Arash Noshadravan

A Methodology for Robust Comparative Life Cycle Assessments Incorporating Uncertainty

We propose a methodology for conducting robust comparative life cycle assessments (LCA) by leveraging uncertainty. The method evaluates a broad range of the possible scenario space in a probabilistic fashion while simultaneously considering uncertainty in input data. The method is intended to ascertain which scenarios have a definitive environmentally preferable choice among the alternatives being compared and the significance of the differences given uncertainty in the parameters, which parameters have the most influence on this difference, and how we can identify the resolvable scenarios (where one alternative in the comparison has a clearly lower environmental impact). This is accomplished via an aggregated probabilistic scenario-aware analysis, followed by an assessment of which scenarios have resolvable alternatives. Decision-tree partitioning algorithms are used to isolate meaningful scenario groups. In instances where the alternatives cannot be resolved for scenarios of interest, influential parameters are identified using sensitivity analysis. If those parameters can be refined, the process can be iterated using the refined parameters. We also present definitions of uncertainty quantities that have not been applied in the field of LCA and approaches for characterizing uncertainty in those quantities. We then demonstrate the methodology through a case study of pavements.

Stochastic comparative assessment of life-cycle greenhouse gas emissions from conventional and electric vehicles

PurposeElectric vehicles (EVs) are promoted due to their potential for reducing fuel consumption and greenhouse gas (GHG) emissions. A comparative life-cycle assessment (LCA) between different technologies should account for variation in the scenarios under which vehicles are operated in order to facilitate decision-making regarding the adoption and promotion of EVs. In this study, we compare life-cycle GHG emissions, in terms of CO2eq, of EVs and conventional internal combustion engine vehicles (ICEV) over a wide range of use-phase scenarios in the USA, aiming to identify the vehicles with lower GHG emissions and the key uncertainties regarding this impact.MethodsAn LCA model is used to propagate the uncertainty in the use phase into the greenhouse gas emissions of different powertrains available today for compact and midsize vehicles in the US market. Monte Carlo simulation is used to explore the parameter space and gather statistics about GHG emissions of those powertrains. Spearman’s partial rank correlation coefficient is used to assess the level of contribution of each input parameter to the variance of GHG intensity.Results and discussionWithin the scenario space under study, battery electric vehicles are more likely to have the lowest GHG emissions when compared with other powertrains. The main drivers of variation in the GHG impact are driver aggressiveness (for all vehicles), charging location (for EVs), and fuel economy (for ICEVs).ConclusionsThe probabilistic approach developed and applied in this study enables an understanding of the overall variation in GHG footprint for different technologies currently available in the US market and can be used for a comparative assessment. Results identify the main drivers of variation and shed light on scenarios under which the adoption of current EVs can be environmentally beneficial from a GHG emissions standpoint.

The importance of incorporating uncertainty into pavement life cycle cost and environmental impact analyses

