Building Portfolio Research Articles

Using Machine Learning to Predict Retrofit Effects for a Commercial Building Portfolio

Buildings account for 40% of the energy consumption and 31% of the CO2 emissions in the United States. Energy retrofits of existing buildings provide an effective means to reduce building consumption and carbon footprints. A key step in retrofit planning is to predict the effect of various potential retrofits on energy consumption. Decision-makers currently look to simulation-based tools for detailed assessments of a large range of retrofit options. However, simulations often require detailed building characteristic inputs, high expertise, and extensive computational power, presenting challenges for considering portfolios of buildings or evaluating large-scale policy proposals. Data-driven methods offer an alternative approach to retrofit analysis that could be more easily applied to portfolio-wide retrofit plans. However, current applications focus heavily on evaluating past retrofits, providing little decision support for future retrofits. This paper uses data from a portfolio of 550 federal buildings and demonstrates a data-driven approach to generalizing the heterogeneous treatment effect of past retrofits to predict future savings potential for assisting retrofit planning. The main findings include the following: (1) There is high variation in the predicted savings across retrofitted buildings, (2) GSALink, a dashboard tool and fault detection system, commissioning, and HVAC investments had the highest average savings among the six actions analyzed; and (3) by targeting high savers, there is a 110–300 billion Btu improvement potential for the portfolio in site energy savings (the equivalent of 12–32% of the portfolio-total site energy consumption).

Assessment of the uncertainty in spatial-correlation models for earthquake ground motion due to station layout and derivation method

The evaluation of the aggregate risks to spatially distributed infrastructures and portfolios of buildings requires quantification of the estimated shaking over a region. To characterize the spatial dependency of ground motion intensity measures (e.g. peak ground acceleration), a common geostatistical tool is the semivariogram. Over the past decades, different fitting approaches have been proposed in the geostatistics literature to fit semivariograms and thus characterize the correlation structure. A theoretically optimal approach has not yet been identified, as it depends on the number of observations and configuration layout. In this article, we investigate estimation methods based on the likelihood function, which, in contrast to classical least-squares methods, straightforwardly define the correlation without needing further steps, such as computing the experimental semivariogram. Our outcomes suggest that maximum-likelihood based approaches may outperform least-squares methods. Indeed, the former provides correlation estimates, that do not depend on the bin size, unlike ordinary and weighted least-squares regressions. In addition, maximum-likelihood methods lead to lower percentage errors and dispersion, independently of both the number of stations and their layout as well as of the underlying spatial correlation structure. Finally, we propose some guidelines to account for spatial correlation uncertainty within seismic hazard and risk assessments. The consideration of such dispersion in regional assessments could lead to more realistic estimations of both the ground motion and corresponding losses.

Reliability of Seismic Performance Assessments for Individual Buildings and Portfolios

Seismic performance and loss assessments are required in areas of Insurance, Finance and Public Policy. Providers are Structural Engineers and Risk Management Firms. There are no current procedures to evaluate the epistemic and aleatory uncertainties for such assessments. The essential issue is whether or not there is sufficient reliability in the result to use the result as the basis for risk management decisions and actions. For a single building this may be whether or not a prescribed earthquake performance level is met, life safety or if a portfolio’s vulnerability level is acceptable, whether the. loss for a given time period is less than a stated value. A method based in part on Federal Emergency Management Agency P-695, is developed for evaluating the reliability of performance and/or loss assessments for both individual and portfolios of buildings. Consideration is given to how well the building investigation and corresponding evaluation process have been performed, the qualifications of the person(s) doing the assessment, the thoroughness of the building evaluation, the technical validity of the assessment procedure or model and what computational reliabilities are presented. The method characterizes the uncertainty of each component of the assessment procedure for each building by qualitative determined assignments. The resulting reliability measure is likely to be most useful for determining whether/or not a building has acceptable life safety performance, or if a portfolio has an acceptably low loss risk over a given period of time. In both cases, the reliability must either be sufficient to warrant action, or serve to indicate need for improved assessment.

