INVESTIGATION AND DEVELOPEMNET OF ENVIRONMENTAL MODELING FOR MANAGEMENT OF REGIONAL AIR POLLUTION CONTROL SYSTEM
Environmental pollution problems associated with social and economic development at regional- or industrial-scales have become critical concerns facing both national and local governments around the world for several decades. These issues usually relate to a number of factors, with multi-source, multi-stage, multi-objective, interactive, and uncertain characteristics. For such complex environmental systems, when decisions regarding management practices are to be made, integrated consideration that incorporates various processes and factors within a general framework rather than examining them in isolation would be useful for generating effective management schemes and strategies.
We want to develop a set of environmental modeling methodologies have been developed and applied to the management of several type of environmental pollution control systems, which include the development of an inexact chance-constrained mixed-integer linear programming (ICCMILP) model and its application to a case study for managing a regional environment system.
Environmental pollution problems associate with rapid social and economic development have been a critical concern facing both national and local governments around the world for several decades. The increasing environmental issues can not only pose a variety of impacts and risks on public health, but also lead to significant vulnerability for sustainable regional development in the future. With the demands of both deepening regional development and raising public awareness of environmental problems, planners and decision makers have been facing increasing pressures for more effectively responding to a number of environmental concerns. Consequently, identifications of decision schemes with sound environmental and socio-economic efficiencies need for effectively managing pollution control systems (PCS). In this study, according to spatial size of study area and human activities involved in the study systems, the PCS is sorted into one type: regional pollution control systems (RPCS). This categorization can provide convenience for tasks of model development and case application.
Management of Regional Pollution Control Systems
In the management of regional pollution control systems (RPCS), there are a number of factors to be considered by planners and decision makers, such as environmental, economic, social, technical, legislational, institutional and political issues, as well as the use and conservation of resources (Wilson 1995). These factors may affect the behavior of the regional system, and lead to conflicts among different system components. A variety of system activities are interacted with and interrelated to each other, resulting in complicated pollution control systems with interactive, dynamic, multi-objective and uncertain features. As a result, the decision-making process may require a sound understanding of the significant contributors to regional environmental problems and the way a pollution control system will react to particular problems.
Environmental simulation and optimization models have been often used to satisfy the above-mentioned requirements. These models were primarily developed for reflecting the impacts of human activities, exploring interactions among various system components, making tradeoffs among different objectives, and thus supporting decisions that lead to maximized environmental and economic efficiencies (Thomas et al. 2001). One of the biggest dilemmas encountered is that many modeling parameters generally show high degrees of intrinsic variability and substantial levels of uncertainty, since a number of system components and their interrelationships in real-world problems may not be known with certainty (Robin et al. 1999; Woodbury and Dudicky 2000; Rotmans et al. 2002). This makes the study systems more complicated to quantify. Thus, effective examination, identification and reflection of the uncertainties, which is essential for generating highly efficient and reliable outcomes, have been a major concern in the development of environmental modeling methodologies for supporting management of the RPCS. Previously, the major approaches for dealing with uncertainties included fuzzy, stochastic, and interval programming (Zadeh 1998; Morgan et al. 1993; Huang et al. 1996,1999; Jairaj and Vedula 2002; Ballestero 2001; Mohan and Nguyen 2001). These methods have been extensively applied for the planning of environmental pollution control systems (Teller 1989; Werczberger 1974; Singpurwalla 1975; Guldmann and Shefer 1977; Macchiato et al. 1994; Teng and Tzeng 1994; Lejano et al. 1997; Jairaj and Vedula 2000; Chang and Chen 2000; Kouwenbery 2001). However, the above-mentioned individual methods have varied limitations in managing uncertain inputs/outputs and reflecting dynamic and interactive system characteristics. For example, several researchers considered uncertainties through the applications of stochastic methods based on probability theory. Under this situation, some uncertainties are expressed as probability density functions (PDFs), while the others are assigned deterministic values followed by some post-optimality analyses (Fortin and McBean 1983; Guldmann 1999; Ellis 1991; Ellis and Bowman 1994). The simplification into deterministic values could possibly mislead the planners, resulting in poor environmental management recommendations (Beale 1955; Macchiato et al. 1999).
