H. Christopher Frey

Identification and Review of Sensitivity Analysis Methods

Identification and qualitative comparison of sensitivity analysis methods that have been used across various disciplines, and that merit consideration for application to food-safety risk assessment models, are presented in this article. Sensitivity analysis can help in identifying critical control points, prioritizing additional data collection or research, and verifying and validating a model. Ten sensitivity analysis methods, including four mathematical methods, five statistical methods, and one graphical method, are identified. The selected methods are compared on the basis of their applicability to different types of models, computational issues such as initial data requirement and complexity of their application, representation of the sensitivity, and the specific uses of these methods. Applications of these methods are illustrated with examples from various fields. No one method is clearly best for food-safety risk models. In general, use of two or more methods, preferably with dissimilar theoretical foundations, may be needed to increase confidence in the ranking of key inputs.

On-Road Measurement of Vehicle Tailpipe Emissions Using a Portable Instrument

Abstract A study design procedure was developed and demonstrated for the deployment of portable onboard tailpipe emissions measurement systems for selected highway vehicles fueled by gasoline and E85 (a blend of 85% ethanol and 15% gasoline). Data collection, screening, processing, and analysis protocols were developed to assure data quality and to provide insights regarding quantification of real-world intravehicle variability in hot-stabilized emissions. Onboard systems provide representative real-world emissions measurements; however, onboard field studies are challenged by the observable but uncontrollable nature of traffic flow and ambient conditions. By characterizing intravehicle variability based on repeated data collection runs with the same driver/vehicle/route combinations, this study establishes the ability to develop stable modal emissions rates for idle, acceleration, cruise, and deceleration even in the face of uncontrollable external factors. For example, a consistent finding is that average emissions during acceleration are typically 5 times greater than during idle for hydrocarbons and carbon dioxide and 10 times greater for nitric oxide and carbon monoxide. A statistical method for comparing on-road emissions of different drivers is presented. Onboard data demonstrate the importance of accounting for the episodic nature of real-world emissions to help develop appropriate traffic and air quality management strategies.

Uncertainties in predicted ozone concentrations due to input uncertainties for the UAM-V photochemical grid model applied to the July 1995 OTAG domain

The photochemical grid model, UAM-V, has been used by regulatory agencies to make decisions concerning emissions controls, based on studies of the July 1995 ozone episode in the eastern US. The current research concerns the effect of the uncertainties in UAM-V input variables (emissions, initial and boundary conditions, meteorological variables, and chemical reactions) on the uncertainties in UAM-V ozone predictions. Uncertainties of 128 input variables have been estimated and most range from about 20% to a factor of two. 100 Monte Carlo runs, each with new resampled values of each of the 128 input variables, have been made for given sets of median emissions assumptions. Emphasis is on the maximum hourly-averaged ozone concentration during the 12–14 July 1995 period. The distribution function of the 100 Monte Carlo predicted domain-wide maximum ozone concentrations is consistently close to log-normal with a 95% uncertainty range extending over plus and minus a factor of about 1.6 from the median. Uncertainties in ozone predictions are found to be most strongly correlated with uncertainties in the NO2 photolysis rate. Also important are wind speed and direction, relative humidity, cloud cover, and biogenic VOC emissions. Differences in median predicted maximum ozone concentrations for three alternate emissions control assumptions were investigated, with the result that (1) the suggested year-2007 emissions changes would likely be effective in reducing concentrations from those for the year-1995 actual emissions, that (2) an additional 50% NOx emissions reductions would likely be effective in further reducing concentrations, and that (3) an additional 50% VOC emission reductions may not be effective in further reducing concentrations.

Characterizing, simulating, and analyzing variability and uncertainty: An illustration of methods using an air toxics emissions example

Abstract Variability is the heterogeneity of values with respect to time, space, or a population. Variability may be quantified using frequency distributions. Uncertainty arises due to lack of knowledge regarding the true value of a quantity. Uncertainty may be quantified using probability distributions. These two concepts are distinct and, therefore, should be treated separately in an analysis. In this paper, methods for quantifying variability and uncertainty in model inputs, simulating variability and uncertainty in a model, and analyzing the results are presented. The method is demonstrated via an illustrative case study involving emissions characterization. The analysis of input data illustrates methods for characterizing variability and uncertainty when uncertainties arise due to sampling error (i.e., small sample sizes) or measurement error, as well as for dealing with differences in averaging times among multiple datasets. Correlation structures between sampling distributions representing uncertai...

Coal blending optimization under uncertainty

Abstract Coal blending is one of several options available for reducing sulfur emissions from coal-fired power plants. However, decisions about coal blending must deal with uncertainty and variability in coal properties, and with the effect of off-design coal characteristics on power plant performance and cost. To deal with these issues, a multi-objective chance-constrained optimization model is developed for an illustrative coal blending problem. Sulfur content, ash content and heating value are treated as normally distributed random variables. The objectives of the model include minimizing the: 1) expected (mean) costs of coal bending; 2) standard deviation of coal blending costs; 3) expected sulfur emissions; and 4) standard deviation in sulfur emissions. The cost objective function includes coal purchasing cost, ash disposal cost, sulfur removal cost, and fuel switching costs. Chance constraints include several risk measures, such as the probability of exceeding the sulfur emission standard. Several results are presented to illustrate the model.

