Search

We consider the problem of identifying solutions to a stochastic system under multiple constraints. The objective function and the constraints are expressed in terms of performance measures of the system that are observable only via a simulation model parameterized by a finite number of decision variables. In solving for such a system, one faces the much harder challenge of verifying the feasibility of a potential solution. Toward this, we present cgR-SPLINE, a multistart simulation optimization (SO) algorithm on integer spaces. cgR-SPLINE sequentially solves random restarts of a gradient-based local search routine with increasing precision. The local search routine in turn solves progressively stricter outer approximations of the underlying problem. The local solution estimator from a recently ended restart is probabilistically compared against an incumbent solution, thus generating a sequence of global solution estimators. The optimal convergence rate of the solution iterates is observed to be sub-exponential, slower than the exponential rate observed for SO problems on unconstrained discrete spaces. Additionally, efficiency for cgR-SPLINE dictates that the number of multistarts and the total simulation budget be sublinearly related, implying an increased emphasis on exploration than is prescribed in the continuous context. Heuristics for choosing constraint relaxations and solution reporting demonstrate good finite-time performance on three SO problems, of which two are nontrivial. The extension of cgR-SPLINE's framework to mixed spaces seems a natural next step. The presence of infeasible points arbitrarily close to the stochastic boundary, however pose challenges for consistency. We present a general framework for mixed spaces that is very much along the lines of cgR-SPLINE and propose ideas for specific algorithmic refinements and solution reporting. Strategically locating the restarts of a multistart SO algorithm appears to be a largely unexplored research topic. Toward achieving efficiency during the exploration phase, we present ideas for ``antithetically" generating the restarts from probability measures constructed from the SO algorithm's performance trajectory. Asymptotic behavior of the proposed sampling strategy and policies for optimal parameter selection are presently conjectural, but appear promising based on the outcomes of preliminary experiments. Ph. D.

The research work presented in this dissertation relates to lot streaming in two-stage flow shops and assembly shops. Lot streaming refers to the process of splitting a production lot into sublots, and then, processing the sublots on different machines simultaneously in an overlapping manner. Such a strategy allows finished material at each stage to be transferred downstream sooner than if production and transfer batches were restricted to be the same size. In the case when each sublot consists of just one item, a single-piece-flow is obtained. Such a continuous flow is a key element of the Toyota Production System. However, single-piece-flow increases the number of transfers and the total transportation cost (time). As a result, it may not be economically justifiable in many cases, and therefore, material may have to be transferred in batches (called transfer batches, or sublots). Lot streaming addresses the problems of determining optimal sublot sizes for use in various machine environments and optimizes different performance measures.Given this relationship between lot streaming and the Toyota Production System, lot streaming can be considered a generalization of lean principles. In this dissertation, we first provide a comprehensive review of the existing literature related to lot streaming. We show that two-stage flow shop problems have been studied more frequently than other machine environments. In particular, single-lot two-machine flow shops have been very well researched and efficient solution techniques have been discovered for a large variety of problems. While two-stage flow shop lot streaming problems have been studied extensively, we find that the existing literature assumes that production rates at each stage remain constant. Such an assumption is not valid when processing rates change, for example, due to learning. Learning here, refers to the improvements in processing rates achieved due to experience gained from processing units. We consider the case when the phenomenon of learning affects processing and setup times in a two-stage flow shop processing a single lot, and when, sublot-attached setup times exist. The decrease in unit-processing time, or sublot-attached setup time, is given by Wright's learning curve. We find closed-form expressions or simple search techniques to obtain optimal sublot sizes that minimize the makespan when the effect of learning reduces processing times, sublot-attached setup times, or, both. Then, we provide a general method to transform a large family of scheduling problems related to lot streaming in the presence of learning, to their equivalent counterparts that are not influenced by learning. This transformation is valid for all integrable learning functions (including the Wright's learning curve). As a result, a large variety of new problems involving learning can be solved using existing solution techniques. We then consider lot streaming in stochastic environments in the context of sourcing material. Such problems have been well studied in the literature related to lot streaming for cost-based objective functions when demand is continuous, and when processing times are deterministic, or, for material sourcing problems when the time required to procure a lot is stochastic but is independent of the lot size. We extend this study to the case when the time required to produce a given quantity of products is stochastic and dependent on the number of units produced. We consider the case when two sublots are used, and also compare the performance of lot streaming to the case when each sublot is sourced from an independent supplier. Next, we address a new problem related to lot streaming in a two-stage assembly shop, where we minimize a weighted sum of material handling costs and makespan. We consider the case when several suppliers provide material to a single manufacturer, who then assembles units from different suppliers into a single item. We assume deterministic, but not necessarily constant, lead times for each supplier, who may use lot streaming to provide material to the manufacturer. Lead times are defined as the length of the time interval between a supplier beginning to process material and the time when the first sublot is delivered to the manufacturer; Subsequent sublots must be transported early enough so that the manufacturer is not starved of material. The supplier may reduce this lead time by using lot streaming, but at an increased material handling cost. The decrease in lead time is also affected by other factors such as lot attached/detached setup times, transportation times etc. We allow these factors to be different for each supplier, and each lot processed by the same supplier. We refer to this problem as the Assembly Lot Streaming Problem (ALSP). We show that the ALSP can be solved using two steps. The first step consists of solution to several two-stage, single-lot, flow shop, makespan minimization problems. The solution to these problems generate prospective sublot sizes. Solution methods outlined in the existing literature can be used to complete this step. The second step obtains optimal number of sublots and production sequence. For a given production sequence, this step can be executed in polynomial-time; otherwise, the second step problem is NP-hard and integer programming formulations and decomposition-based methodologies are investigated for their solution. We make very limited assumptions regarding the handling cost and the relationship between the supplier lead time and number of sublots used. As a result, our solution methodology has a wide scope. Ph. D.