Therefore, it is desired that an integration of the existing uncertainty-handling methods be considered. This is due mainly to the following facts: (a) while some uncertain parameters can be presented as PDFs, the others are often not detailed enough to be presented as
PDFs; (b) it may be easier for engineers and planners to specify
possibilistic distributions than to identify probabilisitc ones that are often unavailable; (c) uncertainties in some parameters can only be represented as intervals under situations where both probability and possibility information are unavailable; (d) it is hard to solve large stochastic models with many uncertain parameters being expressed as PDFs, even if these functions are available; and (e) the extensive data requirements for specifying the needed distributions may affect practical applicability of many models. Thus, effective incorporation of different methods within a general framework would be a potential alternative for allowing the uncertainties to be effectively communicated and reflected in the modeling efforts.
As an extension of the previous efforts in studying the management of environmental pollution control systems, this research puts its emphasis on the development of innovative environmental optimization and simulation modeling approaches that reflect parameter uncertainties, as well as their applications to a number of hypothetical and real-world case studies. This objective entails the following aspects:
(i) To develop an inexact chance-constrained mixed-integer linear programming (ICCMILP) model, and apply it to a case study for managing a regional air pollution control system.
(ii) To develop a fuzzy-stochastic robust programming (FSRP) model, and apply it to a case study for the management of a regional air pollution control system.
(iii) To develop a system dynamics model (ErhaiSD) assoicated with a multi-objective program, and apply it to the planning of a watershed environmental system
Statement of Problems
Regional air pollution problems have been the major concern for almost one century since they not only directly relate to human activities and economic development, but also pose a serious threat on public health. There are many emission sources of air pollutants in a regional system. Among them, power plants and industries are major contributors. For air pollution control projects, it is desired that effective planning and management with an efficient and cost-effective manner be undertaken for keeping local air quality at a healthy level.
The phenomenon of regional air pollution involves many factors, such as properties of pollutants, locations of emission sources and receptors, meteorological conditions, and control measures. It involves a sequence of events (i) generation of pollutants and their emission, (ii) pollutant transport, transformation and removal in the atmosphere, and (iii) effects on human beings, materials and ecosystems (Flagan and Seinfeld 2000). For a regional system, air pollution problems are usually characterized by one or several large sources and a number of relatively small sources, leading to adverse impacts on receptors (Boubel et al. 1999). Large sources, such as power plants and industrial sources, can pose a series of threats on the surrounding communities, especially under stable meteorological conditions that cause portions of the plumes to reach the ground with high concentrations. Since it is generally either economically infeasible or technically impossible to design processes leading to zero emission of air pollutants, local authorities and decision makers always seek to control the emissions to levels at which the effects are minimized. Thus, an air pollution control strategy for a region should contain a specification of allowable levels of pollutant emissions and a scheme for making efficient use of environmental loading capacity. To obtain such a strategy it is necessary to be able to estimate the atmospheric fate of the emissions, and thus the ambient concentrations, so that the concentrations can be compared with those considered to give rise to adverse effects. The ultimate mix of control actions and devices employed to achieve this objective might have to be decided on an economic basis that means, seeking cost-effective control measures plays a key role in air pollution abatement planning for identifying promising management decisions
In this study, we develop a set of environmental modeling methodologies have been developed and applied to the management of several type of environmental pollution control systems, including a regional air pollution control system, a watershed environmental management system, and two aquifer-contaminated subsurface systems. For each method, research efforts to devote to system conceptualization, model development, complexity analysis, uncertainty reflection, solution-algorithm study, and case application. One major purpose of this study is to help decision makers to effectively manage the environmental systems through providing rational decision support bases.
.Also we focuses on dealing with uncertainties existed in a regional air quality management system .This study integrate the existing fuzzy-robust programming and chance-constrained programming methods into a general optimization modeling framework. Thus, fuzzy coefficient A and uncertain capacity B with random distributions could be effectively represented within the optimization process. The solution process delimits the fuzzy decision space as a more confidential one by specifying the uncertainties through dimensional enlargement of its original fuzzy constraints. The delimited decision space makes the model solutions more robust and realistic. The development method firstly applies to a simple example for illustrating its applicability and the detailed solution process, and then further applies to a hypothetical case study. The obtain results can support the planning of air-quality-related activities, and can reflect complex tradeoffs between environmental and economic considerations. The results show that willingness to pay higher operating costs will guarantee meeting environmental objectives; however, a desire to reduce the costs will run into the risk of potentially violating the emission and/or ambient-air-quality standards.
