Research Activities

Overview

Research Topics

• Life-cycle-based Solid Waste Management
  Development, implementation and application of an integrated solid waste management decision support tool for generating and evaluating alternative strategies to efficiently meet economic and environmental goals. (Collaborators: Barlaz, Brill)
  [Related Publications]

• Water Resources Planning & Management
  Enabling systems analytic procedures for solving water resources management problems through the development and implementation of prototype decision support tools.
  [Related Publications]

• Transportation and Environmental Systems Analysis
  Study of the implications of transportation infrastructure policy, design and operation on the environment through the use of systems analytic and computational frameworks. (Collaborators: Brill)
  [Related Publications]

• System Identification/Inverse Modeling
  Investigation of systems methods and computational implementations for solving contaminant source characterization problems in groundwater and water distribution systems. (Collaborators: Mahinthakumar)
  [Related Publications]

• Multi Objective Evolutionary Algorithms
  Development and testing of evolutionary algorithms for identifying the Pareto optimal solutions for multiobjective problems.
  [Related Publications]

• Algorithms for Design Innovation
  Investigation of search algorithms to identify alternatives that consist of unusual combinations of design choices but with similar design performance, potentially creating innovative solutions to engineering design problems.
  [Related Publications]

• Model Error Correction Procedures
  Development and testing of quantitative methods to improve the predictability of simulation models by accounting for automatic parameter calibration and model error correction.
  [Related Publications]

List of Projects and Abstracts

DDDAS-TMRP (COLLABORATIVE RESEARCH): An Adaptive Cyberinfrastructure for Threat Management in Urban Water Distribution Systems
Mahinthakumar, Brill, and Ranjithan (NCSU), von Laszewski (Univ Chicago/Argonne Natl Lab), Harrison (Univ of SC), Uber (Univ of Cincinnati)
NSF
01/2006-12/2008
Abstract:
Contamination threat management in drinking water distribution systems involves real-time characterization of the contaminant source and plume, identification of control strategies, and design of incremental data sampling schedules. This requires dynamic integration of time-varying measurements of flow, pressure and contaminant concentration with analytical modules including models to simulate the state of the system, statistical methods for adaptive sampling, and optimization methods to search for efficient control strategies. For realistic distribution systems, the analytical modules are highly compute-intensive, requiring multi-level parallel processing via computer clusters. While data often drive the analytical modules, data needs for improving the accuracy and certainty of the solutions generated by these modules dynamically change when a contamination event unfolds. Since such time-sensitive threat events require real-time responses, the computational needs must also be adaptively matched with available resources. Thus, a software system is needed to facilitate this integration via a high-performance computing architecture (e.g., the TeraGrid) such that the measurement system, the analytical modules and the computing resources can mutually adapt and steer each other. The goal of this multi-disciplinary research is to develop a cyberinfrastructure system that will both adapt to and control changing needs in data, models, computer resources and management choices facilitated by a dynamic workflow design. Using virtual simulation and a field study, this cyberinfrastructure will be tested on illustrative scenarios for adaptive management of contamination events in water distribution systems. Urban water distribution systems are vulnerable to accidental and intentional contamination incidents that could result in adverse human health and safety impacts. The pipe network in a typical municipal distribution system includes redundant flow paths to ensure service when parts of the network are unavailable, and is designed with significant storage to deliver water during daily peak demand periods. Thus, a typical network is highly interconnected and experiences significant and frequent fluctuations in flows and transport paths. These design features unintentionally enable contamination at a single point in the system to spread rapidly via different pathways through the network, unbeknown to consumers and operators. When a contamination event is detected via the first line of defense, e.g., data from a water quality surveillance sensor network and reports from consumers, the municipal authorities are faced with several critical questions as the contamination event unfolds: Where is the source of contamination? When and for how long did this contamination occur? Where additional hydraulic or water quality measurements should be taken to pinpoint the source more accurately? What is the current and near future extent of contamination? What response action, such as shutting down portions of the network, implementing hydraulic control strategies, or introducing decontaminants, should be taken to minimize the impact of the contamination event? What would be the impact on consumers by these actions? Real-time answers to such complex questions will present significant computational challenges. This project will address these challenges by developing an adaptive cyberinfrastucture that will enable real-time processing and control through dynamic integration of computational components and real-time sensor data. This system will be evaluated using contamination scenarios based on field-scale data from a large metropolitan area.