This dissertation addresses the development of optimization models for hospital evacuation logistics, as well as the analyses of various resource management strategies in terms of the equity of evacuation plans generated. We first formulate the evacuation transportation problem of a hospital as an integer programming model that minimizes the total evacuation risk consisting of the threat risk necessitating evacuation and the transportation risk experienced en route. Patients, categorized based on medical conditions and care requirements, are allocated to a limited fleet of vehicles with various medical capabilities and capacities to be transported to receiving beds, categorized much like patients, at the alternative facilities. We demonstrate structural properties of the underlying transportation network that enables the model to be used for both strategic planning and operational decision making. Next, we examine the resource management and equity issues that arise when multiple hospitals in a region are evacuated. The efficiency and equity of the allocation of resources, including a fleet of vehicles, receiving beds, and each hospital\'s loading capacity, determine the performance of the optimal evacuation plan. We develop an equity modeling framework, where we consider equity among evacuating hospitals and among patients. The range of equity of optimal solutions is investigated and properties of optimal and equitable solutions based on risk-based utility functions are analyzed. Finally, we study the integration of the transportation problem with the preceding hospital building evacuation. Since, in practice, the transportation plan depends on the pace of building evacuation, we develop a model that would generate the transportation plan subject to the output of hospital building evacuation. The optimal evacuation plans are analyzed with respect to resource utilization and patient prioritization schemes. Parametric analysis of the resource constraints is provided along with managerial insights into the assessment of evacuation requirements and resource allocation. In order to demonstrate the performance of the proposed models, computational results are provided using case studies with real data obtained from the second largest hospital in Virginia. Ph. D.

