Technical Note Seismic Risk Reduction Strategy for Public School Buildings in Peru February 2017 Project led by the World Bank, and developed by: UNIVERSIDAD DE LOS ANDES WORLD BANK Luis Eduardo Yamin Fernando Ramírez Project Lead, Professor Task Team Leader, Senior Disaster Risk Management Specialist José Raúl Rincón Technical Coordination Juan Carlos Atoche Juan Carlos Reyes Technical Management Coordination Advisor Laisa Daza Obando Álvaro Iván Hurtado Technical and Administrative Support Specialized Engineer Julián Tristancho Specialized Engineer Andrés Felipe Becerra Engineer Laura Lunita López Engineer Jonathan Estrada Administrative Assistant © 2017 International Bank for Reconstruction and Development/World Bank 1818 H St. NW Washington, DC, 20433 USA Telephone number: 202-473-1000 Web site: www.worldbank.org This work has been carried out by World Bank staff, with external contributions. The opinions, interpretations, and conclusions expressed herein do not necessarily reflect the views of the World Bank, its Board of Executive Directors or the countries represented by it. The World Bank does not guarantee the accuracy of the data contained in this publication. The boundaries, colors, names, and any other information shown in the maps of this document do not imply the expression of any opinion whatsoever on the part of the World Bank concerning the legal status of any territories, nor the endorsement or acceptance of such boundaries. Rights and Permissions The material contained in this work is copyrighted. The World Bank encourages the dissemination of its knowledge, and authorizes the partial or total reproduction of this report for non-profit purposes, provided the source is cited. All queries on rights and licenses, including subsidiary rights, should be addressed to: World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax number: 202-522-2625; e-mail address: pubrights@worldbank.org. Cover photo: “Emblematic educational institution Alfonso Ugarte, built in 1927 and structurally reinforced in 2010” Cover design: FCI Creative FOREWORD In August 2007, an earthquake with a magnitude of 7.8 (MW) struck the south of Peru with a death toll of 550 people, plus 2,000 people affected and economic losses of around USD 1 billion. With this disaster as a starting point, the World Bank put in motion a new process to support and provide technical assistance to the Government of Peru in the design of policies that would allow to reduce the impact of earthquakes on both the population and the economy. In particular, reducing the seismic vulnerability of critical infrastructure—including buildings from the health, education, transport and government sectors, among others—was set as a priority. This note presents a summary of the seismic risk assessment of the school infrastructure countrywide and a strategy for reducing its vulnerability. This study is an integral part of the main results of a program funded by the Government of Japan and the GFDRR (Global Facility for Disaster Reduction and Recovery), the main objective of which is to integrate disaster risk management into infrastructure sectors. For the first time in the history of the country, Peru has a quantitative analysis of the potential damages and losses on the country’s school infrastructure network in the event of an earthquake, as well as a risk reduction strategy. Considering the challenge posed to Peru given the need to make interventions in tens of thousands of school buildings, either for seismic retrofitting or replacement, this study provides an example on the approach, methodology and design of a seismic risk reduction strategy which may be useful for other countries with similar conditions. Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  i TABLE OF CONTENTS FOREWORD i 1. INTRODUCTION 1 2. SUMMARY OF THE METHODOLOGY 2 2.1 Seismic Hazard 2 2.2 Exposure 4 2.3 Seismic Vulnerability 6 2.4 Risk Assessment 8 3. SEISMIC RISK REDUCTION STRATEGY 11 3.1 General Objectives and Priorities of the Plan 11 3.2 Intervention Options by Structural Typology 11 3.3 Estimation of the Total Cost of Interventions 12 3.4  Intervention Cost for a 10-Year Seismic Risk Reduction Program 13 3.5 Specific Prioritization of Interventions at the School Facility Level 14 3.6 Disaggregation of Interventions by Region 15 4. CONCLUSIONS 18 5. REFERENCES 19 ii  Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru 1. INTRODUCTION The Ministry of Education of Peru (MINEDU), through the General Directorate for School Infrastructure (DIGEIE), prepared the National School Infrastructure Plan to 2025 (PNIE). Within this