ITR: A Prototype to Enable Near Real-time Environmental Characterization on the Grid
Mahinthakumar and Ranjithan (NCSU), Karonis (NIU)
NSF
09/2003-08/2006
Abstract:
The overall goal of this project is to investigate formal computational approaches that can readily harness grid computing for the solution of environmental characterization problems. To this end, we will develop a grid-enabled software framework. Two alternative paradigms, based on the grid-enabled version of MPI and Java, respectively, will be explored. The framework will be applied to groundwater and air pollution problems, both of which are of prime societal importance. Acknowledging that, even within a grid environment, inverse problems are computationally challenging, model surrogates will be explored. Further, modeling to generate alternatives (MGA), a technique that identifies a set of good yet very different solutions, will be investigated for dynamically steering the search process through human-computer interaction. Alternatives generated via MGA will be used also to address the commonly encountered non-uniqueness issue in inverse problems.

Solid Waste Management Planning for the State of Delaware
Barlaz and Ranjithan
Delaware Solid Waste Authority
10/2003-12/2005
Abstract:
The objective of this project is to assist Delaware Solid Waste Authority (DSWA) identify long-term solid waste management (SWM) alternatives. Using a Life-cycle inventory (LCI) based SWM model that was developed at NC State University, a series of alternate SWM strategies will be analyzed with respect to cost, waste diversion from landfills, energy consumption and environmental emissions. Tradeoff information will be generated to examine the incremental cost of increased recycling, decreased dependence on landfills and reduced emissions. A variety of options, including curbside recycling, composting, combustion, and in-state disposal or out-of-state landfill disposal, for managing waste will be considered. The results from this research are expected to provide valuable input to the DSWA staff as they develop and evaluate future SWM plans for the State of Delaware.

Urban Watershed Management Tools
Ranjithan
US EPA/MCNC
10/2001-9/2004
Abstract:
U.S. EPA's Multimedia Integrated Modeling System (MIMS) provides a unified computing framework to simulate the cycling of environmental pollutants within and across all media. The processes associated with urban storm water runoff and wastewater discharge are to be available for simulation in MIMS. The simulation models are to be coupled with systems analytic methods (including uncertainty analysis and optimization-based search procedures) to explore and identify efficient strategies to manage the urban runoff related problems. In addition, decision-makers require cost/benefit tradeoffs and reliability associated with different strategies. The objective of this project is to integrate urban watershed decisions support tools into MIMS, and demonstrate their use. Related goals include integration of an appropriate urban storm water model into MIMS, and testing the broader applicability of the decision support tools for use with other process simulation models, including US EPA's Storm Water Management Model (version 5).

NSF CAREER Award: Development of a Computer-Based Methodology to Assist in Environmental Systems Analysis and Decision Making and Its Applications in Watershed Management
Ranjithan
National Science Foundation
5/1998-4/2003
Abstract:
This project will: 1) investigate ways to enhance the capabilities of genetic algorithms (GAs) for complex environmental systems analysis, 2) develop and integrate into existing courses an interactive training module to assist in teaching the fundamentals of GAs and their uses in environmental systems analysis, 3) explore applications of the methodology in watershed management and 4) integrate the applications and their findings in courses related to environmental systems analysis. The academic plans include the development of teaching modules and interactive techniques for both students and practitioners. To demonstrate the practical applicability, an array of realistic watershed management applications will be investigated, including a case study of the Neuse River Basin in North Carolina.
Key Words watershed management, BASINS, modeling, evolutionary computation, genetic algorithms, uncertainty, decision support, TMDL, planning and management