Managing traffic during oversaturated conditions is a current challenge for practitioners due to the lack of adequate tools that can handle such situations. Unlike under-saturated conditions, operation of traffic signal systems during congestion requires careful consideration and analysis of the underlying causes of the congestion before developing mitigation strategies. The objectives of this research are to provide a practical guidance for practitioners to identify oversaturated scenarios and to develop a multi-objective methodology for selecting and evaluating mitigation strategy/ or combinations of strategies based on a guiding principles. The research focused on traffic control strategies that can be implemented by traffic signal systems. The research did not considered strategies that deals with demand reduction or seek to influence departure time choice, or route choice. The proposed timing methodology starts by detecting network's critical routes as a necessary step to identify the traffic patterns and potential problematic scenarios. A wide array of control strategies are defined and categorized to address oversaturation problematic scenarios. A timing procedure was then developed using the principles of oversaturation timing in cycle selection, split allocation, offset design, demand overflow, and queue allocation in non-critical links. Three regimes of operation were defined and considered in oversaturation timing: (1) loading, (2) processing, and (3) recovery. The research also provides a closed-form formula for switching control plans during the oversaturation regimes. The selection of optimal control plan is formulated as linear integer programming problem. Microscopic simulation results of two arterial test cases revealed that traffic control strategies developed using the proposed framework led to tangible performance improvements when compared to signal control strategies designed for operations in under-saturated conditions. The generated control plans successfully manage to allocate queues in network links. Ph. D.

A deterministic change to a time-varying queueing model is described as either changing the number of entities, the queue capacity, or the number of servers in the system at selected times. We use a surrogate distribution for N(t), the number of entities in the system at time t, to approximate deterministic changes to the Ph(t)/Ph(t)/1/c and the Ph(t)/M(t)/s/c queueing models. We develop a solution technique to minimize the number of state probabilities to be approximated. Master of Science

This research is focused on driver behavior in traffic, especially during car-following situations and safety critical events. Driving behavior is considered as a human decision process in this research which provides opportunities for an artificial driver agent simulator to learn according to naturalistic driving data. This thesis presents two mechine learning methodologies that can be applied to simulate driver naturalistic driving behavior including risk-taking behavior during an incident and lateral evasive behavior which have not yet been captured in existing literature. Two special machine learning approaches Backpropagation (BP) neural network and Neuro-Fuzzy Actor Critic Reinforcement Learning (NFACRL) are proposed to model driver behavior during car-following situation and safety critical events separately. In addition to that, as part of the research, state-of-the-art car-following models are also analyzed and compared to BP neural network approach. Also, driver heterogeneity analyzed by NFACRL method is discussed. Finally, it presents the findings and limitations drawn from each of the specific issues, along with recommendations for further research. Master of Science