Real-World In-Use Activity, Fuel Use, and Emissions for Nonroad Construction Vehicles: A Case Study for Excavators

Abstract A study design was developed and demonstrated for deployment of a portable emission measurement system (PEMS) for excavators. Excavators are among the most commonly used vehicles in construction activities. The PEMS measured nitric oxide, carbon monoxide, hydrocarbons, carbon dioxide, and opacity-based particulate matter. Data collection, screening, processing, and analysis protocols were developed to assure data quality and to quantify variability in vehicle fuel consumption and emissions rates. The development of data collection procedures was based on securing the PEMS while avoiding disruption to normal vehicle operations. As a result of quality assurance, approximately 90% of the attempted measurements resulted in valid data. On the basis of field data collected for three excavators, an average of 50% of the total nitric oxide emissions was associated with 29% of the time of operation, during which the average engine speed and manifold absolute pressure were significantly higher than corresponding averages for all data. Mass per time emission rates during non-idle modes (i.e., moving and using bucket) were on average 7 times greater than for the idle mode. Differences in normalized average rates were influenced more by intercycle differences than intervehicle differences. This study demonstrates the importance of accounting for intercycle variability in real-world in-use emissions to develop more accurate emission inventories. The data collection and analysis methodology demonstrated here is recommended for application to more vehicles to better characterize real-world vehicle activity, fuel use, and emissions for nonroad construction equipment.

Requirements and Incentives for Reducing Construction Vehicle Emissions and Comparison of Nonroad Diesel Engine Emissions Data Sources

Nonroad construction vehicles and equipment powered by diesel engines contribute to mobile source air pollution. The engines of this equipment emit significant amounts of carbon monoxide, hydrocarbons, nitrogen oxides, and particulate matter. These pollutants pose serious problems for human health and the environment. Therefore, it is necessary to regulate and control the levels of these pollutants. Furthermore, there are emerging requirements and incentives for “greening” of construction vehicle fleets and operations. Currently, there are two types of standards that regulate air pollution for these types of vehicles: technological standards for engines and quality standards for air. It is also necessary to quantify the levels of emissions that nonroad construction vehicles and equipment produce. Quantification may be based on existing data sources (such as the EPA NONROAD model) or by collecting data directly from the vehicles as they work in the field. The purpose of this paper is to introduce the chall...

Speed- and Facility-Specific Emission Estimates for On-Road Light-Duty Vehicles on the Basis of Real-World Speed Profiles

Estimating the emissions consequences of surface transportation operations is a complex process. Decision makers need to quantify the air quality impacts of transportation improvements aimed at reducing congestion on the surface street network. This often requires the coupling of transportation and emissions models in ways that are sometimes incompatible. For example, most macroscopic transportation demand and land use models, such as TransCAD, TranPlan, and TRANUS, produce average link speed and link vehicle miles traveled (VMT) by vehicle and road class. These values are subsequently used to estimate link-based emissions by using standard emissions models such as the U.S. Environmental Protection Agencys MOBILE6 model. In contrast, recent research with portable emissions monitoring systems indicates that emissions are not directly proportional to VMT but are episodic in nature, with high-emissions events coinciding with periods of high acceleration and speed. This research represents an attempt to brid...

Comprehensive Field Study of Fuel Use and Emissions of Nonroad Diesel Construction Equipment

Limited field data are available for analyses of fuel use and emissions of nonroad diesel construction equipment. This paper summarizes the results of field research that used a portable emissions monitoring system to collect fuel use and emissions data from eight backhoes, six bulldozers, three excavators, four generators, six motor graders, three off-road trucks, one skid-steer loader, three track loaders, and five wheel loaders while they performed various duty cycles. These tests produced approximately 119 h of field data for petroleum diesel and approximately 48 h for B20 biodiesel. Engine attribute data including horsepower, displacement, model year, engine tier, and engine load were collected to determine these factors’ influence on fuel use rates and emission rates of nitrogen oxides, hydrocarbons, carbon monoxide, carbon dioxide, and opacity. Mass per time fuel use rates were developed for each item of equipment, as were mass per time and mass per fuel used emission rates for each pollutant. For petroleum diesel, fuel use and emission rates of each pollutant were found to increase with engine displacement, horsepower, and load and to decrease with model year and engine tier. The results were qualitatively similar for B20 biodiesel. Fuel-based emission rates were found to have less variability and less sensitivity to engine size and load than time-based emission rates. Where possible, development of emission inventories based on fuel consumed, rather than time of activity, is preferred.

Sensitivity Analysis of a Two-Dimensional Probabilistic Risk Assessment Model Using Analysis of Variance

This article demonstrates application of sensitivity analysis to risk assessment models with two-dimensional probabilistic frameworks that distinguish between variability and uncertainty. A microbial food safety process risk (MFSPR) model is used as a test bed. The process of identifying key controllable inputs and key sources of uncertainty using sensitivity analysis is challenged by typical characteristics of MFSPR models such as nonlinearity, thresholds, interactions, and categorical inputs. Among many available sensitivity analysis methods, analysis of variance (ANOVA) is evaluated in comparison to commonly used methods based on correlation coefficients. In a two-dimensional risk model, the identification of key controllable inputs that can be prioritized with respect to risk management is confounded by uncertainty. However, as shown here, ANOVA provided robust insights regarding controllable inputs most likely to lead to effective risk reduction despite uncertainty. ANOVA appropriately selected the top six important inputs, while correlation-based methods provided misleading insights. Bootstrap simulation is used to quantify uncertainty in ranks of inputs due to sampling error. For the selected sample size, differences in F values of 60% or more were associated with clear differences in rank order between inputs. Sensitivity analysis results identified inputs related to the storage of ground beef servings at home as the most important. Risk management recommendations are suggested in the form of a consumer advisory for better handling and storage practices.