A computational framework for large-scale seismic simulations of residential building stock

Urban areas reveal particularly vulnerable due to the high concentration of people and, in many cases, their hazard-prone location. Indeed, according to data from the United Nations, about 2/3 of the population will live in large cities by 2050, and the majority of the world’s cities are highly exposed to disasters. This paper presents a computational framework to assess the seismic vulnerability and the damage of residential building portfolio in urban areas. First, a surrogated model is proposed to estimate the global capacity of building structures. Monte Carlo simulations are implemented to take into account the uncertainties associated with the material, mechanical, and geometrical parameters. The proposed approach is validated through nonlinear finite element models and a real case study. Then, the proposed computational framework is implemented and applied to a virtual city that is envisioned for being representative of a typical Italian residential building stock. The main achievement of this work is to introduce a new simplified approach for large scale structural analyses to limit the computational efforts while providing reasonable results.

Spatial correlation of broadband earthquake ground motion in Norcia (Central Italy) from physics-based simulations

This paper investigates the spatial correlation of response spectral accelerations from a set of broadband physics-based ground motion simulations generated for the Norcia (Central Italy) area by means of the SPEED software. We produce several ground-motion scenarios by varying either the slip distribution or the hypocentral location as well as the magnitude to systematically explore the impact of such physical parameters on spatial correlations. We extend our analysis to other ground-motion components (vertical, fault-parallel, fault-normal) in addition to the more classic geometric mean to highlight possible ground-motion directionality and therefore identify specific spatial correlation features. Our analyses provide useful insights on the role of slip heterogeneities as well as the relative position between hypocentre and slip asperities on the spatial correlation. Indeed, we found a significant variability in terms of both range and sill among the considered case studies, suggesting that the spatial correlation is not only period-dependent, but also scenario-dependent. Finally, our results reveal that the isotropy assumption may represent an oversimplification especially in the near-field and thus it may be unsuitable for assessing the seismic risk of spatially-distributed infrastructures and portfolios of buildings.

Computer-Aided Decision Making for Regional Seismic Risk Mitigation Accounting for Limited Economic Resources

Seismic risk mitigation levels for an existing building are a balance between the reduction of risk and the cost of rehabilitation. Evidently, the more that is paid the more risk is reduced; however, due to limited public budgets a practical approach is needed to manage the risk reduction program when a portfolio of buildings is concerned. Basically, decision makers face a challenge when there are a large number of vulnerable buildings and there is no plan for how to allocate the appointed budget. This study develops a technological platform that implements a decision-making procedure to establish how to optimally distribute the budget in order to achieve the maximum possible portfolio risk reduction. Decisions are made based on various presumed intervention strategies dependent on building’s level of risk. The technological platform provides an interactive, user-friendly tool, available online, that supports stakeholders and decision makers in understanding what the best economic resource allocation will be after selecting the available budget for a specific portfolio of buildings. In addition, the ease of use enables the user to analyze the extent of risk reduction achievable for different budget levels. Therefore, the web platform represents a powerful tool to accomplish two challenging tasks, namely optimal budget selection and optimal budget allocation to gain territorial seismic risk mitigation.

The implementation of the Italian National Seismic Prevention Plan: A focus on the seismic upgrading of critical buildings

The Italian National Seismic Prevention Plan (NSPP) was issued following the destructive 2009 L'Aquila earthquake, with a total endowment of 965 million euros granted by law, under the coordination of the Italian Civil Protection Department (ICPD). The NSPP addresses seismic risk reduction according to a broad view, through multi-task programs. Among the measures granted by NSPP, one is devoted to the seismic retrofit of strategic or critical public buildings and infrastructures, whose damage or collapse may affect emergency management as well as produce high consequences in terms of human lives.The paper provides a preliminary assessment of such seismic prevention measure, through some statistical elaborations on a sample of around 1000 interventions, 375 of which so far concluded. The analyses carried out on this sample provide interesting information to verify the effectiveness of the measure as well as to enhance further actions similarly purposed.Analyses on seismic risk indices of critical buildings included in the NSPP are presented and discussed in the paper, together with an early analysis of costs and benefits associated with different structural types, and intervention types. The results obtained highlight interesting aspects in terms of seismic safety increases and marginal costs for the masonry and reinforced concrete buildings of the considered sample. These preliminary results, though requiring further calibrations on the larger datasets foreseen in the future, provide useful hints to envisage how they could be used in the definition of future strategies for seismic risk reduction. To this aim, a final paragraph provides an exemplification and a simplified procedure based on the findings shown in the paper, for comparing costs and benefits of different strategies aimed at reducing the risk of a given portfolio of buildings.