framework, the MINEDU commissioned the National Institute of Statistics and Information (INEI) with carrying out the School Infrastructure Census (CIE) [1] which started on September 2013 and was completed in 2014. The MINEDU requested the World Bank’s technical assistance for the analysis of the results obtained from the CIE as well as for devising a strategy to reduce seismic vulnerability and for the formulation of the PNIE. Under this program, a nationwide probabilistic seismic risk assessment of school infrastructure was carried out, which constitutes the basis for defining the seismic risk reduction strategies and for setting intervention priorities to optimize the required investments. In turn, the risk reduction strategy aims mainly to reduce the risk of death or injuries in the educational community to reduce damages to the property and infrastructure, and to minimize disruption of the educational services in the event of an earthquake. Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  1 2. SUMMARY OF THE METHODOLOGY The probabilistic seismic risk assessment of the Peruvian public school infrastructure requires quantifying the seismic hazard in the area under analysis, having a thorough knowledge of the exposed components and their replacement cost, and having detailed information on the seismic vulnerability of the main structural typologies. The probabilistic risk assessment considers the whole range of potential events that may occur in the future. In the risk assessment process, the probabilistic models take into account uncertainties which are inherent to the analytical models, and to the severity and frequency of occurrence of events. The risk model is the result of the integration of various modules associated to hazard, exposure and vulnerability, as illustrated in Figure 2-1. Reference 2 presents the detailed methodology for the analysis of risk derived from seismic events. Figure 2-1  General outline of the probabilistic risk analysis Exposure Vulnerability Hazard modules module module Risk module 2.1 SEISMIC HAZARD Seismic hazard is depicted using maps of distribution of the seismic intensity parameters, such as peak ground acceleration or peak acceleration of school buildings. Due to its influence, seismic hazard should include the effects of soil deposits in each particular location. Intensity maps are assessed for a sufficiently wide set of possible events that might occur, taking into account the possible magnitude ranges in the different seismic sources and the relative distances between these and the buildings under analysis. As regards this case, the seismic intensity considered for the analysis is the peak acceleration response of each structural typology. Moreover, every event is characterized by the annual average frequency of occurrence, which is obtained from the analysis of the historical frequency of events. Figure 2-2 shows the seismic hazard zones defined by the National Building Code (RNE) [3] updated in the year 2016, and the zoning proposed at country level in order to consider the soil dynamic amplification effects according to the different soil types (from 1 to 8 in accordance with Reference 4), where type 1 soil is the hardest, and type 8 soil is the softest. 2  Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru Seismic hazard zoning and soil dynamic amplification effects Figure 2-2   Type of soil 1 2 ZONE 3 4 4 0.45 5 3 0.35 6 2 0.25 7 1 0.10 8 a) Map of seismic hazard in RNE (2016): b) Classification of Peruvian soils Peak ground accelerations without according to methodology Vs30 [4] amplification 2.5 2.0 Amplification factor 1.5 1.0 0.5 0.0 0 0.5 1.0 1.5 2.0 Structural period (sec.) 8 7 6 5 4 3 2 1 c) Final amplification spectra for the eight types of soil defined Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  3 2.2 EXPOSURE Exposure is calculated based on a georeferenced database of the exposed school buildings which may sustain damages due to the occurrence of seismic events. The information gathered includes: ID, geographical location, replacement cost and associated seismic vulnerability function. Additionally, for the purpose of evaluating vulnerability, information regarding the structural system, height, level of seismic-resistant design (seismic code level), quality of the design and construction is included, as well as supplementary information of each school building. Replacement costs are defined according to the geographical location and the school setting1 of each school facility, and are based on the statistical analysis of the information available regarding direct construction costs (see Reference 5). Table 2-1 provides a quantitative and economic description of the exposure of public school infrastructure to earthquakes in Peru. Table 2-1  Exposure of school infrastructure to earthquakes Characteristics Value Number of public school facilities 40,475 Total number of buildings 187,312 Number of buildings for educational use 2 152,660 Economic valuation of buildings for educational use USD 8.4 billion Figure 2-3 shows the distribution of the replacement costs by country regions and by structural typology for the complete inventory of buildings. The predominant structural typologies are the following: P = Precarious; A = Adobe; UCM = Unconfined masonry; SS = Steel structure; T = Timber; PROV =Provisional; MCF = Concrete frames with masonry walls built by Parents’ Associations (APAFAs); LSU = Large school units; 780-PRE = 780 modular system built before the 1998 seismic standard; and 780-POST = 780 modular system built after the 1998 seismic standard. Modular 780-PRE buildings are especially relevant, as they include all the buildings with moment resisting reinforced concrete frames built between 1978 and 1997 by the national or regional governments according to the CIE. As regards their seismic response, they are characterized by a great flexibility and problems with their short columns, which lead to an anticipated structural failure in case of seismic events. 1. The methodology applied to cost estimation took into account an adjustment in the urban-rural distribution considering five school settings: big cities, mid-size cities, urban centers, connected villages, and scattered communities. 2. Buildings with lower occupancy (such as warehouses, storage rooms, restaurants, security booths, staircases, among others) are excluded from the group of “buildings for educational use”. 4  Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru Replacement cost by region and structural typology Figure 2-3   PERU Replacement cost (millions of USD) 56-120 120-241 241-380 380-639 639-1,146 a) Distribution of replacement cost by department 60,000 30 In percentage (%) 40,000 20 No. of buildings 20,000 10 0 0 P A UCM SS T PROV MCF LSU 780- 780- PRE POST No. 26,559 47,788 16,437 500 4,112 3,527 19,354 351 21,104 47,580 % 14% 26% 9% 0% 2% 2% 10% 0% 11% 25% b) Distribution of exposure value at national level according to structural typology Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  5 2.3 SEISMIC VULNERABILITY Seismic vulnerability is illustrated by means of functions that relate the damage or expected losses expressed as a percentage with the seismic intensity selected. These functions represent the expected seismic behavior of the buildings from each particular typology, so their use is statistically representative when there is a wide inventory of exposed assets. Each vulnerability function is defined by a mean damage level and its variance, which makes it possible to estimate the probability function of the respective losses for different seismic intensities. This study makes use of the vulnerability functions proposed in references 6, 7, and 13. Figure 2-4 a) shows a table of the structural typologies defined for the determination of their seismic vulnerability, while Figure 2-4 b) and c) show the representative vulnerability functions used for the risk analysis. Description of vulnerability functions used for the analysis Figure 2-4   Typical height Seismic code level No. of No. Structural typology Description Range stories P L M H 1 Adobe (A) Adobe Low 1+ X X — — Unconfined masonry Low 1-2 X X — — 2 Load-bearing walls in simple masonry (UCM) Medium 3-5 X X — — Informal precarious constructions 3 Precarious (P) Low 1+ X — — — (plywood, quincha, etc.) 4 Steel structures (SS) Steel frames Low 1-3 X X X — 5 Timber structures (TS) Timber constructions Low 1+ X X — — Reinforced concrete Concrete structures with concrete frames; Low 1-3 X X X — 6 frames (RCF) highly uncertain seismic response Medium 4-7 X X X — Concrete frames built before the Low 1-3 — X X — 7 Large school unit (LSU) institution of the Peruvian building standards Medium 4-7 — X — — 780 pre-code (PRE) 780 module prior to the 1998 standard; 8 Low 1-3 — X X — modules problems related to short columns 780 post-code (POST) Low 1-3 — — X X 9 780 module after the 1998 standard modules Medium 4-7 — — X X Provisional classrooms Provisional classrooms built by the Low 1-3 — X X — 10 (PROV) government after the 1998 standard Medium 4-7 — X X — 6  Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru 1.0 Informal-P Adobe-P Adobe-L Simple masonry-P Simple masonry-L 0.8 Simple masonry-M Conf. masonry-P Conf. masonry-L Conf. masonry-M Damage (%) 0.6 Conf. masonry-A Steel struct.