In this dissertation, we develop novel approaches to independently address two issues that are commonly encountered in machine scheduling problems: uncertainty of problem parameters (in particular, due to job processing times), and batching of jobs for processing on capacitated machines. Our approach to address the uncertainty issue regards the indeterminate parameters as random variables, and explicitly considers the resulting variability of a performance measure. To incorporate variability into the schedule selection process, we develop a method to evaluate both the expectation and variance of various performance measures for a given schedule. Our method is based on the use of mixture models to approximate a variety of distribution types. The Expectation-Maximization algorithm of Dempster et al. (1977) is applied to derive mixture models of processing time distributions. Our method, then, utilizes these mixture models to calculate the distributions of other random variables in order to derive the expectation and variance of various scheduling performance measures, assuming that the job sequencing decisions are known a priori. To make our method more computationally efficient, we adapt a mixture reduction method to control the number of mixture components used in the intermediate steps. We apply our method to two different scheduling problems: the job shop makespan scheduling problem and the single machine total weighted tardiness scheduling problem, and compare its performance with that of Monte-Carlo method. The results show the efficacy of our mixture approximation method. It generates fairly accurate results while requiring significantly less CPU times. The proposed method offers a good compromise between the Monte Carlo method, which requires extensive effort, and use of simple normal approximation, which produces lower-quality results. Next, we introduce and demonstrate for the first time in the literature the use of conditional-value-at-risk (CVaR) as a criterion for stochastic scheduling problems in order to obtain risk-averse solutions. This criterion has the tendency of minimizing both the expectation and variance of a performance measure simultaneously, which is an attractive feature in the scheduling area as most of the literature in this area considers the expectation and variance of a performance measure separately. Also, the CVaR has an added advantage of maintaining a linear objective function. We develop a scenario-based mixed integer programming formulation to minimize CVaR for the general scheduling problem involving various performance measures, and employ a decomposition-based approach for its solution. Furthermore, a set of valid inequalities are incorporated to strengthen the relaxed master problem of this decomposition scheme. The proposed approach is demonstrated on the single machine total weighted tardiness scheduling problem. Our computational investigation reveals the efficacy of the proposed decomposition approach and the effectiveness of using the CVaR as an optimization criterion for scheduling problems. Besides providing an exact approach to solve our stochastic scheduling problem, we also develop an efficient heuristic method to enable the use of CVaR for large-sized problems. To that end, we modify the Dynasearch method of Grosso et al. (2004) to minimize CVaR for a stochastic scheduling problem. Furthermore, we extend the application of CVaR to a parallel-machine total weighted tardiness problem. The use of CVaR appears to be quite promising for simultaneously controlling both the expected value and variability of a performance measure in a stochastic scheduling environment. Scenario-based formulations have frequently been used for stochastic scheduling problems. However, the determination of a lower bound can be a time-consuming task for this approach. Next, we develop a new method for scenario generation that is computationally competitive and that assures attainment of an exact lower bound. Our approach is based on discretization of random parameter distributions of job processing times. We use the idea of Recursive Stratified Sampling to partition the probability space, so that the conditional expectations in each region yield scenario-wise parameter values. These scenarios are, then, used to formulate a two-stage stochastic program, which yields a lower bound for the original stochastic problem. We provide theoretical basis of our bounding approach for both the expectation and CVaR objectives. Our discrete bounding method generates exact lower bounds, as against the probabilistic bounds generated by Sample Average Approximation. We also present results of our numerical experimentation to compare the performances of these two approaches in terms of the bound value obtained and the CPU time required. The problem pertaining to integrated batching and scheduling of jobs on capacitated parallel machines that we consider arises in the primary manufacturing sector of a pharmaceutical supply chain. We, first, develop a comprehensive mathematical programming model that can accommodate various realistic features of this problem. These features include batch production, sequence-dependent setup time/cost, and inter-period carryover of setup status. We further derive several valid inequalities that are based on the embedded subproblem structure. We also consider an alternative formulation (termed the Plant Location model) based on the lot-sizing perspective of the problem. Noting the resemblance of the campaign sequencing subproblem to the high multiplicity asymmetric traveling salesman problem (HMATSP), we adapt various ideas from the HMATSP to enforce the connectivity of the sequencing graph. Due to the complexity of this problem, we also explore the possibility of applying column generation technique for its solution. Various schemes of problem decomposition are considered, along with the use of dual stabilization technique to improve the convergence of the column generation procedure. We also develop heuristic methods to generate initial feasible solutions that further enhance the performance of the column generation method. A computational experimentation has been conducted on a data set that mimics real-life problem instances. It illustrates the effectiveness of using the proposed column generation method. Ph. D.

