This paper presents a multiobjective integer goal programming model for analyzing problems involving planning and design of regional hazardous waste management systems. The proposed model can be used to select treatment and disposal facilities along with allocation of hazardous wastes and waste residues from generators to treatment and disposal facilities and transportation routes in order to achieve the minimum cost and/or minimum risk based solution. The model formulation addresses important practical issues such as diverse characteristics of the hazardous wastes, waste-waste noncompatibility and waste-treatment technology compatibility. The objectives considered are the minimization of (1) total risk (including risk associated with transportation, treatment, and disposal); and (2) total cost (sum of the treatment, disposal, and transportation costs). An illustrated example is presented to demonstrate the usefulness of the proposed formulation.

An infectious waste collection system planning in Taiwan requires solving routing and scheduling problems simultaneously due to an environmental regulation. Previous work on this type of problem that considered a single criterion is extended in this study to a multiple criteria optimization approach. In addition to minimization of the cost, minimization of transportation risk and balancing of workers’ carrying loads are considered for hazardous waste collection. Dynamic programming and integer linear programming methods are incorporated as an integral part of decision analysis in which a compromise programming method is adopted to integrate the three criteria of concern. Finally, an illustrative example of infectious waste collection in Tainan City is presented to demonstrate the application potential of the proposed approach. A geographical information system is used to aid in relevant spatial analyses, such as the location of medical institutions, actual travel distances, and district population density, in the optimization steps. The findings indicate that the incorporation of more objectives in decision analysis may provide a scientifically credible solution for medical waste management in urban areas.

This study presents an integrated approach with multiphase modeling, factorial design, and Monte Carlo simulation by which to deal with uncertainties in the fate and transport process of petroleum contaminants in a subsurface environment. A set of $2^4$ factorial experiments is conducted for 4 factors, including the permeability and porosity of sandy loam and clay soils, in order to examine the effects of uncertain input characteristics on the modeling outputs. For each experimental set, a number of Monte Carlo simulation runs are undertaken through a well-developed multiphase compositional simulator. Thus, the problem of multiphase flow with uncertain information can be solved. The proposed methodology is applied to a petroleum-contaminated site in western Canada to illustrate its applicability. The simulation results reveal that the uncertainties in input parameters impose considerable influence on the predicted output: variations in the permeability of sandy loam soil especially will have significant impact on the modeling output. The proposed method offers an effective tool for evaluating uncertainty in multiphase flow simulation.

Surfactant-enhanced remediation is an effective approach for dealing with subsurface-contaminated sites. However, in studying and controlling the related processes, difficulties exist in incorporating a complicated numerical simulation model that is needed for process forecasting within a real-time nonlinear optimization framework that is critical for supporting process control. In this study, an integrated simulation-optimization approach was developed for supporting real-time dynamic modeling and process control of surfactant-enhanced remediation at petroleum-contaminated sites. Subsurface modeling is combined with a dual-response surface method to develop a system for generating optimum operation conditions under various site conditions, through the support of a nonlinear optimization model. The developed methodology was applied to a real-world case study in western Canada. Using the developed three-dimensional multiphase and multicomponent model, the surfactant-enhanced remediation process that is being implemented on the study site was simulated. The results provide useful information for further dual-response surface analysis to support the development of an optimization model to determine optimum process operation conditions. Under each initial contaminant concentration, optimum operation conditions can then be identified through this combined dual-response surface method optimization approach. Thus, a decision support system can then be produced to guide decisions of remediation process control under various site conditions.

Dual-phase vacuum extraction (DPVE) is an emerging cost-effective technology used to enhance remediation by recovering petroleum hydrocarbon from the subsurface. To improve its efficiency, insight into the complex processes and phase/component interactions in the remediation system needs to be gained before the system can be properly designed and operated. In this paper, a numerical simulation model is proposed to analyze the behavior of the complex DPVE remediation system. The simulator developed is then applied to a petroleum-contaminated site in western Canada that is undergoing the DPVE process. The influence of vacuum pressure employed at extraction wells on system performance is investigated. Moreover, the effects of parameter uncertainties in modeling results are also examined. The proposed approach offers an effective tool for evaluating remediation system performance and supporting decisions on site remediation and management practices.