A Markov framework for generalized post-event systems recovery modeling: From single to multihazards

State-dependent models can be used to represent the system recovery process as a series of stochastic transitions from lower to higher functional states. However, the applications of these models have been limited in scope and there is a lack of a generalized recovery modeling framework. A generalized framework would permit a robust forecasting of systems and system-of-systems recovery under multiple hazards, and more broadly, would contribute to community disaster preparedness. This paper develops a generalized post hazard-event recovery modeling framework based on state-dependent Markov-type processes. We then apply the proposed framework to solve a spectrum of problems that range from hind-casting single-system recovery following a single hazard event to forecasting post-event trajectories under multiple hazards and modeling the recovery of a system-of-systems. First, Markov chains are used to hind-cast the observed recovery for a portfolio of buildings affected by the 2014 South Napa, California, earthquake. Next, Markov processes are used to formulate a parametric post hazard-event recovery model, which can be updated using Bayesian statistics when relevant datasets become available. Semi-Markov processes are then used to develop a more general model of single hazard recovery, which accounts for the intensity of the loading and level of damage caused by the event. Semi-Markov processes with non-renewal features are then used to account for multihazard interactions in a post-event recovery model, and applied to a case study that involves a community in Charleston, South Carolina. Lastly, Markov-type processes are combined with Bayesian networks to model the recovery of residential, commercial, educational, and industrial buildings (system-of-systems) following a hazard event. These applications demonstrate the versatility of the Markov framework towards handling recovery problems with varying levels of complexity.

Nonlinear static characterisation of masonry-infilled RC building portfolios accounting for variability of infill properties

The seismic assessment of existing masonry infilled reinforced concrete (RC) building portfolios is a highly relevant issue in Italy and other Mediterranean countries, particularly in regions where a large part of the built environment was not designed according to modern seismic codes. In the past, masonry infills were generally considered as non-structural elements and were not accounted for in the design process. However, the results of experimental tests and past earthquake evidence demonstrated the influence of masonry infills on the global and local behaviour of RC buildings. Furthermore, the acknowledgment of different construction practices points out the significant variability affecting the masonry infills in terms of material properties, thickness, presence of openings and manufacturing techniques. This paper deals with the large-scale performance analysis of existing masonry infilled RC building portfolios accounting for the uncertainty related to the variability in the mechanical properties and the modelling of masonry infills. To accomplish this goal, a fully integrated building stock, representative of masonry infilled RC buildings designed according to the Italian codes in force between 1960 and 1980, was developed, using a simulated design approach. Statistical data and results of experimental tests were used to identify the main masonry infill typologies adopted in the Italian context. Detailed numerical models, whose prediction capability was appraised through the results of cyclic quasi-static tests, were developed to estimate the lateral response of the analysed RC buildings. Different response parameters, which can be easily implemented in seismic risk assessment frameworks, were considered and quantified, at the portfolio level, to understand the uncertainty propagation and impact of the masonry infills’ variability on the expected performance of existing RC buildings. The results demonstrated that the variability related to the definition and modelling of the masonry infills induces significant dispersion, with respect to the mean, on the lateral response of different classes of buildings within a given portfolio. Moreover, the provided parameters for multi-linear idealised capacity curves can be used in simplified seismic fragility and risk models for existing buildings, including a proper estimate of the associated uncertainty, leading to a more accurate quantification of economic losses.

Tropical cyclone damage assessment of distributed infrastructure systems under spatially correlated wind speeds

Probabilistic damage evaluation and loss estimation of spatially distributed infrastructure systems (e.g. an electric power system or a building portfolio) under a scenario tropical cyclone must consider the spatial correlation of cyclone wind speeds to estimate their impact on the built environment statistically. Previous studies have seldom considered the impact of this spatial correlation on damage assessments of distributed civil infrastructure. In this paper, a stochastic field model is developed to capture the uncertainty of cyclone wind speeds and their spatial correlation for the North Atlantic region. A series of recorded wind speed fields of historical cyclone events are examined. The bias values between the recorded wind speeds and computed wind speeds based on a widely used cyclonic wind field model are obtained. The statistics of the wind field bias are estimated using geostatistical tools. The effect of wind speed uncertainty and spatial correlation on performance assessment of distributed infrastructure systems is illustrated using an electric power system, where the metrics of performance are damage ratio, outage ratio and outage cost.

