis that focuses specifically on the sustainable development of resilient communities Special issue on the Centerville Virtual Community
  • 16.
    16  The CentervilleVirtual Community: A fully integrated decision model of interacting physical and social infrastructure systems Bruce R. Ellingwood, Harvey Culter, Paolo gardoni, Walter gills Peacock, john W. van de Lindt & Naiyu Wang  Building portfolio fragility functions to support scalable community resilience assessment Peihui Lin & Naiyu Wang  A multi-objective optimization model for retrofit strategies to mitigate direct economic loss and population dislocation Weili Zhang & Charles Nicholson  Probabilistic framework for performance assessment of electrical power networks to tornadoes Vipin U. Unnikrishnan & John W. van de Lindt  Modeling the resilience of critical infrastructure: The role of network dependencies Roberto Guidotti, hana Chmielewski, Vipin U. Unnikrishnan , Paolo Gardoni, Therese McAllister & John van de Lindt  Integrating engineering outputs from natural disaster models into a dynamic spatial computable general equilibrium model of Centerville Harvey Cutler, Martin Shields, Daniele Tavani & Sammy Zahran In orange the names of the presenting authors The special issue includes the following contributions
  • 17.
    The Centerville VirtualCommunity: A Fully Integrated Decision Model of Interacting Physical and Social Infrastructure Systems Bruce R. Ellingwood, Harvey Cutler, Paolo Gardoni, walter Gills Peacock, John W. van de Lindt and Naiyu Wang 17 COE Semiannual Meeting 04/26/2017
  • 18.
  • 19.
    19 The Centerville Testbed Motivation •A simple community model with essential physical, social and economic infrastructure components and systems and interdependencies • Software developers at NCSA can begin coding IN-CORE modules while awaiting for more realistic and complex algorithms and datasets • The common community model requires engineers, economists and sociologists to begin working together immediately. Sustainable and Resilient Infrastructure, Vol. 1, Nos. 3-4, Sept-Dec, 2016.
  • 20.
  • 21.
    21 Population demographics Demographic Value TotalPopulation 50,000 Total Households 19,684 Total male (proportion) 0.5 Total Female (proportion) 0.5 0 to 17 years (proportion) 0.19 18 to 64 years (proportion) 0.64 65 years and over (proportion) 0.17 Median Household Income $51,646 Total Owner Occupied (proportion) 0.5 Total renter Occupied (proportion) 0.5
  • 22.
    Employment by sector 22 SectorEmployees Employment (%) Retail 8087 25 Manufacturing 1923 6 Services 7921 25 Construction 2375 7 Health Care 3081 10 Education 589 2 P r o f e s s i o n a l Business Services 2795 9 Utilities 519 2 Government 4842 15 Total 32132 Health Care 9% Retail 25% Manufacturing 6% Services 25% Construction 7% Education 2% Professional Business Services 9% Utilities 2% Government 15%
  • 23.
    23 Building Construction Occupancy IDBuilding Types Residential R1 SF, 1-story, Wood, 1,400 ft2, 1945-1970 R2 SF, 1-story, Wood, 2,400 ft2, 1985-2000 R3 SF, 2-story, Wood, 3,200 ft2, 1985-2000 R4 SF, 1-story with basement, Wood, 2,400 ft2, 1970-1985 R5 MF, 3-sto, 48-unit multi-family (12,000 ft2/floor), 1985 R6 SF, Mobile home Commercial/retail C1 1-story, Steel Braced frame, 50,000 ft2, 1980 C2 2-story, RC frame, 50,000 ft2, 1980 C3 2-story, Reinforced masonry, 25,000 ft2, 1960 C4 Tilt-up concrete 125,000 ft2, (similar to a Walmart or Home Depot), 1995 Industrial I1 2-story, steel braced frame, 100,000 ft2, 1975 I2 1-story, steel braced frame, 500,000 ft2, 1995 Special S1 4-story, Hospital, RC, 30,000 ft2/floor, 1980 S2 2-story, Fire Station, Reinf. masonry, 10,000 ft2, 1985 S3 3-story, School, RC, 100,000 ft2, 1990 S4 1-story, School, lightly reinforced masonry, 100,000 ft2
  • 24.
