Information technology tools for coping with uncertainty and imprecision in environment-oriented decision support systems

Piotr Holnicki (Systems Research Institute)    Piotr.Holnicki >at>

Description:The recent European regulations on air quality standards assume an intensive application of air quality forecasting models for:
  1. assessment of ambient air quality,
  2. quantitative evaluation of the violation of human health or environmental protection thresholds,
  3. simulation of the respective mitigation solutions.
The more complex Integrated Assessment Models (IAM) provide tools for assessment and analysis of the air quality related problems as well as for design of the efficient pollution abatement strategies.
The architecture of an integrated system includes the input databases (emission inventories and projections, emission reduction technologies, efficiency and the unit costs of emission abatement technologies), pollution transport model (analysis of emission-concentration relationships), decision module (objectives and constraints of the decision problem, cost-benefit or cost-effectiveness analysis, formulation of the optimization problem, and implementation of the algorithm). Due to the interdisciplinary character and significant complexity of such decision support systems, as well as mixed continuous-discrete problem formulation, finding of satisfactory solutions can be only achieved using heuristic approaches. Moreover, there exist numerous sources of uncertainty in the modeling of environmental effects of atmospheric pollution (e.g. uncertainty and imprecision of meteorological and emission data, or that related to the model parameters). The resulting uncertainty affects the system generated measures of adverse environmental effects as well as the related protection strategies.
The project addresses the problems of computer modeling of air pollution dispersion and also the related task of supporting strategic decisions of reduction or minimization of the respective harmful effects. The realization of the project, due to its complex character, requires an intensive use of computer modeling and simulation techniques, imprecision and uncertainty handling, analysis of decision support and management problems in uncertainty and imprecision conditions which may involve the use of adequate techniques, like the fuzzy set or the possibility rooted methods.