Preventive Maintenance Model for National School Buildings in Indonesia Using a Constraint Programming Approach

The Indonesian government needs to maintain around 231,000 school buildings in active use. Such a portfolio of buildings given the diversity of locations, limited maintenance budget, and deterioration rates varied by different building conditions presents many challenges to effective maintenance planning. Many of those schools had been reported to be aging and in a degenerated condition. However, contemporary practice for the planning method of Indonesia’s building maintenance program applies reactive maintenance strategies with a single linear deterioration rate. Such methodology cannot properly guarantee the sustainability of those school buildings. Therefore, this study attempts to examine a different approach to Indonesia’s building maintenance planning by adopting a preventive maintenance strategy using the deterioration rate model proved by historical data from a previous study. This study develops an optimization model with varied deterioration rates and considers the budget limitation, by utilizing a Constraint Programming (CP) approach. The proposed model achieves the minimum maintenance cost for a real case of 41 school buildings under different deterioration rates to ensure adequate building conditions and maintain expected levels of service. Finally, research analysis also proves that this new preventive maintenance model has potential to deliver superior capability for assisting building maintenance decisions in Indonesia’s government.

Building portfolios based on machine learning predictions

This paper demonstrates that portfolio optimization techniques represented by Markowitz mean-variance and Hierarchical Risk Parity (HRP) optimizers increase the risk-adjusted return of portfolios built with stocks preselected with a machine learning tool. We apply the random forest method to predict the cross-section of expected excess returns and choose n stocks with the highest monthly predictions. We compare three different techniques—mean-variance, HRP, and 1/N— for portfolio weight creation using returns of stocks from the S&P500 and STOXX600 for robustness. The out-of-sample results show that both mean-variance and HRP optimizers outperform the 1/N rule. This conclusion is in the opposition to a common criticism of optimizers’ efficiency and presents a new light on their potential practical usage.

Estimating Building Airtightness from Data – A Case Study

The focus on energy conservation in buildings is increasing. Despite that, the yearly building renovation rate is only at around 1 %. To increase the renovation rate, new and time-efficient methods used for screening of large building portfolios’ energy saving potential are needed. In this paper, a re-engineered take on the classical energy signature method is applied to two renovated apartments in Denmark. The energy signature model relies on time-series measurements of space heat consumption, outdoor temperature, solar irradiation, and wind speed. The estimates obtained from it consist of—among other things—heat loss coefficient and wind-induced heat loss. This paper focuses on the latter. To validate the model estimate, the airtightness has been quantified by blower door-tests in both apartments: the results showed that one apartment is reasonable airtight, while the other suffers from significant air leakages. The energy signature and two other infiltration models, based on blower door test results, were compared. Good agreement between the results obtained from the data-driven energy signature and the blower door test were found. With use of a simple linear relation between the average infiltration and the blower door test result (q50), from the Danish national building code, the energy signature was found to overestimate the blower door test result (q50) by 33 % for the leaky apartment and underestimate the same air flow by 18 % for the other apartment. Both estimates are within the standard error of the infiltration model in the Danish national building code.

Assessing post-hazard damage costs to a community’s residential buildings exposed to tropical cyclones

Assessment of the potential for damage to community’s residential buildings exposed to extreme natural hazards is essential for measuring and optimising resilience enhancement strategies for communities. This paper develops a probability-based method to estimate the potential for damage and losses to a community’s residential building portfolio due to tropical cyclones (TC) by aggregating the performance of individual buildings quantitatively. The proposed approach reflects the fact that the total building damage of a community depends on the specific TC scenario and the correlation in performance of different individual buildings due to correlations in intensity in the wind field and common construction practices. The estimated damage potential, conditioned on the maximum TC wind speed, is determined from the TC fragilities. The mean value and variance of the cyclone-induced damage losses are estimated using an approximate closed-form approach. The applicability of the proposed method is demonstrated for a small coastal community in North Australia and the sensitivity of the estimated TC damage losses to the key characteristics of the TC wind field and building portfolio is investigated. It is found that correlations in building performance due to common TC demand and construction practices have a significant impact on losses to the community.