  • 25.
  • 26.
  • 27.
    Social and economicmetrics 27 • Economic damages to physical infrastructure • Population dislocation • Loss of employment • Decline in household income/distribution • Change in domestic output
  • 28.
    Peihui Lin Naiyu Wang,Ph.D. University of Oklahoma COE Semiannual Meeting 04/27/2017 Building Portfolio Analysis 28
  • 29.
    Building Portfolio Analysis 29 Spatialand Temporal Hazzard Characterization At Community Scale Portfolio-level Damage Assessment Portfolio-level Functionality Loss Estimation Portfolio Recovery Time and Trajectory Modeling Risk-based decisions To Enhance Resilience of Community Building Portfolios inform Uncertainty And spatial correlation Functionality dependency on utility availabilities Engineering process and impact of social- economic systems Resilience objectives, gaps and mitigation strategies and policies Spatial and Temporal Hazzard Characterization At Community Scale Portfolio-level Damage Assessment Portfolio-level Functionality Loss Estimation Portfolio Recovery Time and Trajectory Modeling Challenges: 1. Uncertainty propagation and spatial correlation modeling 2. Building functionality assessment considering utility availabilities 3. Building portfolio recovery time and trajectory estimation 4. Ensuring scalability of analysis algorithms
  • 30.
    Building Portfolio PerformanceMetrics 30 Portfolio Metrics Abbrev. Definition Immediately Occupancy Ratio IOR The percentage of a building portfolio that can provide safe occupancy immediately following a disaster Household Dislocation Ratio HDR The percentage of displaced households in the community due to loss of housing habitability Direct Loss Ratio DLR The ratio of total direct loss to total assessed value of a building portfolio Portfolio Recovery Index PRI The percentage of a buildings that are in any of the five predefined functionality states at any given time following a disaster Portfolio Recovery Time PRT The time takes for a target percent of buildings in the portfolio to regain their full functionality states Sustainable and Resilient Infrastructure, 1(3-4), 108 -122.
  • 31.
    Testbed: Centerville 31 • 14,890wood buildings in 7 residential zones • 151 steel and RC buildings in 2 retail and business zones • 70 steel buildings in 2 industrial zones • 19 RC and masonry buildings for schools, hospitals and fire stations Sustainable and Resilient Infrastructure, 1(3-4), 108 -122.
  • 32.
    Damage and FunctionalityLoss 32 Damage to Functionality Mapping
  • 33.
    Building Portfolio RecoveryModel (Follow-up Work) 33 We developed a building portfolio recovery model (BPRM) which can output: • Functionality state probabilities at any time t following an event, for different building types; • Portfolio recovery time and recovery trajectory; • Uncertainties associated with the recovery trajectory. • Temporal evolution of the spatial variation in the portfolio recovery;
  • 34.
    34 Special Issue ofSustainable and Resilient Infrastructure on the Centerville Testbed
  • 35.
    Mitigation Resource Allocation 35 Sustainableand Resilient Infrastructure, 1(3-4), 123 -126. Allocate limited resources to retrofit building types in each zone, to specific code levels w.r.t. competing objectives: ‐ minimize economic loss (i.e., structure, non- structure, and content loss, provided by the Building team) ‐ minimize total dislocation (OLS model provided by the Social Science team) Requirement: ‐ Overall disparity in the dislocation rates by socio- economic status does not increase
  • 36.
    Baseline 36 Sustainable and ResilientInfrastructure, 1(3-4), 123 -126. • Heterogeneous building types, code levels • Ideal budget ($346M) is sufficient to improve all buildings to highest code level • Magnitude 7.8 earthquake, 35km to southwest ‐ Total direct economic loss: $856M ‐ Total dislocation: 3,203 HH Evaluation of optimization results under three budget levels: ‐ Low: 15% of ideal budget ‐ Medium: 30% of ideal budget ‐ High: 60% of the ideal budget
  • 37.
    Pareto Optimality 37 Sustainable andResilient Infrastructure, 1(3-4), 123 -126. Budget Level: 15% ($52M) Policy I reduces expected loss by $80M and reduces dislocation by about 700 HH from baseline Policy II reduces expected loss by $125M and reduces dislocation by about 480 HH from baseline
  • 38.