Power consumption is one of the most important aspects in mobile and wireless communications. Existing research has shown significant power reduction through limited radio reconfiguration based on the channel conditions, especially for short range sensor network applications. A cognitive radio (CR) is an intelligent wireless communication system which is able to determine the most favorable operating parameters (cognition) based on the radio environment and its own capabilities and characteristics (awareness) and reconfigure the radio accordingly (reconfigurability). This work leverages the advances in cognitive radio technology to dynamically implement favorable trade-offs in radio parameters to achieve more efficient use of radio resource (e.g., minimizing power consumption) on the required Quality of Service (QoS) of an application and channel. A CR-based approach enables us not only to adjust modulation, coding, and radiated power as in a conventional radio, but also to learn and to control component characteristics (e.g., the power amplifier (PA) efficiency characteristic) to minimize power consumption. Significant power savings using this approach are shown in this work for single input single output (SISO) systems and multiple input multiple output (MIMO) systems. This work has a broad potential impact on the research of improving power efficiency of communication systems. It establishes a cognitive radio based methodology for system power consumption optimization. It emphasizes the difference between radiated power (power radiated from the transmit antenna) and the consumed power (power drawn from the power source, such as a battery). It provides a way to connect communication (which usually cares about radiated power, received signal to noise ratio, etc.) to hardware (which focuses on speed, efficiency, power consumption, etc.) and software (which emphasizes complexity, speed, etc.). This design methodology enhances the capability to jointly optimize communication, hardware, and software. In addition, this CR-based framework can be adapted for general radio resource management with various radio operation optimization targets, such as spectrum utilization. Ph. D.

This dissertation addresses three important open questions in the context of generating random vectors having discrete support. The first question relates to the "NORmal To Anything" (NORTA) procedure, which is easily the most widely used amongst methods for general random vector generation. While NORTA enjoys such popularity, there remain issues surrounding its efficient and correct implementation particularly when generating random vectors having denumerable support. These complications stem primarily from having to safely compute (on a digital computer) certain infinite summations that are inherent to the NORTA procedure. This dissertation addresses the summation issue within NORTA through the construction of easily computable truncation rules that can be applied for a range of discrete random vector generation contexts. The second question tackled in this dissertation relates to developing a customized algorithm for generating multivariate Poisson random vectors. The algorithm developed (TREx) is uniformly fast—about hundred to thousand times faster than NORTA—and presents opportunities for straightforward extensions to the case of negative binomial marginal distributions. The third and arguably most important question addressed in the dissertation is that of exact nonparametric random vector generation on finite spaces. Specifically, it is wellknown that NORTA does not guarantee exact generation in dimensions higher than two. This represents an important gap in the random vector generation literature, especially in view of contexts that stipulate strict adherence to the dependency structure of the requested random vectors. This dissertation fully addresses this gap through the development of Maximum Entropy methods. The methods are exact, very efficient, and work on any finite discrete space with stipulated nonparametric marginal distributions. All code developed as part of the dissertation was written in MATLAB, and is publicly accessible through the Web site https://filebox.vt.edu/users/pasupath/pasupath.htm. Ph. D.