Enhancements to the Strategy Development Tool to Support Regulatory Analysis
Loughlin, Baugh, Brill, and Ranjithan
North Carolina Department of Environment and Natural Resources - Division of Air Quality
9/1998-12/2001
Abstract:
The Strategy Development Tool (SDT) is a prototype decision support system for air quality management. Components of the SDT include tools for: 1) visualizing emissions inventories, 2) designing and testing control strategies, 3) modeling the costs and impacts of incentive-based control approaches such as emissions trading programs, and 4)identifying low-cost management alternatives through optimization. The goal of this project is to upgrade the features of the SDT so that it can be used by the State of North Carolina in developing state implementation plans towards meeting the Federal air quality standards.
Key Words air quality management, modeling, evolutionary computation, genetic algorithms, optimization/search procedures, uncertainty, decision support, planning and management, control strategy development

Development of an integrated systems model to explore environmentally beneficial alternatives for product manufacturing and waste management
Ranjithan and Barlaz
National Science Foundation
10/1998-10/2001
Abstract:
As societies become more focused on minimizing the impact of human activity on the environment, many industries are trying to move towards environmentally responsible manufacturing practices. The objective of this work is to develop a model, integrating a product life-cycle inventory (LCI) with a newly developed life-cycle model of waste management alternatives to examine how the combined LCIs are important and can be influential in product decision making. An integrated systems model will be developed to represent life cycle considerations of products from manufacturing through waste management. The model will then be illustrated through a series of case studies.
Key Words solid waste management, modeling, life cycle analysis, mathematical programming, optimization, uncertainty, planning and management

Advancement of Environmental Decision Support Systems through HPCC
Brill, Baugh, Ranjithan
US EPA
9/1996-9/1999
Abstract:
The goal of the proposed research is to overcome the current computational resource limitations by developing Decision Support System (DSS) tools for use within a high performance distributed computing and communications (HPCC) environment. There are four research objectives: 1) to explore the role of optimization techniques in a DSS framework for complex environmental problems; 2) to examine ways of making better use of existing and expected future computational power to increase performance, 3) to continue the development of better DSS prototypes as true joint-congnitive systems, and 4) to evaluate each prototype's performance and user interface with respect to user needs.
Key Words air quality management, modeling, evolutionary computation, genetic algorithms, optimization/search procedures, uncertainty, decision support, planning and management, control strategy development, high performance computing

Models for Evacuation Planning and Network Analysis System
Rouphail and Ranjithan
Lockheed Martin Energy Systems, Inc./Oak Ridge National Laboratory
6/1996-5/1999
Abstract:
This research is directed at developing network-specific evacuation plans for dynamic incidents. The focus is on the development of algorithm that can produce minimum evacuation times over a highway network subject to constraints on shelter locations and capacity, and link capacity. Models will be developed which consider the priority status of evacuation subzones, and selection of link reversals to expedite the evacuation process.
Key Words evacuation management, transportation planning and management, mathematical programming, network optimization

Assessing Pavement Layer Condition Using Deflection Data
Kim and Ranjithan
National Cooperative Highway Research Program, TRB
2/1997-4/2000
Abstract:
The objective of the research is to develop procedures to assess the condition of pavement layers based on deflection measurements. This research is concerned with all layers of rigid and flexible pavements that include an asphalt concrete surface layer. The proposed research approach describes a mechanistic-empirical method of developing a simple, practical deflection interpretation procedure for condition assessment of distressed pavement layers. Validation and implementation of this procedure will be conducted using data collected from distressed pavement sections in various locations in the United States.
Key Words nondestructive evaluation, artificial neural networks, pavement management