    38 Investment Decisions: Residentialvs. Non-Residential Sustainable and Resilient Infrastructure, 1(3-4), 123 -126.
  • 39.
    39 Pareto Optimal Frontier:Capital Stock Loss and Dislocation Sustainable and Resilient Infrastructure, 1(3-4), 123 -126. Zone1: W2: 1,072 buildings from level 2 to 3 W4: 2,196 buildings from level 1 to 2 Zone2: W1: 767 buildings from level 1 to 3 Zone4: W1: 2,567 buildings from level 1 to 3 W5: 25 buildings from level 2 to 3 Zone6: W5: 59 buildings from level 2 to 3
  • 40.
    40 Special Issue ofSustainable and Resilient Infrastructure on the Centerville Testbed
  • 41.
    41  It proposesa unified methodology for modeling dependent infrastructure networks and uses a 6-step probabilistic procedure to assess system resilience.  To estimate system recovery as a function of time at a community scale, the proposed methodology simulates direct physical damage, cascading effects due to dependencies, and recovery levels of network functionality.  The methodology for infrastructure resilience assessment is general, can be applied to other infrastructure networks and hazards, and can model different types of dependencies between networks.  As an illustration, the paper presents the direct physical effects of seismic events on the functionality of a potable water distribution network (WN) and the cascading effects of a damaged electric power network (EPN) on the WN. The paper makes the following contributions
  • 42.
    42  The maindiagonal of A has the symmetric adjacency tables of each network, that represent the pairwise connections within each network; the out-of-diagonal, generally rectangular, tables are used to represent pairwise connections between nodes of different networks  The likelihood table L describes the likelihood of failure of a node, given the failure of a different node  The dependency table P captures the failure propagation, combining tables A and L The dependency of networks is modeled through an augmented adjacency table A, a likelihood table L and a dependency table P
  • 43.
    43 The 6-step probabilisticprocedure to assess infrastructure resilience is as follows: Generate network modelStep 1 Generate hazard for the network areaStep 2 Assess direct physical damage to network componentsStep 3 Propagate cascading effects due to dependenciesStep 4 Assess functionality lossStep 5 Predict recovery time for network functionalityStep 6 Iterate to describe level of functionality within specified Tolerance
  • 44.
    44  WN systemis dependent on the EPN system  Insets A, B, and C show the WN and EPN node dependencies  The dependencies between WN and EPN are modeled with the out-of- diagonal adjacency tables which have values equal to 1 for the node pairs just described and zeroes for all of the other terms Centerville’s WN and EPN are used as testbed of the proposed procedure
  • 45.
    45  Two casesare used to describe the one- way dependency of the WN on the EPN:  Perfect independence: the WN is decoupled from the EPN (WN)  Perfect dependence: failure at any EPN nodes fails the dependent WN node (WN + EPN)  The minimum values of the functionality metrics are lower for the WN + EPN than for the WN  Although recovery begins around the same time, it is slower for the dependent system (WN + EPN) than for the independent system (WN) The case study results show the role of dependencies in assessing network resilience
  • 46.
    46  The lossof functionality and delay in the recovery process for network dependency can be quantified.  Simulating the dependency between networks can be important in estimating the resilience of critical infrastructure.  The cascading failures in the WN due to EPN failures can lead to increased recovery times or increased extent of functionality loss.  The modularity of the proposed methodology provides flexibility and allows the user to update fragilities, component and network models, hazards, and repair rates. Concluding Remarks
  • 47.
    47 Figure 1: FlowDiagram of a Computable General equilibrium Model
  • 48.
    48 East and WestCenterville
  • 49.
    49 Restrict the earthquakeshock to hit East and West Centerville separately • Damages commercial and residential buildings • Understand the value of the spatial CGE model Dynamic outcomes • Allow the earthquake to hit both sides of Centerville simultaneously • Damages to ‐ Buildings and transportation network ‐ Workers and shoppers have to use alternative routes Motivation
  • 50.
  • 51.
  • 52.
