Air Quality Planning: A General Chance‐Constraint Model

Abstract

One complex aspect of environmental planning is the stochastic planning nature of environmental impacts. Planners can evaluate alternatives more fully if the stochastic impacts of various alternatives are incorporated into the analyses. Chance‐constrained modeling techniques that are currently applied to real problems cannot incorporate many stochastic environmental impacts because of random variables that are non‐normal and statistically dependent. In addition, confidence limits of distribution parameters cannot be reflected in solutions. The planning technique proposed in this paper uses linear programming and simulation interactively to resolve these difficulties. The technique is demonstrated using a model that selects power plant sites to fulfill two planning objectives: (1) satisfaction of air quality standards; and (2) minimization of the costs of flue gas treatment and electrical transmission. The stochastic environmental impacts that must be controlled are the ${\mathrm{SO}}_2$ impacts from power plants. Relying on the nonparametric method of order statistics, this modeling technique can incorporate random variables of any distribution. Furthermore, confidence limits selected by the model user are obtained. This modeling technique is applicable to any environmental planning problem where the stochastic environmental impacts can be modeled with impact coefficients.