Case study

An air transportation association

Airport emissions modeling project

In response to local, state and federal air pollution authorities’ injunctions to reduce airport emissions, an air industry consortium with representatives from government, airport associations and private business wanted to gain a better understanding of the quantity of emissions produced at airports. PA’s Market Analytics team was asked to develop uncertainty estimates around each emission input, and for the larger model describing variation in airport emissions, using uncertainty analysis. 

PA’s Market Analytics team developed a method to specify confidence intervals surrounding emissions estimates. The uncertainty surrounding a class of events, such as the emission of pollutants, is characterized probabilistically, as a proportion of total events expected to occur within a specified range. In the terminology of statistics, this is the problem of establishing a confidence interval for the measurement of an event. Wherever a variety of factors contribute to the event, each with its own distribution of probabilities, uncertainty analysis combines those probabilities in order to determine a probability distribution for aggregated events. 

PA created a computer algorithm that could calculate emissions levels for every possible combination of attributes, or make a random sampling of 100 million scenarios when the number of permutations reached into the trillions. By attaching probabilities to the scenarios it was possible to construct a probability distribution and, using probability theory, find confidence intervals for the estimated outcomes.

The study provided numerical estimates of airport emissions uncertainty. This allowed the air industry association to evaluate its emissions model and develop an equitable framework for the aviation industry’s participation in voluntary initiatives to reduce air pollutants at US airports.