Nondestructive Evaluation of Structural Condition of Timber Piles
Kim and Ranjithan
North Carolina Department of Transportation
7/1997-12/1999
Abstract:
The principal objective of the proposed research is to develop a nondestructive test and analysis method, that can be readily implemented, for determining structural condition of installed timber piles using the stress wave test method currently used by NCDOT for the evaluation of in-place length of timber piles. The outcome of this research will be a comprehensive stress wave testing system that can quantitatively evaluate the structural condition of installed timber piles. Test results obtained from this system will enable NCDOT calculating the load bearing capacity of deteriorated timber piles. The resulting system will be a user-friendly analysis program that can be used for routine inspection of timber piles.
Key Words nondestructive evaluation, artificial neural networks, transportation infrastructure management

An Environmental Assessment of Management Alternatives for Municipal Solid Waste
Barlaz, Brill, and Ranjithan
Research Triangle Institute
8/1995-4/2000
Abstract:
Research is in progress to evaluate the cost and life-cycle inventory (LCI) of municipal solid waste management alternatives. The overall objective of the research is to develop a user-friendly decision support system prototype to identify efficient alternatives for solid waste management (SWM) in consideration of the cost and LCI parameters. The decision support tool will utilize a combination of a site-specific user input data and default data to identify SWM strategies that are optimal with respect to a given objective such as cost, energy consumption, or the emissions of a number of air pollutants including NOx, Sox, CO and particulates. Components of the decision support system include process models for each solid waste unit operation included in the management system, a mathematical model of the SWM system, an optimization routine and a graphical user interface. The SWM system includes waste collection and transfer, recycling, composting, waste-to-energy, refuse derived fuel, anaerobic digestion and landfills. In addition, an offset analysis is used to include manufacturing processes associated with the conversion of recycled materials to new products.
Key Words solid waste management, modeling, life cycle analysis, mathematical programming, optimization, planning and management

Interpretation of FWD Data When Pavement Layers Are Not Intact
Kim and Ranjithan
North Carolina Department of Transportation
7/1995-7/1997
Abstract:
The falling weight deflectometer (FWD) is a principal means of evaluating the structural condition of pavements. Effort has been made to interpret FWD deflection basins for determining rehabilitation strategies and overlay thickness. This research interprets FWD information for condition evaluation of flexible pavements. The objectives are: 1) to determine the effects of cracked or broken layers in flexible pavement systems based on FWD deflection basins; 2) to develop a method for determining the layers in the existing pavement structure at which rehabilitation must be directed; and 3) to verify the recommended procedure using the deflection measured from cracked pavements with known conditions.
Key Words nondestructive evaluation, artificial neural networks, pavement management

A Decision Support System for Highway Construction Considering Environmental Impacts
Rouphail and Ranjithan
Center for Transportation and the Environment
8/1995-12/1997
Abstract:
This project represents a collaborative research with the Technical University of Lisbon, with support from the Portuguese-American Foundation (FLAD) and the Center for Transportation and Environment (CTE). The project is designed to develop a decision support system (DSS) for scheduling highway construction projects in a metro area, subject to annual budget, resource and precedence constraints. The Lisbon Metropolitan Area will be the case study for the development and application of the DSS.
Key Words transportation planning and management, mathematical programming, optimization

Decision Support Systems for Air Pollution Management
Brill, Baugh, and Ranjithan
US EPA via MCNC-North Carolina Supercomputing Center and Industrial Extension Service
9/1993-9/1996
Abstract:
The capabilities of high-performance computing systems are often used to build analysis models with increasing detail and realism. However, these computing resources also provide a key technological component for interactive decision support systems. This project explores the design and development issues necessary to create computer systems that support open ended, iterative decision making processes. A wide range of support system tools will be considered for use in such a system, including mathematical optimization, genetic algorithms, neural networks, and formal methods to examine tradeoffs among multiple objectives and to generate alternative solutions.
Key Words air quality management, control strategy development, mathematical programming, evolutionary computation, optimization, high performance computing, decision support