-P Steel struct.-L Steel struct.-M Syst. APAFA-P 0.4 Syst. APAFA-L Timber-P Timber-L LSU-L PRE-L 0.2 PRE-M POST-L POST-M POST-A 0.0 Prov. classrooms-L 0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 2,000 Elastic spectral acceleration (Gal) 1.0 Informal-P Adobe-P Simple masonry-L 0.8 Conf. masonry-P Conf. masonry-L Conf. masonry-M Damage (%) 0.6 Conf. masonry-A Syst. APAFA-P Syst. APAFA-M 0.4 LSU-L PRE-L PRE-M 0.2 POST-L POST-M POST-A 0.0 Prov. classrooms-L 0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 2,000 Elastic spectral acceleration (Gal) Note: The “Seismic Code Level” as used in the text is defined as follows: P = pre-code; L = low code; M = medium code; and H = high code. Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  7 2.4 RISK ASSESSMENT The risk assessment using probabilistic techniques with a CAPRA-type approach is widely documented. References 8, 9, 10 and 11 present in detail the methodological bases of the procedures used in this study. For the risk assessment, the hazard and vulnerability of the exposed elements are included in order to obtain parameters that indicate level of damage, physical impacts, and overall impact on the infrastructure or its occupants. Once the expected physical damage (potential average value and its dispersion) has been estimated for each of the exposed buildings, whether as a percentage or as an absolute value, it is possible to estimate different metrics, such as average annual loss (AAL) or probable maximum loss (PML) in absolute (USD) or relative (%) terms as regards the exposed value3. Figure 2-5 shows the basic results of the seismic risk assessment for the building inventory in the education sector in Peru. Figure 2-6 shows the geographical distribution of average annual loss by department in absolute and relative terms as regards the respective replacement costs. In turn, Figure 2-7 shows the distribution of average annual loss by structural typology and school setting. Figure 2-5  Results of the seismic risk assessment. Average annual loss and probable maximum loss (PML) curve Results Exposed value USD x106 9,087 Average annual loss USD x106 190.0 ‰ 20.91 PML Return period (AAL) Loss Years USD x106 % 100 308 3.4 250 408 4.5 500 497 5.5 1,000 590 6.5 800 9.00 Probable maximum loss 600 6.75 In percentage (%) (millions of USD) RP 1,000 PML (6.5%) RP 500 PML (5.5%) 400 4.50 RP 250 PML (4.6%) RP 100 PML (3.4%) 2.25 200 0 0.00 0 250 500 750 1,000 1,250 1,500 1,750 2,000 Return period (years) 3. In a simple insurance scheme, the AAL represents the annual premium of insurance considering all possible earthquakes (rare and highly frequent). The PML is the loss that may occur as a result of rare earthquakes (or earthquakes with a high return period [RT], for example, 100 years, 250 years, etc.). 8  Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru Geographical distribution of average annual loss by department Figure 2-6   PERU PERU Average annual Average loss (millions annual loss of USD) (‰) 0-2 0-3 2-5 3-14 5-10 14-20 10-17 20-24 17-28 24-31 Total and relative average annual loss by (a) structural typology, and (b) school setting Figure 2-7   for the national portfolio of exposure 90 50 72 40 AAL (millions of USD) 54 30 AAL (‰) 36 20 18 10 0 0 P A UCM SS T PROV MCF LSU 780- 780- PRE POST AAL (millions 31 82 14 1 5 2 29 2 15 9 of USD) AAL (‰) 43 39 33 28 42 25 24 15 10 3 (a) Structural typology Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  9 100 1,046,699 50 students (21%) 80 40 AAL (millions of USD) 229,544 60 1,061,132 students 30 1,499,700 AAL (‰) students (5%) students (21%) 1,172,926 (30%) students 40 (23%) 20 20 10 0 0 Big Capital cities Urban Connected Scattered cities centers villages communities AAL (millions 22 23 43 87 15 of USD) AAL (‰) 23 17 17 24 25 (b) School setting Note: See description of structural typologies in 2.2. As can be observed in these figures, risk tends to be geographically concentrated according to the seismic hazard level and the dominant construction characteristics in each region of the country. On the other hand, the most vulnerable structural typologies—and the more frequent ones—such as precarious, adobe, unconfined masonry, concrete frames with masonry walls, and 780-PRE systems are the ones that accumulate the greater risk. From the school settings perspective, connected villages concentrate a higher risk in terms of AAL. 10  Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru 3. SEISMIC RISK REDUCTION STRATEGY In order to define an optimal seismic risk reduction strategy, it is necessary to carry out the following tasks: 1. Define the main objectives and priorities. 