The use of mixed integer programming for the evaluation of some alternate air pollution abatement policies

Abstract

The model presented in this paper must be considered to be an effective tool for establishing bases for corrective alternatives for an abatement problem of air pollution. It should also be considered a very useful instrument for qualifying, within the development policies for a given area, the standards that are more and more indispensible for protecting our air environment. It should be noted that the basic statistical parameter considered in the formulation of the model is the maximum probability allowed that the concentration in a given grid square exceed the maximum limit allowed, in contrast to models which use averages as their standards of quality. This methods avoids the danger of large concentrations being masked with smaller concentrations. This probability depends conjointly on the probabilistic matrix of the typology by which the different meteorological factors have been stratified and the probability that for a theoretical concentration estimated on the basis of a predicted set of emissions the real concentration might exceed the maximum limit permited. It is of interest to point out that in order to estimate the concentration in each grid square, stochastic diffusion models have been used for each meteorological stratum, depending on the emissions, so that the tabular form of the model is in function with the emitter grid squares. The criterium which minimize the model is the weighted reduction of the emission levels for each contributing grid square in accord with the effect it has on the pollutant concentration in the sum of the grid squares which make up the polluted area.