We present an approach for conducting probabilistic life cycle cost analyses (LCCA) and life cycle assessments (LCA) and demonstrate its value with case study results. We define uncertainty quantities and methods for characterizing uncertainty for different types of parameters. The approach includes leveraging outputs from Pavement-ME to characterize uncertainty in pavement performance over time. Uncertainty in the input data and scenarios is used in a Monte Carlo analysis to quantify the uncertainty in life cycle costs and environmental impacts. The probabilistic results are then used to calculate several comparative metrics, including the statistical confidence that one alternative has a lower cost or environmental impact than another alternative, and to determine the parameters that contribute most to the variance of the results. The approach enables a wide analysis of the scenario space to determine which scenarios are most relevant to the comparison of alternatives, and iterative analyses that feature refined data selected in the influential parameter analysis. We demonstrate the value of the approach and the benefits of incorporating uncertainty into LCCAs and LCAs via results from cases in the literature. 2 UNCERTAINTY QUANTITIES, CHARACTERIZATION, AND ANALYSIS 2.1 Uncertainty quantities The LCA literature (see (Lloyd & Ries 2007) for a summary) has coalesced around three types of uncertainty for both life cycle inventories (LCI) and life cycle impact assessment (LCIA) methods: parameter (uncertainty in input data), scenario (uncertainty in choices), and model (uncertainty in mathematical relationships) uncertainty. Differentiating these types of uncertainty can be challenging because of the overlap among them. All forms of uncertainty are expressed as uncertainty in a parameter value, even if there is an aggregate of multiple types of uncertainty. We found guidance on uncertainty quantities from the work of Morgan and Henrion (1990), which defines quantities used in uncertainty analyses for risk and policy analysis. They define eight types of uncertainty quantities. The five that are of most relevance to LCA and LCCA are listed in Table 1. There will likely be a single decision variable and only a few outcome criteria for each analysis. However, there will almost certainly be numerous empirical, model domain, and value parameters. Some empirical quantities will be used directly in life cycle inventories, such as quantities of material inputs or emission outputs; these are inventory parameters. However, other empirical quantities are actually model parameters, such as pavement thickness or vehicle fuel efficiency, which are used to calculate inventory parameters. Table 1. Summary of types of quantities in LCAs and LCCAs. Content adapted from (Morgan & HenrionTo advance the adoption of strategies to reduce the greenhouse gas and air pollutant emissions and urban heat island effects of pavement systems within California, a collaborative research project was conducted between the University of California Pavement Research Center (UCPRC), Lawrence Berkeley National Laboratory (LBNL), and University of Southern California (USC) to develop a tool for comparing environmental impacts of alternative decisions at the local government level in California. This paper details results of the pavement management survey, albedo of different pavement treatment materials, dynamic modeling of albedo of public pavement for different local governments in California, and life cycle assessment (LCA) models of such materials and common pavement surface treatments to capture their environmental impacts. This information is intended for use in the urban heat island LCA tool and for inputs into climate modeling in that tool’s use stage. Board (CARB) meet its shortand long-term greenhouse gas emissions reduction targets; help regions and the state meet air pollution standards; and help local governments adapt to increasing temperatures. University of California Pavement Research Center (UCPRC) in a collaboration with Lawrence Berkeley National Laboratory (LBNL) and University of Southern California (USC) conducted a study on benefits and environmental impacts of cool pavements in urban areas in California. The project, funded by California Air Resources Board and Caltrans, was aiming at developing a tool to compare alternative pavement management strategies for reducing urban heat island. With an analysis period of 50 years, the scope of the tool consisted of pavement material production, transportation to the site, pavement construction activities, the changes in urban temperature due to cool pavement strategies implemented, and the resulting changes in building energy consumption throughout the analysis period. In addition to the total primary energy demand GHG emissions, air quality impacts were investigated by comparing smog and particulate matter formation under each scenario. The research project seeks to progress the adoption of strategies to reduce the greenhouse gas and air pollutant emissions and urban heat island effects of pavement systems within California. The following tasks will be completed to achieve this objective: 1. Review the existing literature for cool pavements and pavement LCA, and convene an expert panel that will inform the goal and scope of the LCA analysis. 2. Develop a scenario-modeling tool to analyze, for a wide range of pavement characteristics, the GHG emissions inventories and the air quality, urban heat island (UHI), and building energy use impacts of pavement albedo over a wide range of California city characteristics. 3. Create a pavement strategy guidance tool for local government officials based on the scenario results that can be used to estimate the potential impact of cool pavement adoption. 4. Create clear guidelines for the continual maintenance of the modeling and guidance tools. This paper covers part of Tasks 2 and 3. 2 PAVEMENT MANAGEMENT PRACTICE A pavement management survey was conducted with several local California governments to obtain general information about the pavement treatment practices in current use. The main questions included in the pavement management survey for different local governments concerned the following: 1. The size of the pavement network managed by the local government (any units, lanemiles, square feet, centerline miles, etc.). 