2. Define the intervention options by structural typology according to their level of risk. 3. Estimate the total cost of interventions. 4. Define the criteria to prioritize the interventions. 5. Carry out the optimization of the intervention strategy and the prioritization of the intervention subprograms. 6. Disaggregate the intervention plan by region for implementation. 3.1 GENERAL OBJECTIVES AND PRIORITIES OF THE PLAN The risk reduction strategy for school infrastructure in Peru is designed to meet the following specific objectives and priorities: 1. Reduce the risk of death or injuries in the community resulting from seismic events (maximizing the number of benefited students). 2. Reduce damages to the infrastructure and protect the property. 3. Minimize disruption of the educational service. 3.2  INTERVENTION OPTIONS BY STRUCTURAL TYPOLOGY The intervention of school buildings is aimed at correcting possible structural defects and at providing the structure with an appropriate combination of stiffness, resistance and ductility which may ensure its adequate structural behavior in future seismic events under the terms established in the seismic-resistant design standard E030 of the Peruvian National Building Code [3]. Four main intervention alternatives are defined: • Conventional retrofitting: The retrofitting intervention is made in a single phase and in such a way that the school building reaches the level of seismic response established by the E030 Standard [3]. I Incremental retrofitting: The structural intervention is carried out in two or more phases with a targeted performance level to be achieved in each phase. • Replacement of school buildings by new seismic-resistant buildings: It is applied when there is no technical and/or financial feasibility for structural retrofitting. It involves the demolition of the existing building, the installation of temporary classrooms, and the design and construction of a new building. • Contingent intervention to prevent collapse: It is a type of intervention for highly vulnerable structural typologies with the sole purpose of preventing collapse. It is a temporary intervention that would be carried out when the above alternatives are technically, financially or logistically impossible. Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  11 Table 3-1 summarizes the recommended intervention options according to the level of risk of the structural typologies. Table 3-1  Possible types of structural intervention Buildings with expected Types of adequate intervention Buildings with high risk of collapse (HRC) Buildings with high damage potential (HDP) performance Definition and Poor seismic response; their intervention Regular seismic response under Seismic- characteristics implies major technical difficulties, high earthquakes of medium/high magnitude. resistant costs, and few guarantees of functionality. Intervention is technically, functionally buildings and economically feasible. Structural • Adobe (A) • Large school units (LSU) • 780-POST typology • Unconfined masonry (UCM) • Moment resisting concrete frames (MCF) modules • Precarious (P) • 780-PRE modules • Provisional (PROV) Intervention a) Replacement by new seismic-resistant a) Incremental retrofitting with Not required options buildings. interventions carried out in phases. b) Replacement by temporary classrooms Compliance with fundamental (in the short term) while modular requirements established by the alternatives are defined. country regulations must be achieved c) Contingent intervention to prevent in the initial phase. collapse. b) Conventional retrofitting with a single intervention phase for the building to fully comply with all requirements established in the country regulations. c) Contingent intervention in buildings located in medium and low hazard zones. 3.3  ESTIMATION OF THE TOTAL COST OF INTERVENTIONS Based on the groups of structural typologies previously defined and the associated lines of intervention, intervention subprograms and their approximate implementation cost were defined. Table 3-2 summarizes the information for each of the subprograms. Table 3-2  Summary of the total cost of interventions Total cost No. of buildings in millions of USD Program for seismic vulnerability reduction in school infrastructure 139,732 6,032 Cost by subprogram Subprogram No. 1: Replacement of HRC buildings 97,110 4,660 Subprogram No. 2: Conventional retrofitting 39,933 1,353 Subprogram No. 3: Buildings in low seismic hazard zones 2,689 19 3.4.  