    52 • Spatial characteristicsare important • Commuting in and out (Joplin) • Dynamic estimates will be used to model recovery
  • 53.
    53 • Building portfolioanalysis (Earthquake) • Electric Power Network (EPN) damage analysis (Tornado) • Water network recovery analysis with EPN dependency (Earthquake) • Mitigation Resources Allocation (MRA) analysis • Computable General Equilibrium (CGE) analysis IN-CORE v1 Demo for Centerville Analyses
  • 54.
    54 • Uses IN-COREscenario earthquake • Computes these for a designated zone in Centerville Building Portfolio Analysis • Immediately Occupancy Ratio • Direct Loss Ratio • Household Dislocation Ratio
  • 55.
  • 56.
    56 • Uses IN-COREscenario tornado • Computes damages of Electric power network EPN Damage Analysis
  • 57.
  • 58.
    58 • Uses IN-COREscenario earthquake • Water network recovery analysis with EPN dependency • Computes changes of avg. of pressure, demand, and contamination Water Network Recovery Analysis
  • 59.
  • 60.
    60 • Multi-objective optimization •Computes solutions of resource allocation Mitigation Resource Allocation Analysis
  • 61.
  • 62.
    62 • Uses IN-COREscenario earthquake • Computes the impact of the earthquake through the building infrastructure and total factor productivity transportation channels Computable General Equilibrium Analysis
  • 63.
  • 64.
  • 65.
    65 Outline • 2011 JoplinMO Tornado • Field Study Data Collection & Mapping • Single Sector Validation: Electric Power Network • Single Sector Validation: Buildings • Multisector Validation: Buildings-EPN • Post-Disaster Functionality Methodology • Business Interruption • Socially Vulnerable Populations • Population Dislocation • Future Plans
  • 66.
    66 Hindcasting Hindcasting is asystematic process by which we check the accuracy of mathematical (and other) models using a past event. For Joplin, this will be accomplished through a robust series of checks and hindcasts ranging from the single building or infrastructure component level, economic component, to single infrastructure sector validation – this will be done at points in time to validate damage modeling, loss estimation, population dislocation, and functionality and recovery models.
  • 67.
    67 City of Joplin,Missouri Hindcast is a way of testing mathematical models; known or closely estimated inputs for past events, are entered into the model to see how well the output matches the known results. • Land area in square miles: 35.56 • Population: 51,316 (2014 estimate) • Housing units: 23,322 • Median value of housing: $103,300 Joplin tornado – May 22, 2011 • EF5 multiple-vortex • Fatalities: 161 • Injured: 1150 • Total loss: $2.8 billion • Costliest single tornado in US history
  • 68.
    68 2011 Joplin TornadoOverview • Touched down at 5:34 PM CDT, Sunday, May 22, 2011.1 Stayed on ground for about 22 miles (6 miles in City of Joplin) and 15 minutes • Enhanced Fujita Scale EF-5 tornado; estimated maximum wind speeds: 200+ mph • Damaged/destroyed ~ 8,000 buildings. • Affected ~41% of City’s population (20,820 of 50,173). Costliest tornado on record (~$1.8 billion insured loss). • 161 fatalities and over 1,000 injuries. Deadliest single tornado on record. Exceeded U.S. average deaths/year for all tornados (91.6), hurricanes (50.8), & earthquakes (7.5). • Official warning time of 17 minutes (national average is 14 minutes).
  • 69.
    69 NIST Objectives • Determinethe tornado hazard characteristics and associated wind fields • Determine the response of residential, commercial, and critical buildings, including the performance of designated safe areas • Determine the performance of lifelines relative to the community of operations of residential, commercial, and critical buildings • Determine the pattern, location, and cause of fatalities and injuries, and associated emergency communications and public response • Identify areas in current building, fire, and emergency communications codes, standards, and practices that warrant revision
  • 70.
    70 NIST Findings • 40%of the fatalities and as much as 90 % of the tornado area were associated with EF-3 or lower wind speeds. • Of the buildings damaged, 7,411 were residential and 553 were non-residential. • The failure of building envelopes at St. John’s Regional Medical Center led to its loss of functionality; the structural frame withstood the tornado without structural collapse. • All utilities (water, gas, power) were lost in the areas most damaged by the May 22. The utility providers restored service to critical buildings (SJRMC, water treatment plant) within 24 hours.