This study carries out a system-based assessment in search of the optimal operating strategy for light hydrocarbon recovery in a contaminated aquifer using bioslurping technology. A two-stage analytical framework involving a combined simulation/regression/optimization (S/R/O) modeling approach was designed to analyze, predict, and optimize the pumping volume with regard to the separate oil, water, and soil gas phases in the bioslurping process. The aim of the first-stage analysis is to calibrate and verify this newly developed bioslurping simulation model and to assess light nonaqueous phase liquids (LNAPLs) recovery efficiency in a heterogeneous, anisotropic unconfined aquifer. Recovery system simulations using derived linear regression submodels present high potential for exhibiting, eliciting, and summarizing the nonlinear behavior between successive batch operations in predicting residual oil levels in the surface groundwater table for model calibration and verification. The research findings indicate that the statistical paired-sample T-tests confirm the prediction accuracy of the bioslurping model during calibration and verification procedures. The simulation results presented in this paper are proved useful for finding a relatively better operational scheme to minimize the indirect operational risk throughout the on-site remediation time period. The simulation results also provide a firm database to assist in an advanced optimization study in the second stage analysis for generating risk-informed, scientifically credible, and cost-effective solutions in remediation.

The bioslurping model was successfully applied to simulate the recovery efficiency of light nonaqueous phase liquids (LNAPLs) in a companion remediation study. This paper carries out an advanced environmental systems analysis for handling the application challenges related to a seamless integration of simulation outputs, regression analyses, and optimization practices to search for the best remediation solution. In view of the inherent complexity of integrating simulation outputs at various scales for building the optimization steps and searching for the ultimate risk-informed and cost-effective solutions, a set of regression submodels describing the system response to water and gas pumping and oil skimming in hydrocarbon recovery are derived and employed to bridge the traditional gap between simulations and optimization studies. The bioslurping model is used to generate regression submodels of the system behavior during remediation operations. To test the robustness of the operational scheme, three planning scenarios based on a common linear programming framework are defined and utilized to assess the optimal control mechanics for LNAPL recovery. These scenarios are designed to shed light on improving the predictive capabilities for environmental decision making to handle a complex multiphase and multimedia remediation program. The research findings indicate that water-pumping volume can be minimized and oil and soil gas recovery can be maximized whether the candidate well locations are fixed beforehand or not. However, flexible choices in well locations in the optimization steps merit substantial examination in the future.

The management and regulation of hazardous waste sites are increasingly being carried out by assessing risks via different exposure pathways. The selection of an appropriate multimedia fate and transport for carrying out screening level risk assessment is a challenging task given the paucity of data and the need for modeling rigor. The utility of a multicriteria decision-making technique, called the analytic hierarchy process (AHP), is explored in this study to select an appropriate model that satisfies the needs of the stakeholders in an optimal manner. The criteria for model selection focused on various issues ranging from scientific credibility to data availability and end-user requirements, and were compiled from a survey of modeling practices and needs of various agencies of the California/Environmental Protection Agency that was carried out previously. Seven different public-domain multimedia modeling codes were evaluated against these criteria using AHP. The study indicated that the highest ranked model (CalTOX) only satisfied about 31% of the needs of the stakeholders for the site conditions assumed. This result indicates that there is a wide scope for further improvement in all the models that are considered here. The local priority matrix obtained for each criteria was seen to be useful to identify limitations of individual models. Certain improvements are suggested to enhance the available public-domain multimedia models to meet stakeholder needs. AHP is seen as a consistent, flexible, and intuitive approach to assess necessary model complexity and is recommended for use in similar policy and regulatory applications.