2. The portion of the network that in a typical year gets any kind of treatment. For example, “treat 7.5 lane-miles per year, or treat 5 percent of the network per year.” 3. The approximate breakdown of the treatments used, for example: slurry seal, 70 percent or 7 lane-miles Table 1 summarizes the results of the pavement management survey and Table 2 shows a summary of the pavement treatment surface materials, the recommended thickness or the user specifies the thickness, and approximate ranges of the expected time between replacements. 3 ALBEDO DATA FOR DIFFERENT PAVEMENT TREATMENTS There are two ASTM standard test methods for determining the solar reflectance of a surface: ASTM C1549 (Standard Test Method for Determination of Solar Reflectance near Ambient Temperature Using a Portable Solar Reflectometer) (ASTM 2009) and ASTM E1918 (Standard Test Method for Measuring Solar Reflectance of Horizontal and Low-Sloped Surfaces in the Field) (ASTM 2006). A modified method was developed by UCPRC that is in accordance with ASTM E1918. This modified method essentially follows the standard method except for two differences: it uses a dual-pyranometer instead of a single pyranometer and it uses a data acquisition system (DAS) composed of a datalogger powered by a battery and connected to a computer to record data automatically. These modifications provide a way to monitor the solar reflectivity of a surface over long time periods. Table 1. Summary of pavement treatment practice currently used by local governments in California City Public Pavement Network Lane-Miles (Centerline Miles) 1 Portion of Network Treated Every Year Portion of Each Treatment Used in Total Network Treated Slurry Seal Sand Seal Chip Seal Cape Seal Asphalt Overlay Reconstruction City of Bakersfield (1,264) 20% 75% 13% 12% City of Berkeley 453 (216) 7.4% 31% 41% 28% City of Chula Vista (461) 3.9% 28.3% 46.4 % 0.5% 21.8% 3% City of Fresno 2 (1,548) 1.3% 100% City of Los Angeles 28,000 7.4% 60.7% 35.4% 3.9% City of Richmond 576 5.2% 47.1% 0.7% 0.5% 45.9% 5.9% City of Sacramento 3,065 4.3% 82.4% 17.6% City of San Jose 4,264 5% 80% 20% Average 6.8% 41.2% 9.4% 5.9% 0.1% 36.8% 6.6% 1 Use multiplier 2.2 to convert centerline miles to lane-miles. The lane width is assumed 12 ft. 2 Forty (40) centerline miles asphalt overlay up to 2009, then 20 centerline miles asphalt overlay since 2009. Table 2. Summary of pavement treatments Treatment Type Range of Treatment Life 1 Thickness (mm) or Application Rate (Asphalt, Aggregate) Conventional Asphalt Concrete Overlay 2–12 years (1–2 inch) Varies with traffic and design (> 2 inches) User gives thickness Rubberized Asphalt Concrete Overlay 2–12 years (1–2 inch) Varies with traffic and design (> 2 inches) User gives thickness Asphalt Concrete or Overlay with Reflective Coating 2–12 years (1–2 inch) Varies with traffic and design (> 2 inches) User gives thickness Chip Seal 1–10 years 9 mm stone Slurry Seal 1–10 years 6 mm Cape Seal 2–15 years Chip plus slurry Fog Seal 1–5 years Sand Seal 1–6 years Portland Cement Concrete Varies with traffic and design User gives thickness Whitetopping 10–20 years (3–5 inches) Varies with traffic and design (> 5 inches) User gives thickness User Defined Material User input User input 1 Adapted from Treatment Selection for Flexible Pavements. www.pavementpreservation.org/library/getfile.php?journal_id=941 for local streets, parking lots, etc., not for highways. Albedo data were collected from three sources: LBNL, the Federal Highway Administration (FHWA), and UCPRC. The LBNL Heat Island Group has compiled a pavement albedo database that includes sets of measurements from laboratory samples of various cool pavement treatments taken using spectrophotometer, from field samples taken using the pyranometer test method (ASTM E1918), and compiled from various sources such as field testing and literature. An on-going FHWA project, entitled “Quantifying Pavement Albedo” (Solicitation Number: DTFH61-12-R-000050), is measuring the albedo of different pavement materials. Some initial albedo data were provided by the project contractor, Iowa State University, and included asphalt and concrete materials with different ages measured using the pyranometer test method (ASTM E1918). An on-going study on cool pavements being conducted at UCPRC is devoted to investigating the thermal behavior and cooling effect of different pavement types (including asphalt, concrete, and block paver) and different designs (conventional impermeable and novel permeable designs), to using the field measurement data to validate the heat-transfer modeling, to employing the validated model to simulate the thermal behavior and cooling effect of different pavements in various contexts (climates and surroundings), and to examining the effect of cool pavements on human thermal comfort (Li et al. 2013). Nine test sections were the primary sections for albedo measurements at UCPRC. These nine test sections include three different pavement surfacing materials, namely interlocking concrete pavers (surfacing Type A), open-graded asphalt concrete (surfacing Type B), and portland cement concrete (surfacing Type C). More details on the materials can be found in reference (Li et al. 2013). Along with these nine sections, several extra pavement sections with conventional impermeable asphalt and concrete surfacing were also included in the study for the field measurement of albedo. For comparison, albedo has also been measured on other land cover materials, including gravel, soil and grass. Some of these materials were of different ages when solar reflectivity measurements were conducted on them. In May 2014, more field albedo measurements were performed around Davis, California, and these included slurry seal, fog seal, cape seal, chip seal, and more PCC and AC materials. The steady-state (the final stable albedo value remained after a certain time of weathering and trafficking) albedo of the different pavement materials summarized across all data sources are shown in Table 3. Table 3. Summary of steady-state albedo of different pavement treatment materials with different data sources Material Type Albedo (LBNL) Albedo (FHWA) Albedo (UCPRC) Albedo (Typical) Range Avg. Range Avg. Range Avg. Range Avg. Asphalt Concrete or Overlay 0.1–0.15 0.12 0.05–0.15 0.1 0.06–0.15 0.1 0.05–0.15 0.1 Asphalt Concrete or Overlay with Re