  • 71.
    71 Field Study DataCollection & Mapping
  • 72.
  • 73.
    73 Estimating Service Areaof Each Substation Based on Weighted Cellular Automata Combined with Distribution Lines
  • 74.
  • 75.
    75 Categorizing Buildings inJoplin Single Sector Validation: Buildings Developing Damage Fragilities Bldg. ID Building Description T1 Res Wood Bldg. – Small Rectangular Plan – Gable Roof - 1 Story T2 Res Wood Bldg. – Small Square Plan – Gable Roof - 2 Stories T3 Res Wood Bldg. – Medium Rectangular Plan – Gable Roof - 1 Story T4 Res Wood Bldg. – Medium Rectangular Plan – Hip Roof - 2 Stories T5 Res Wood Bldg. – Large Rectangular Plan – Gable Roof - 2 Stories T6 Business and Retail Buildings T7 Light Industrial Buildings T8 Heavy Industrial Buildings T9 Elementary/Middle School – Unreinforced Masonry T10 High School – Reinforced Masonry T11 Fire/Police Station T12 Hospital T13 Community Center/Church T14 Government Buildings T15 Large Big Box T16 Small Big Box T17 Mobile Homes T18 Shopping Mall T19 Office Buildings
  • 76.
    76 Building Taxonomy forTornado Hazard Derivation Resources: a) Building Attributes Affecting Structural Response [GEM Building Taxonomy (2013)] a) Building Attributes Affecting Wind Load Intensity [ASCE7-10 chapters 26 to 31]
  • 77.
  • 78.
  • 79.
  • 80.
    80 Functionality of buildingBased on Damage to Buildings and Damage to Electric Power Network
  • 81.
    81 Post-Disaster Functionality Methodology •Developed for: • Any type of building (i.e. residential buildings, health care facilities, industrial, commercial etc. • Any hazard (hazard agnostic) • Integrates Performance- Based Engineering (PBE) into Community Resilience Assessment studies • Four-step probabilistic functionality methodology
  • 82.
    82 Post-Disaster Functionality Methodology •Four damage states (Slight, moderate, extensive & complete) • Three different building types: school, heavy industrial & strip mall • Output: • Intrinsic Functionality Fragilities (IFF) accounts only for construction and clean-up time • Semi-intrinsic Functionality Fragilities (SIFF) accounts for construction and clean-up time & time to obtain financing, permits and complete design • Results presented for the assumption of being in Damage State “Extensive” • Results format: Empirical and log-normal fitted cumulative distribution curves
  • 83.
  • 84.
    84 Estimating Probability ofBusiness Closure After 3 Months
  • 85.
  • 86.
    86 Influences of theEvent on Economic Factors • A social accounting matrix (SAM) summarizes the data that reflects the intersection between households, firms and the government. • SAMs will be constructed for 2010, 2011 and 2012 for Joplin • CGE model for 2010 and then use it to forecast future SAM values and compare these forecasts to the values in the 2011 and 2012 SAMs. • The forecast errors (Actual – Forecast) will determine initial ability to forecast the impacts of a tornado.
  • 87.
    87 Socially Vulnerable Populations ‐Census Tract Level example for the city ‐ Population Data Geocoded within Census Tracts ‐ Census Block Group Level example for Joplin Missouri ‐ Population Data Geocoded within Census Block Groups
  • 88.
    88 Socially Vulnerable Populations ‐90% Confidence Interval Population Estimates for population data within the EF0-EF4 2011 Joplin Tornado Path ‐ Block Group Level provides larger estimates due to variation in population density ‐ Census Tract Level estimates underestimate vulnerable populations Factors Census Tract Estimate Block Groups Estimate Percent Difference Total Population 18,325±1,001 20,168±1,535 7.6% - 12.3% 65 years and over 2,698±326 3,130±365 15.6 - 16.5% Single-parent with own children 900±234 1,069±270 18.0% - 20.0% Occupied Housing Units without a vehicle 502±144 628±182 24.8% - 25.5% Total Housing Units with 10 or more units 540±213 661±230 18.3% - 31.8%
  • 89.