Design of Animal Waste Management Strategies to Achieve Regional Environmental Objectives
Ranjithan, Liehr, and Classen
Abstract:
The replacement of anaerobic lagoons that are currently used for swine waste management in North Carolina is expected to be expensive. Recent initiatives calling for such actions have increased the need for characterization of the cost and treatment effectiveness of each alternative, which potentially depends on the location and size of the farm. Thus, identification of the most effective solution to achieve collective target reductions in discharge levels needs an integrated approach that examines and searches among these technology alternatives at all farms simultaneously. The objective of this research is to develop collective or regional management strategies that will aid policy makers, planners, and farmers in making cost effective lagoon replacement decisions to achieve desired treatment and public protection goals. A major component of this activity is the implementation of a cost and treatment efficiency assessment tool (CAWSST) to evaluate alternative animal waste treatment technologies. Building upon the outputs of CAWSST and existing databases, formal optimization techniques to search for good solutions with respect to different goals are being developed. This procedure is presented. In an illustrative example that is used to describe the application of this approach, GIS based data for the farms within the Lower Neuse River Watershed are used. Farm sizes are inputted to CAWSST, and the outputs, together with additional GIS data such as farm locations, are then used in a mixed-integer programming model that helps us examine different management scenarios for the region. We optimized for cost while meeting constraints such as pollutant discharge targets, farm proximity to floodplains, risk levels, or the effect of odor on the area. Example scenarios include the least cost solution, uniform treatment for the whole watershed area, and uniform treatment within certain zones defined by such factors as political boundaries, farm size, type of farming operation or proximity to floodplains.
Key Words animal waste management, treatment technology modeling, mathematical programming, optimization, planning and management

Improving Performance of Water Treatment Plant Operations through Optimization
Knappe and Ranjithan
Abstract:
The primary focus of this research is to develop and test computer-based decision support tools to assist in improving the performance of drinking water treatment plants in the presence of variability and uncertainty. Two aspects of the plant operations are targeted. One, the unit process choices at the plant design stage, and two, the process control settings during daily operations. This investigation attempts to identify, based on pilot studies, these choices such that the overall reliability in meeting effluent conditions is increased. The first phase of this research implemented a procedure based on artificial neural network (ANN) models of unit process trains. Data from a pilot plant was used to test and validate this modeling approach. This ANN-based model was then coupled with a systematic search procedure to identify effective process control choices for controlling turbidity. This procedure is then applied to identify effective process controls to achieve the desired effluent targets assuming deterministic conditions. The performance of these controls, however, is unpredictable in the presence of variable or uncertain influent conditions, measurement errors, and uncertainty in process performance. Applying excessive treatment to build in a safety factor is an option to avoid potential failure to meet effluent targets, but at a significant increase in cost. A systematic consideration of uncertainty is needed to address this problem. Ongoing work investigates ways to characterize the uncertainty in the plant performance, and to identify cost-effective operational controls that improve the robustness in the performance of water treatment plant operations.
Key Words water treatment, artificial neural networks, mathematical programming, optimization, uncertainty

Decision-making under uncertainty: A New Quantitative Method-Bayesian Programming
Harrison and Ranjithan
Abstract:
Our understanding of environmental processes is fraught with uncertainty. Nevertheless, environmental decision-makers must decide a course of action to follow. Often, the question of what to do is falsely framed as having to decide between taking immediate action or to gather more data before addressing the problem. In fact, of course, both actions can be pursued simultaneously. Surprisingly, however, there are few numerical approaches available to explore such options. Of the methods for decision-making under uncertainty that are available, nearly all are either not theoretically sound or are applicable only to the simplest of problems. This poster introduces a new method for decision-making under uncertainty that addresses many of the limitations of existing approaches. In addition to uncertainty, the methodology can be used to address a range of related issues typically encountered in environmental modeling, including the stochastic behavior of environmental systems, and irreversibility. The method, Bayesian Programming (BP), utilizes optimization techniques for an efficient search of the alternatives and incorporates Bayesian methods to model uncertainty reduction as data is collected. The methodology is illustrated with a simple example. Finally, the strengths and limitations of the method are discussed. The technique is expected to increase the decision-maker's ability to address uncertainty.
Key Words uncertainty, decision analysis, optimization, Bayesian methods, heuristic search procedures