INTERVENTION COST FOR A 10-YEAR SEISMIC RISK REDUCTION PROGRAM As the PNIE was drawn up for a 10-year period, a risk reduction program was defined for the same period. In order to optimize the resources for this program, only the buildings classified according to their educational use as common classrooms, restrooms for boys and girls, students and staff, libraries, faculty lounges, and principal’s offices, among others, are included. Based on this, the following are considered second priority buildings: • Buildings with non-educational specific uses, such as pantries, kitchens, cafeterias, waiting areas, educational material warehouses, staircases, print rooms, security and security booths, among others • Buildings with expected adequate seismic performance (ASP) • Buildings located in low seismic hazard zones Based on this, the statistics for the 10-year program are obtained, which are included in Table 3-3. Table 3-3  Financial gap summary for the 10-year seismic risk reduction program Intervention value Program No. of buildings (in millions of USD) Cost of the 10-year program Seismic risk reduction program 108,629 2,778 10-year program gap, differentiated by subprogram Subprogram No. 1: Replacement by new construction  73,645 1,995 Subprogram No. 2: Incremental retrofitting  34,984   783 Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  13 3.5  SPECIFIC PRIORITIZATION OF INTERVENTIONS AT THE SCHOOL FACILITY LEVEL Considering the number of school buildings which require an intervention within a 10-year period and the budget limitations, it is necessary to optimize the intervention strategy and to prioritize the school facilities to be intervened in each of the subprograms. The aim is to maximize the cost-effectiveness of the interventions performed in line with the objectives set, particularly the objective of increasing the number of students benefited by the risk reduction measures. This analysis is made at the level of each school facility since, in practice, this is the level at which each intervention will take place. The prioritization is based on the cost-effectiveness analysis of the interventions, which is defined as follows: NS • (AALi – AALf) CE = C Where CE  =  cost-effectiveness indicator; NS  =  number of potential students in each school facility; AALi = average annual loss at initial state (in millions); AALf = average annual loss at final state (once the proposed intervention has been implemented, in thousands); and C = cost of the proposed intervention. Based on this, the order of intervention priority is determined by school facility so as to maximize the benefits of the risk reduction measures according to the number of students. Priority criteria are consistently applied to each of the intervention subprograms proposed. Figure 3-1 shows the impact of the buildings intervened in Subprogram No. 1: Replacement of HRC buildings, and Subprogram No. 2: Incremental retrofitting of HDP buildings, in terms of intervention cost, number of students benefited and AAL percentage reduction at national level. These figures have been evaluated using the list of facilities sorted form the most critical to the less critical one, and aggregating the listed values. Impact of intervention measures. a) Cost of intervention; (b) Number of students Figure 3-1   benefited; and (c) Risk percentage reduction according to the number of school facilities intervened Investment (millions of dollars) AAL percentage reduction (%) 3,000 5.0 90 No. of students benefited 4.5 80 2,500 4.0 70 (in millions) 2,000 3.5 60 3.0 50 1,500 2.5 40 1,000 2.0 30 1.5 20 500 1.0 10 0 0 0 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 No. of school facilities No. of school facilities No. of school facilities Replacement Retrofitting Total Replacement Retrofitting Total Replacement Retrofitting Total a) b) c) Based on a given budget availability, Figure 3-1 a) allows for the estimation of the number of facilities to be intervened. On this basis, the number of students benefited and the percentage in risk reduction of the portfolio can be also estimated using Figure 3 1 b) and c), respectively. 14  Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru From these figures, it may also be concluded that Subprogram No. 2 is more efficient in terms of the number of students benefited, while Subprogram No. 1 is more efficient in terms of effective seismic risk reduction as regards the AAL. On the other hand, given a certain sum allocated for the optimization in line with the previous criteria, a greater relative investment should be made for the replacement of buildings as compared with the required amount for seismic retrofitting. 