    89 Future Plans • One-waydependencies beyond EPN-buildings • Recovery functions; functionality, permitting • Systematically test each of the algorithms presented in the SRI Special Issue portion of the webinar • Apply the CGE model in one year time steps • Predict population dislocation • Systematically apply decisions made by Joplin leadership over time
  • 90.
  • 91.
    91 Field Studies Thrustin the CoE • In order to measure a community’s resilience, the CoE will collect practical metrics that can be divided into two categories: 1) Metrics that require field study data: a) Population dislocation: Households displaced b) Business interruption: Business closed c) Employee dislocation: Employees failing to report to work d) Critical facilities impact: Critical facilities closed e) Housing loss: Units lost 2) Metrics that support (but do not require) field study data: a) Physical and mental morbidity b) Mortality c) Fiscal impact: Loss of property tax and sales tax • The community dimensions that are described in NIST GCR 16-001 are embedded into these metrics ‐ Dimensions include: sustenance, health, housing and shelter, education and personal development, security and safety, culture and identity, and belonging and relationships
  • 92.
    92 Disaster and FailureStudies (DFS) at NIST Statutory Arm • Evaluate hazard events against deployment criteria – new knowledge or national impact • Federal Advisory Committee (NCSTAC) • Conduct reconnaissance, field studies, or investigations • MOUs with other agencies Standard Operating Procedures • Study objectives • Field and safety protocols • Human subjects protocols (IRB, PRA) • Equipment for data collection and personnel safety • Data preservation and management Research Arm • Develop research program focused on disaster metrology • Coordination with NIST groups • Coordination with NIST’s CoE • Coordination with other federal agencies and research centers • Outreach and committee work
  • 93.
    93 CoE/NIST Collaborative FieldStudies • Data collection ‐ collect data and establish likely technical factors responsible for poor/successful performance of buildings and infrastructure in the aftermath of disasters. ‐ collect data related to community impact and recovery. ‐ collect data to validate models under development. • Field methods ‐ test new field sampling protocols. ‐ assess new field equipment. • Findings ‐ recommend frequency of data collection to capture community functioning over time. ‐ recommend, as necessary, specific improvements to standards, codes, and practices based on field studies.
  • 94.
    94 Hurricane Matthew Overview •Made landfall Oct 8, 2016 as a Cat 1 (75 mph sustained winds), in South Carolina • 27 deaths in NC; more than 2,300 people saved in over 600 rescue operations • Thousands displaced; 42 schools closed for 3 weeks • $1.5 B in damage losses; 900k without power during peak loss • Severe transportation network disruptions (including I-95 & I-40)
  • 95.
    95 Robeson County (Lumberton)Impacts • Demographics: 40% white; 37 percent black; 13% Native American (Lumbee Indian) • Per capita income was $15,321 • 1/3 population in poverty • 5,000 displaced • 7,057 structures affected • >5,000 power customer outages • 133 roads impacted or damaged • 4 public housing developments damaged, totaling 545 units
  • 96.
    96 Field Deployment ResearchGoals 1) Assess damage to school buildings and impact on educational system (e.g., school closures, food programs, other services); 2) Study how the floods impacted housing stock (collect empirical data to compare to existing fragility curves) and population displacement; 3) Investigate the dependency of these important community functions on distributed infrastructure networks (e.g., power, water, wastewater, etc.).
  • 97.
  • 98.
  • 99.
    99 Implementing Field ResearchTools • Engineering ‐ Damage assessment of a random sample of housing units (HUs) to inform fragility curves • Social Science ‐ Household interviews of a random sample of Hus/ households to inform general population models • Engineering + Social Science ‐ Qualitative survey of community leaders and stakeholders to understand impact, response, and recovery planning and activities
  • 100.
    100 IRB Field StudyProtocol • Institutional Review Board (IRB) approval from all participating Universities ‐ Develop protocols for semi-structured and structured interviews, damage assessment, and photography • Recruitment of interviewees • Informed consent • Data collection methodologies and instruments • Types of data collected • Photography: publically viewable versus internal photos and release statements
  • 101.