Abriola, L. M., Dekker, T. J. and Pennell, K. D., (1999). Surfactant enhanced solubilization of residual dodecane in soil columns, 2, Mathematical modeling, Environmental Science and Technology, 27(12), 2341-2351.
Acevedo, J., and Pistikopoulos, E. N., (2000). A hybrid parametric/stochastic programming approach for mix-integer linear problems under uncertainty, Industrial & Engineering Chemistry Research, 36, 2262-2270.
Beale, E. M. L., (1985), On minimizing a convex function subject to linear inequalities,J. Royal Statistical Soc. Ser. B, 17(2), 173-180
Ellis, J. H., McBean, E. A., and Farquhar, G. J., (1985). Chance-constrained stochastic linear programming model for acid rain abatement - I. Complete colinearity and noncolinearity, Atmospheric Environment, 19, 925-937.
Fountain, J. C, Starr, R. C, Middleton, T., Beikirch, M., Taylor, C, and Hodge, D., (2001). A controlled field test of surfactant-enhanced aquifer remediation. Groundwater, 34, 910-916.
Guldmann, J. -M., (1993). Interactions between weather stochasticity and the locations of pollution sources and receptors in air quality planning: A chance-constrained approach, Geo. Anal., 18(3), 198-214.
Guldman, J. -M., (1995). Chance-constrained dynamic model of air quality management, Fuzzy Sets and Systems, 114(5), 1116-1126.
Haas, C, Anotai, J., and Engelbrecht, R., (1996). Monte Carlo assessment of microbial risk associated with landfilling of fecal material, Water Envir. Res. 68(7), 1123- 1131.
Haith, A. D., (1982). Environmental Systems Optimization, John & Sons, Inc., New York. Haley, J. L., Hanson, B., Enfield, C, and Glass, J., (1991). Evaluating the effectiveness of ground water extraction systems, Ground Water Monit. Rev., 111(3), 119-124.
Hallenbeck, W. H. and Flowers, R. E., (1997). Risk analysis for work exposure to benzene, Environmental Management, 16, 415-520
Infanger, G., and Morton, D. P., (1999). Cut sharing for multistage stochastic linear programs with interstage dependency, Mathematical Programming, 75, 241-251.
Kuhner, J. and Harrington, J. J., (1989). Mathematical models for developing regional solid waste management policies, Engineering Optimization, Vl(4), 237-256.
Lai, Y. -J, and Hwang, C. L., (1999). Fuzzy Mathematical Programming. Dordrecht, Netherlands: Springer-Verlag. Lake, L.W., (1989). Enhanced Oil Recovery, Prentice-Hall, Englewood Cliffs, N. J. Lee, E. S., and Li, R. J., (2000). Fuzzy Sets and Systems, 53, 275-288.
Lein, J. K., (1998). Expressing environmental risk using fuzzy variables: a preliminary examination, Environmental Professional 14, 257-67
Miller, C. T., Poirier-McNeill, M. M., and Mayer, A. S., (1995). Dissolution of trapped nonaqueous phase liquids: Mass transfer characteristics, Water Resources
Research, 26(11), 2783-2796.
Powers, S. E., (1995). Dissolution ofNonaqueous Phase Liquids in Saturated Subsurface Systems, PhD Dissertation, The University of Michigan, Ann Arbor, MI. Rakes, T. R., and Reeves, G. P., (1995). Selecting tolerances in chance-constrained programming approach, J. Operations Research Soc, 22, 359-369.
Rotmans, J., van Asselt, and Marjolein, B. A., (2003). Uncertainty management in integrated assessment modeling: towards a pluralistic approach, Environmental monitoring and assessment, 69(2), 101-131
Robin, M. J. L., Sudicky, E. A., GiUham, R. W., and Kachanski, R. G., (1999). Spatial variability of strontium distribution coefficients and their correlation with hydraulic conductivity in the Canadian forces base borden aquifer, Water Resources Research, 27(10), 2619-2632