3.6 DISAGGREGATION OF INTERVENTIONS BY REGION The intervention plan requires the implementation by region, as available resources are usually allocated and executed at the regional level. The following information is required for each region of the country: a) List of prioritized school facilities for intervention purposes b) Intervention proposals for each building and their estimated cost c) Aggregate cost of each of the subprograms proposed For illustrative purposes, comparative charts between two regions, Lima and Amazonas, are shown, which make evident the regional differences that may appear during the implementation of the plan. Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  15 Figure 3-2  Comparison of results for Lima and Amazonas LIMA AMAZONAS 16  Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru Figure 3-2  Comparison of results for Lima and Amazonas LIMA AMAZONAS Based on this information, regional governments can carry out the implementation of the plan following the procedure below: 1. Define the amount of resources to be invested in each of the subprograms. 2. Quantify the following three parameters according to the desired investment in each subprogram: • Impact on the number of students benefited by interventions • Percentage of reduction in risk compared to initial risk • Number of school facilities or buildings intervened Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  17 3. Redistribute the amounts by program until reaching a high impact solution in accordance with specific criteria for the region. 4. Check the list of school facilities prioritization in order to identify the geographical location and the characteristics of the facilities included. In particular, the list indicates the type of intervention recommended and the estimated budget for each building. 5. Based on the above, set the terms for the execution of specific intervention projects and commission the final designs and intervention works. 6. Final phase of the plan implementation. 4.  CONCLUSIONS The analytical work in this present study allows to draw a number of conclusions regarding the elements to consider in the design and implementation of a seismic risk reduction strategy for school facilities. Those elements are listed below. a) The average annual loss of the inventory is USD 190 million, which, in relative terms, equals 2.1% of its replacement cost. This loss does not include loss of content, nor indirect losses derived from the disruption to operations and loss of profit. In comparison to the analysis of similar inventories, this figure is relatively high, which is attributed to the high seismic hazard and the high vulnerability of most of the components in the inventory. b) The average annual loss before and after the intervention, the number of students, and the intervention cost may be combined in a prioritization criterion that maximizes cost-effectiveness given the size of the inventory (187,312 buildings) and the budget limitations. c) Risk is not uniformly distributed in the inventory. The first 15,000  school facilities (38%) concentrate more than 55% of the risk. The distribution of the average annual loss in the country shows that most regions in the south, the capital city and one region in the north have the highest seismic risk (between USD 10 million and USD 28 million). The average annual loss is critical in adobe school buildings and in the country’s rural areas classified as connected villages. d) The probable maximum loss, for events with a return period of 1,000 years, is USD 600 million, which correspond to approximately 6.5% of the inventory replacement cost. This figure is high in comparison with equivalent inventories from other regions and countries. e) The risk metrics estimated for each region allows to define the intervention strategy, which includes the following components: Criteria applied to the definition of the interventions for the different structural typologies •  identified • An estimate of the economic investment for seismic risk mitigation (financial gap) and the definition of an investment plan in line with the available budget • The definition of the optimal intervention strategy 18  Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru • Prioritization criteria for each intervention line proposed, which may allow to maximize the risk reduction objectives • Organization of the technical information required to implement the action plans by region f) The main objectives of the seismic risk reduction plan for school infrastructure are the