This dissertation is a collection of several models to understand airline and passenger dynamics in the National Airspace System (NAS). Agent-based modeling is one of the most widely used modeling simulation-analysis approaches to understanding the dynamic behavior of complex systems. The usefulness of agent-based modeling has been demonstrated by simulating the complex interactions between airlines, travelers, and airports of a small-scale transportation system. Three airlines, one low cost and two network airlines are simulated to examine how each airline behaves over time to maximize their profit margins for a given passenger demand and operation cost structure. Passenger mode choice and itinerary choice sub modules are embedded in the framework to characterize traveler agent's response to the evolved airline schedule. An airport delay model was implemented to estimate the average delay at each airport. The estimated delay fed into the mode choice and itinerary choice models to update the travel time related variables. International passenger demand is a very important component of the air transportation system in the United States. The proportion of international enplanements relative to total enplanements increased from 8% in 1990 to 11% in 2008. Nine linear regression models are developed to forecast the enplanements from the United States to each of nine international regions. The international enplanements from the CONUS to each world region are modeled as a function of GDP and GDP per capita of both the United States and the specific region. A dummy variable is also used to account for the effects of September 11, 2001. The total number of international enplanements is forecast to increase from 74.7 million in 2008 to 184.4 million in 2028. The average annual growth rate is expected to be 4.7%. The European Union – United States Open Skies Agreement, which became effective March 30, 2008. Mathematical models are developed to forecast the effect of EU-US Open Skies Agreement on commercial airline passenger traffic over the North Atlantic Ocean. Nine econometric models were developed to forecast passenger traffic between the United States and nine selected European countries between 2008 through 2020. 68 new nonstop flights between the United States airports and the European airports are predicted by the model in 2020 using the airport pair passenger demand forecast. London, Heathrow is demonstrated as an example for rerouting the excess air travel passengers from one airport to other airports when the airport operational capacity is exceeded. The proportion of international enplanements relative to total enplanements within CONUS increased from 8% in 1990 to 11% in 2008. 51% of the sampled international and U.S. territories passengers served by U.S. carriers had at least one domestic coupon in 2007. The number of DOI passengers through airport-pairs in each of the historical years (1990-2007) is estimated based on the adjusted 100% international itineraries including pure international itineraries plus the non-CONUS itineraries. The total number of DOI enplanements is estimated to grow from 37.3 million in 1990 to 79.4 million in 2007. 193 CONUS airports are estimated to have at least 10,000 DOI enplanements in 2007. The number of DOI enplanements is forecast to grow from 79.4 million in 2007 to 206.2 million in 2030 with average growth rate of 4.2% per year. In recent years, there has been an increasing use of secondary airports both in Europe and the U.S. Regional airports have long been considered as a possible source of relief to reduce airport congestion at the hub airport and to efficiently accommodate future air travel demand. The conditions under which the secondary airports develop in a metropolitan area are examined. Fifteen multi-airport systems including 19 Operational Evolution Plan airports and 25 active secondary airports are identified in the National Airspace System. Diverse trends of traffic distribution among airports in the same metropolitan area are observed. We observed that the number of markets served at the secondary airports is less than that at the primary airport in the same metropolitan area. Most of the secondary airports are currently dominated by the low-cost carriers. The share of seats supplied by the low-cost carriers at the secondary airports has increased during the period 1990-2008. Full service carriers concentrate their service mainly on the primary airport in all the multi-airport systems analyzed. The average seat capacity per aircraft at the secondary airports is higher than that of primary airports in most of the multi-airport systems. The secondary airports mainly serve the domestic O&D passengers. Ph. D.

Position location has become one of the more important tasks for improving communication and networking performance for future commercial wireless systems. It is also the enabling technology for many control and sensing applications envisioned by the wireless sensor networks (WSN). Despite its meaningfulness and many algorithms being developed in the past several years, position location in harsh propagation environments remains to be a challenging issue, due mainly to the lack of sufficient infrastructure support and the prominent phenomenon of non-line-of-sight (NLOS) signal propagation. Recently, adopting the concept of collaborative position location has attracted much research interest due to its potential in overcoming the abovementioned two difficulties. In this work, we approach collaborative position location from two different angles. Specifically, we investigate the optimal performance of collaborative position location, which serves as a theoretical performance benchmark. In addition, we developed a computationally efficient algorithm for collaborative position location and incorporated an effective NLOS mitigation method to improve its performance in NLOS-dense environments. Overall, our work provides insight into both theoretical and practical aspects of collaborative position location. Ph. D.

Dilemma zone (DZ), also called decision zone in other literature, is an area where drivers face an indecisiveness of stopping or crossing at the yellow onset. The DZ issue is a major reason for the crashes at high-speed signalized intersections. As a result, how to prevent approaching vehicles from being caught in the DZ is a widely concerning issue. In this dissertation, the author addressed several DZ-associated issues, including the new stochastic safety measure, namely dilemma hazard, that indicates the vehicles' changing unsafe levels when they are approaching intersections, the optimal advance detector configurations for the multi-detector green extension systems, the new dilemma zone protection algorithm based on the Markov process, and the simulation-based optimization of traffic signal systems with the retrospective approximation concept. The findings include: the dilemma hazard reaches the maximum when a vehicle moves in the dilemma zone and it can be calculated according the caught vehicles' time to the intersection; the new (optimized) GES design can significantly improve the safety, but slightly improve the efficiency; the Markov process can be used in the dilemma zone protection, and the Markov-process-based dilemma zone protection system can outperform the prevailing dilemma zone protection system, the detection-control system (D-CS). When the data collection has higher fidelity, the new system will have an even better performance. The retrospective approximation technique can identify the sufficient, but not excessive, simulation efforts to model the true system and the new optimization algorithm can converge fast, as well as accommodate the requirements by the RA technique. Ph. D.