A fundamental unified framework to quantify and characterise energy flexibility of residential buildings with multiple electrical and thermal energy systems

To date, the energy flexibility assessment of multicomponent electrical and thermal systems in residential buildings is hindered by the lack of adequate indicators due to the different interpretations, properties, and requirements that characterise an energy flexible building. This paper addresses this knowledge gap by presenting a fundamental energy flexibility quantification framework applicable to various energy systems commonly found in residential buildings (i.e., heat pumps, renewables, thermal and electrical storage systems). Using this framework, the interactions between these systems are analysed, as well as assessing the net energy cost of providing flexibility arising from demand response actions where onsite electricity production is present. A calibrated white-box model of a residential building developed using EnergyPlus (including inter alia a ground source heat pump, a battery storage system, and an electric vehicle) is utilised. To acquire daily energy flexibility mappings, hourly independent, and consecutive demand response actions are imposed for each energy system, using the proposed indicators. The obtained flexibility maps give insights into both the energy volumes associated with demand response actions and qualitative characteristics of the modulated electricity consumption curves. The flexibility potential of each studied energy system is determined by weather and occupant thermal comfort preferences as well as the use of appliances, lighting, etc. Finally, simulations show that zone and water tank thermostat modulations can be suitably combined to shift rebound occurrences away from peak demand periods. These insights can be used by electricity aggregators to evaluate a portfolio of buildings or optimally harness the flexibility of each energy system to shift peak demand consumption to off-peak periods or periods of excess onsite electricity generation.

Multi-Hazard Portfolio Loss Estimation for Time-Dependent Shaking and Tsunami Hazards

Megathrust subduction earthquakes generate intense ground shaking and massive tsunami waves, posing major threat to coastal communities. The occurrence of such devastating seismic events is uncertain and depends on their recurrence characteristics (e.g. inter-arrival time distribution and parameters) as well as elapsed time since the last major event. Current standard probabilistic loss models for earthquakes and tsunamis are based on a time-independent Poisson process and uniform earthquake slip distribution. Thereby, considerations of more realistic time-dependent earthquake occurrence and heterogeneous earthquake slip distribution are necessary. This study presents an innovative computational framework for conducting a time-dependent multi-hazard loss estimation of a building portfolio subjected to megathrust subduction earthquakes and tsunamis. The earthquake occurrence is represented by a set of multiple renewal models, which are implemented using a logic-tree approach, whereas earthquake rupture characterization is based on stochastic source models with variable fault geometry and heterogeneous slip distribution. By integrating these hazard components with seismic and tsunami fragility functions, multi-hazard loss potential for a coastal community can be evaluated quantitatively by considering different possibilities of earthquake recurrence and rupture characteristics. To demonstrate the implementation of the developed time-dependent multi-hazard loss model, the Tohoku region of Japan is considered.

Analytical fragility curves for masonry school building portfolios in Nepal

Schools represent a reference point for communities in any part of the world. Therefore, their safety and resilience against natural catastrophes is of paramount importance. The recent 2015 Gorkha earthquake has unfortunately shown that Nepalese school buildings are highly vulnerable to seismic actions. Most of them are indeed constituted by low-quality unreinforced masonry (URM). The quantification of URM vulnerability is fundamental to estimate the risk associated to school building portfolios at territorial scale. This work discusses statistics available for Nepalese schools and then presents analytical fragility curves for three recurrent URM typologies covering more than 50% of the school building stock. The methodology adopted to derive fragilities is spectral-based and accounts for out-of-plane and in-plane damage potential in a single easy-to-use analytical framework. Inter-building, intra-building and record-to-record variabilities are directly considered in the analysis. The obtained fragilities integrate the studies available for the region and can be used for pre-/post-earthquake risk assessment and prioritization of interventions at country level.