Operational Strategies for Increasing Secondary Materials in Metals Production Under Uncertainty

Increased use of secondary raw materials in metal production offers several benefits including reduced cost and lowered energy burden. The lower cost of secondary or scrap materials is accompanied by an increased uncertainty in elemental composition. This increased uncertainty for different scraps, if not managed well, results in an increased risk that the elemental concentrations in the final products fall outside customer specifications. Previous results show that incorporating this uncertainty explicitly into batch planning can modify the potential use of scrap materials while managing risk. Chance-constrained formulations provide one approach to uncertainty-aware batch planning; however, typical formulations assume normal distributions to represent the compositional uncertainty of the materials. Compositional variation in scrap materials has been shown to have a skewed distribution, and therefore, the performance of these models, in terms of their ability to provide effective planning, it may then be heavily influenced by the structure of the compositional data used. To address this issue, this work developed several approximations for skewed distributional forms within chance-constrained formulations. We explored a lognormal approximation based on Fenton’s method; a convex approximation based on Bernstein inequalities; and a linear approximation using fuzzy set theory. Each of these methods was formulated and case studies executed using compositional data from an aluminum remelter. Results indicate that the relationship between the underlying structure/distribution of the compositional data and how these distributions are formulated in batch planning can modify the use of secondary raw materials.

A Lifecycle Cost Analysis of Residential Buildings Including Natural Hazard Risk

AbstractDespite isolated efforts in the cost assessment of design strategies for energy-efficient and resilient buildings, there is still a need for an integrated assessment that incorporates major...

A probabilistic mesoscale damage detection in polycrystals using a random matrix approach

This article is concerned with a probabilistic mesoscale damage detection in polycrystals. For this purpose, we make use of a stochastic model describing the linear elasticity matrix of material at the mesoscale. The model is constructed using a maximum entropy principle and random matrix theory and allows one to directly construct a probabilistic model for the system random matrices characterizing the constitutive behavior of the system. First, the theoretical framework and upscale scheme in the construction of the model are briefly reviewed. For each case of healthy and damaged materials, where the damage is introduced in the form of intergranular microcavities, the random matrix model is calibrated by performing simulations on an ensemble of statistical volume elements of microstructure. The calibrated models are then used in a simple coarse-scale simulation in order to explore the sensitivity of the model in detecting the location of mesoscale damages. The result shows that in most cases, one can identify the location of cracks by comparing the probabilistic description of a suitable response quantity of interest predicted for both healthy and damaged systems. Such a probabilistic description is suitable for detecting signature of fine-scale defects where the consequences are reflected at the coarse scale in the form of random fluctuations around the mean behavior. The model can be used as a predictive tool in the context of structural health monitoring and damage prognosis of metallic systems.

Identfication and validation of a stochastic model for mesoscale material description of metallic polycrystals

This paper is concerned with the identi cation and validation of a stochastic mesoscale material description for metallic polycrystals. For this purpose we make use of a bounded random matrix model characterizing the mesoscale elasticity tensor of heterogeneous material. The bounded random matrix exhibits uctuations that are connected to ne scale features through a calibration process performed using a micromechanical framework. The experimental calibration and validation of the model requires testing of a statistically meaningful number of samples possessing statistically similar microstructures. To that aim we rst employ a statistical model for generating 2D polycrystals that are consistent with the available microstructural measurements on the geometry and crystallographic orientations. The calibration of the mesoscale probabilistic model using realizations of digitally generated polycrystalline microstructures is brie y discussed. We then present validation of the probabilistic model from simulated data resulting from subscale simulations. It is found that the probabilistic model for bounded mesoscale elasticity matrix is adequate to predict the response quantity of interest. The scatters in the model predictions are found to be consistent with the ne scale response. The proposed probabilistic model combined with the nite element analysis can be used as a predictive tool in the system level in the context of structural health monitoring and damage prognosis.

Characterization of random heterogeneities in polycrystalline microstructures using wave propagation simulation

The quantification of variability in the mechanical behavior of metallic materials is important in the design and reliability assessment of mechanical components. A combination of experimental and computational approaches is often required to alleviate the experimental burden and lack of data in constructing a probabilistic formalism for material design. The present work aims at integrating material characterization and computational modeling for the evaluation of variability in the elastodynamic response of random polycrystals. First, a procedure is presented for simulation of random 2D polycrystalline microstructures from limited experimental data. Second, the capability of the numerical model in capturing the variation of the scattered waves due to the random heterogeneities is investigated by introducing a suitable quantity of interest characterizing the intensity of the fluctuations of the stochastic waveforms. Two important types of heterogeneities are considered. The first is the inherent heterogeneity due to the mismatch in the grain orientations. The second is the heterogeneity due to fine scale defects in the form of random intergranular micro-cavities. The numerical model presented in this paper can be useful for the interpretation of experimental ultrasonic measurements for random heterogeneous material. The result is also applicable to the validation of multiscale probabilistic models for material prognosis.