    101 Random sample ofHousing Units/Households Target area was the largest school attendance zone (Lumberton Jr. High) Developed sampling scheme to capture: 1. Both heavily and lightly/ none flooded areas 2. Obtain random representative sample of residential housing units and household 3. Stratified two stage random sampling procedure based on census blocks and housing units.
  • 102.
    102 Used Google Mappingtools to guide teams To facilitate teams finding the sampled Hus we employed Google Maps (Google My Maps) that could be view on smart phones.
  • 103.
    103 Used Google Mappingtools to guide teams The sample linked to a Google Doc Spreadsheet – providing information on each HU and updated each evening on status of damage assessment and HH interview.
  • 104.
    104 Daily Activities • Briefingon the day’s activities and goals • Division into qualitative and survey sampling teams • Assignments of areas or stakeholders/meetings to be attended • Lunch time check in, update, and assessment • Evening check in: • Debriefing • update of sample/survey activities • initial development of next day’s activities and assignments • Set-up for next day
  • 105.
    105 Damage Assessment Data •Most of the damaged homes are wood light-frame structures (many with brick veneer). • Average quality and typically one to two stories. • Almost two-thirds have crawlspaces. • Generally single floor • Majority single family: multi-family often duplexes or triplexes. Foundation Type (with Number of Buildings) Building Type Distribution of Buildings (%) Distribution of Number of Stories (%) Distribution of Construction Type (%) One Two Wood Masonry Both Crawlspace (273) Single 96.3 90.1 8.4 60.0 28.1 8.0 Multi 1.8 100 0.0 80.0 0.0 20.0 Slab (116) Single 33.6 89.7 10.3 56.4 25.6 15.4 Multi 64.7 86.7 12.0 41.3 29.3 18.7
  • 106.
    106 Damage Assessment Data DS Level Description 0No damage although water enters crawlspace or touches foundation (crawlspace or slab on grade). No contact to electrical or plumbing, etc. in crawlspace. No contact with floor joists. No sewer backup into living area. 1 Minor water enters house; damage to carpets, pads, baseboards, flooring. Approximately 1”, but no drywall damage. Touches joists. Could have some mold on subfloor above crawlspace. Could have minor sewer backup and/or minor mold issues. 2 Drywall damage up to approximately 2 feet and electrical damage, heater and furnace and other major equipment on floor damaged. Lower bathroom and kitchen cabinets damaged. Doors or windows need replacement. Could have major sewer backup and/or major mold issues. 3 Substantial drywall damage, electrical panel destroyed, bathroom/kitchen cabinets and appliances damaged; lighting fixtures on walls destroyed; ceiling lighting may be ok. Studs reusable; some may be damaged. Could have major sewer backup and/or major mold issues. 4 Significant structural damage present; all drywall, appliances, cabinets etc. destroyed. Could be floated off foundation. Building must be demolished or potentially replaced.
  • 107.
    107 Empirical Flood-Damage Fragilities Nextstep is to include social factors into engineering fragilities!
  • 108.
    108 Household Dislocation, Damage, andSocio-Economic Data • There was considerable dislocation and that dislocation was clearly a function of damage • We also see variations with respect to tenure, race/ethnicity, and income Household Dislocation Total Sample Damage State DS0 DS1 DS2 DS3+ Yes 48.6% 25.0% 71.5% 90.4% 94.2% No 51.4% 75.0% 28.5% 9.6% 5.8% Household Dislocation Tenure Household Racial Status Renter Owner White Black Native Am. Yes 68.5% 48.3% 26.5% 74.2% 84.1% No 31.5% 51.7% 73.5% 25.8% 15.9%
  • 109.
  • 110.
    110 Considering Damage andSocio-Economic Factors can be Important: two examples • Here we see that the probabilities of household dislocation vary not only by damage, but also by tenure and race.
  • 111.
    111 Where are weHeaded • Continue analyses ‒ linking Engineering and social science.. • Consequences of infrastructure disruption • Dislocation time • Aid and initial recovery efforts ‒ Undertake additional data collection activities to assess resiliency
  • 112.