The issues and concerns facing recreation managers, academicians, and other practitioners are now often complex and important enough that solving them requires more than the sum of parts from social and physical disciplines. To that end, this dissertation document identifies and addresses three research projects that in varying proportions draw from the social and ecological aspects of recreation management. The first of three articles in this dissertation examined approximation of cross sectional soil profiles on foot trails. Monitoring this ecological indicator with current field techniques can be expensive and time-consuming for managers. Therefore, this article described a modified procedure for assessing trail soil loss and discusses several potentially useful geometric curves for approximating the cross-section of a trail at a given sampling point and in aggregate across a trail network. Differences in profiles for each study area and implications for inventorying and monitoring were discussed. The second article examined integration of soundscape and hiker spatial modeling. GPS data were used to generate a spatial model of hiker travel, soundscape modeling software calibrated with field data was used to generate a spatial model of sound, and the models were integrated in a geographic information system to provide insights for baseline and an alternative management option scenario. The findings suggested that small changes in soundscape, based on altered management practices, can have large effects on visitors' hiking experiences in terms of soundscape. The third article discussed an observational study examining several integrative and additive, information/education and site management approaches to preventing natural resource damage along backcountry trails. Video surveillance equipment unobtrusively captured hiker behaviors within the study area for each treatment. The findings suggested that direct, obtrusive measures (e.g., low symbolic rope fencing) in some cases can outperform multiple concurrent measures that are less direct and/or obtrusive. Implications on aesthetics, experiences, and management decision-making were discussed. Ph. D.

Developing equations to compute congestion measures for the general G/G/s/c queueing model and networks of such nodes has always been a challenge. One approach to analyzing such systems is to approximate the model-specified general input processes and distributions by processes and distributions from the more computationally friendly family of phase-type processes and distributions. We develop numerical approximation methods for analysis of general time-dependent queueing nodes by introducing new approximations for the time-dependent first two moments of the number-in-system and departure-count processes. Ph. D.

Several studies in the field of outdoor recreation management and planning have used computer simulation modeling to demonstrate its utility as a tool to help managers monitor encounters and similar visitor use-related indicators of quality. However, previous applications of computer simulation modeling to outdoor recreation planning and management have generally done little to assess the reliability, or precision, of model estimates. The purpose of this research is to explore several questions concerning the reliability of computer simulation model estimates for monitoring wilderness solitude-related indicators of quality. In particular, can reliable estimates of solitude-related indicators be generated for low use recreation environments, such as backcountry and wilderness areas? Is there a spatial component to questions about the reliability of computer simulation estimates for low use visitor landscapes? The research presented in this thesis examines the reliability of computer simulation estimates of wilderness solitude indicators that account for the timing and location of hiking and camping encounters in the backcountry of Great Smoky Mountains National Park. This study was designed to model visitor use and inter-group encounters in the Cosby and Big Creek areas of Great Smoky Mountains National Park, which are located within the park's proposed wilderness area. Two primary types of information about visitor use in the study area were collected to construct the computer simulation model in this study. First, information was gathered about the amount of visitation to the study area; second, information was collected about visitors' travel routes within the study area. Three alternative methods were used to estimate the number of replications needed to obtain desired levels of precision for the visitor-based and spatially based computer simulation model outputs. The results suggest that computer simulation models of visitor use can generate precise estimates for a small to moderate number of visitor-based and spatially-based outputs. However, there are constraints to generating precise estimates of use-related outputs as the number of outputs estimated simultaneously becomes large. This challenge is particularly pronounced in cases where at least some of the outputs are derived for low use attractions, trails, or camping locations. Master of Science