New building portfolio – assessment model for sustainable and future-proof company-owned building stocks

An essential task for residential property companies to maintain their long-term viability is to secure the future rentability/marketability of their buildings. On top of that, they should also be willing to take on more responsibility for the environment and society to contribute to sustainable development. Both tasks are closely interlinked. The question that arises is how and with which tools residential property companies should control the further development of their building stocks in order to meet these requirements. Residential property companies use tools for the analysis and management of their existing building portfolio. Models that link location characteristics and building performance with rental success provide one approach. In the context of using methodological approaches such as cause-effect relationships, scoring methods and regression analysis, the paper discusses how dependencies between megatrends, rental success, location characteristic and property performance can be better modelled. The paper also attempts to introduce additional indicators that are directly related to sustainable development objectives. The result is a concept that supports the advancement of portfolio analysis methods, which both integrates additional sustainability aspects and takes into account a dynamically changing environment (shift in values towards sustainability, political framework conditions, climate change). In particular, the paper presents the interdependencies between external environment and location / building on the one hand, and location and building on the other. By modelling the correlation between the impact of the respective megatrend and the underlying portfolio assessment, the model can support portfolio managers of residential property companies in their work. Results can serve as an early warning indicator and form the basis for recommendations for action regarding the portfolio strategy.

Portfolio Approach in Green Building Certification

Hong Kong has over 9,000 high-rise buildings, of which over 1,500 are skyscrapers standing taller than 100m with 350 buildings over 150m. These buildings contribute to about 60% of the total carbon emissions of Hong Kong, therefore in Hong Kong’s Climate Action Plan 2030+, existing building energy saving is one key focus where potential for energy saving is very significant. However, it is often difficult for existing building especially aged buildings to undergo green transformation due to the existing site constraint such as space and capital limitation and at the same time requiring continuous operation of the building. This paper presents a case study of a portfolio of high-rise buildings and comprehensive building development with buildings ranging from 60 to less than 5 years in operation, situated in the most densely populated commercial district in Hong Kong undergone a “Green Body Check”, benchmarking using Hong Kong’s Green Building accreditation - BEAM Plus for Existing Building and identified appropriate green measures to improve its performance. The aim of the developer is not only trying to reduce the carbon emission of its portfolio, but to lead by example and showcasing the impact to the portfolio but also to the tenants and beyond. This study aims to highlight the importance of continuous monitoring and benchmarking, adopting green building accreditation as a framework in evaluating the existing performance and effectively identify improvement measures. Lastly to appreciate major players in the building industry to lead and drive the industry greener and more sustainable in order to tackle the global climate change crisis.

Physics‐based probabilistic seismic hazard and loss assessment in large urban areas: A simplified application to Istanbul

AbstractA set of 3D physics‐based numerical simulations (PBS) of possible earthquakes scenarios in Istanbul along the North Anatolian Fault (Turkey) is considered in this article to provide a comprehensive example of application of PBS to probabilistic seismic hazard (PSHA) and loss assessment in a large urban area. To cope with the high‐frequency (HF) limitations of PBS, numerical results are first postprocessed by a recently introduced technique based on Artificial Neural Networks (ANN), providing broadband waveforms with a proper correlation of HF and low‐frequency (LF) portions of ground motion as well as a proper spatial correlation of peak values also at HF, that is a key feature for the seismic risk application at urban scale. Second, before application to PSHA, a statistical analysis of residuals is carried out to ensure that simulated results provide a set of realizations with a realistic within‐ and between‐event variability of ground motion. PBS results are then applied in a PSHA framework, adopting both the “generalized attenuation function” (GAF) approach, and a novel “footprint” (FP)‐based approach aiming at a convenient and direct application of PBS into PSHA. PSHA results from both approaches are then compared with those obtained from a more standard application of PSHA with empirical ground motion models. Finally, the probabilistic loss assessment of an extended simplified portfolio of buildings is investigated, comparing the results obtained adopting the different approaches: (i) GMPE, (ii) GAF, and (iii) FP. Only FP turned out to have the capability to account for the specific features of source and propagation path, while preserving the proper physically based spatial correlation characteristics, as required for a reliable loss estimate on a building portfolio spatially distributed over a large urban area.