following: • Reduce the risk of death or injuries in the community resulting from seismic events (maximizing the number of benefited students). • Reduce damages to the infrastructure and protect the property. • Minimize disruption of the educational service. g) The direct costs assigned according to the program, climatic zone and school setting allow to estimate that the investment needs that the Government of Peru will have to cover in the next 10 years amounts to about USD 2,778 millions. h) The Government of Peru faces a significant challenge as 51% of the buildings belong to the structural typology with high risk of collapse and 21% of the buildings have a high damage potential. i) Based on this categorization, the study proposes replacement, retrofitting and contingent intervention programs as strategies to reduce seismic risk in school buildings. As part of the retrofitting program, the implementation of incremental retrofitting is suggested as an innovative and cost-effective technique. The proposed methodology represents an important contribution to the optimization of seismic risk mitigation programs for school buildings in Peru and in other countries with similar problems. 5.  REFERENCES 1. Instituto Nacional de Estadística Informática (INEI) y Ministerio de Educación (MINEDU), “Censo de Infraestructura Escolar (CIE)”. 2013, Lima: Peru. 2. Universidad de los Andes, “Evaluación del riesgo sísmico del portafolio de edificaciones en locales escolares en Lima Metropolitana y Callao a partir de los resultados del CIE”. Report prepared for the World Bank, 2014, Bogotá D.C: Colombia. 3. Norma E-030. 2015 “Diseño Sismorresistente”. Lima: Ministerio de Construcción, Vivienda y Saneamiento- SENCICO. 4. Allen T., and Wald D. (2009). Short Note. “On the Use of High-Resolution Topographic Data as a Proxy for Seismic Site Conditions”, U.S. Geol. Surv. Open-File Rept. 2009 5. Universidad de los Andes. “Asesoría técnica en la definición y establecimiento de criterios de intervención y priorización que contribuyan a un mejor planeamiento, eficiencia y sostenibilidad en materia de infraestructura educativa”. Centro de investigaciones y obras civiles (CIMOC). Report prepared for the World Bank, 2016. Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru  19 6. Yamin, L., Hurtado A., Barbat A., Bernal G., Cardona O., Earthquake Vulnerability Assessment of Buildings for Catastrophic Risk: Analysis in Urban Areas, in 15 World Conference on Earthquake Engineering (WCEE). 2012, Lisbon, Portugal. 7. ISDR, U.N., UNISDR. GAR13: Global Assessment Report on Disaster Risk Reduction. 2013 Geneva, Switzerland. 8. ERN-AL. Metodología de Modelación Probabilista de Riesgos Naturales. Informe Técnico ERN‑CAPRA-T1-1. Consorcio Evaluación de Riesgos Naturales - América Latina; 2011. 9. Cardona, O.  D., Ordaz M., Yamin L., Marulanda M., Barbat A., “Earthquake Loss Assessment for Integrated Disaster Risk Management”. Journal of Earthquake Engineering, 2008. 12(sup2): p. 48–59. 10. Cardona, O.D., and Yamin L. E., Seismic Microzonation and Estimation of Earthquake Loss Scenarios: Integrated Risk Mitigation Project of Bogota, Colombia. Earthquake Spectra, 1997. 13(4): p. 795–814. 11. Yamin, L., Ghesquiere F., Cardona O., Ordaz M., “Modelación probabilista para la gestión del riesgo de desastre, El caso de Bogotá Colombia”. 2013, Bogota, Colombia: Banco Mundial, Universidad de los Andes. 12. Pontificia Universidad Católica de Perú (PUCP), “Evaluación probabilista del riesgo sísmico de locales escolares en la ciudad de Lima”. Report prepared for the World Bank, 2014, Lima: Peru. 13. Yamin L. E., Hurtado A., Barbat A. H., Cardona O. D. “Seismic and wind vulnerability assessment for the GAR-13 global risk assessment”. International Journal of Disaster Risk Reduction. 2014;10:452–60. 14. Instituto Nacional de Estadística Informática (INEI) y Ministerio de Educación (MINEDU), “Manual del Evaluador Técnico” Censo de Infraestructura Escolar (CIE). 2013, Lima: Peru. 15. Building Seismic Safety Council. 2009. FEMA 420. ”Engineering Guideline for Incremental Seismic Rehabilitation”. Washington D.C: Federal Emergency Management Agency. 16. Pontificia Universidad Católica del Perú, Universidad Nacional de Ingeniería “Evaluación simplificada de las tres técnicas de reforzamiento más apropiadas. Propuesta de actualización del reglamento nacional de edificaciones para incorporar el reforzamiento incremental en edificaciones escolares tipo 780 construidas antes de 1997”. Report prepared for the World Bank, 2016, Lima, Peru. 20  Technical Note. Seismic Risk Reduction Strategy for Public School Buildings in Peru