Tuition and fees are a major source of income for colleges and universities and a major portion of the cost of a student's education. The university administration's task of making sound and effective tuition and fee policy decisions is becoming both more critical and more complex. This is a result of the increased reliance on student-generated tuition-and-fee income, the declining college-age student population, reductions in state and Federal funds, and escalating costs of operation. The comprehensive computerized decision support system (DSS) developed in this research enhances the administration's planning, decision-making, and policy-setting processes. It integrates data and reports with modeling and analysis in order to provide a systematic means for analyzing tuition and fee problems, at a detailed and sophisticated level, without the user having to be an expert in management science techniques or computers. The DSS with its imbedded multi-year goal programming (GP) model allocates the university's revenue requirements to charges for individual student categories based on a set of user-defined objectives, constraints, and priorities. The system translates the mathematical programming model into a valuable decision-making aid by making it directly and readily accessible to the administration. The arduous tasks of model formulation and solution, the calculation of the model's parameter values, and the generation of a series of reports to document the results are performed by the system; whereas, the user is responsible for defining the problem framework, selecting the goals, setting the targets, establishing the priority structure, and assessing the solution. The DSS architecture is defined in terms of three highly integrated subsystems - dialog, data, and models - that provide the following functions: user/system interface, program integration, process control, data storage and handling, mathematical, statistical, and financial computations, as well as display, memory aid, and report generation. The software was developed using four programming languages/systems: EXEC 2, FORTRAN, IFPS, and LINDO. While the system was developed, tested, and implemented at Virginia Polytechnic Institute and State University, the concepts developed in this research are general enough to be applied to any public institution of higher education. / Ph. D.

Fairness emerges as a vital concern to decision makers as crucial as efficiency, if not more important. Fair operations decisions are aimed at distributive justice in various scenarios. In this dissertation, we study two examples of distributively fair decision making in operations research, a dynamic fair allocation problem and a subpopulational robustness assessment problem for machine learning models. We first study a dynamic allocation problem in which 𝑇 sequentially arriving divisible resources are to be allocated to a number of agents with concave utilities. The joint utility functions of each resource to the agents are drawn stochastically from a known joint distribution, independently and identically across time, and the central planner makes immediate and irrevocable allocation decisions. Most works on dynamic resource allocation aim to maximize the utilitarian welfare, i.e., the efficiency of the allocation, which may result in unfair concentration of resources on certain high-utility agents while leaving others' demands under-fulfilled. In this work, aiming at balancing efficiency and fairness, we instead consider a broad collection of welfare metrics, the Hölder means, which includes the Nash social welfare and the egalitarian welfare. To this end, we first study a fluid-based policy derived from a deterministic surrogate to the underlying problem and show that for all smooth Hölder mean welfare metrics it attains an 𝑂 (1) regret over the time horizon length 𝑇 against the hindsight optimum, i.e., the optimal welfare if all utilities were known in advance of deciding on allocations. However, when evaluated under the non-smooth egalitarian welfare, the fluid-based policy attains a regret of order 𝛩 (√𝑇). We then propose a new policy built thereupon, called Backward Infrequent Re-solving (𝖡𝖨𝖱), which consists of re-solving the deterministic surrogate problem at most 𝑂 (log 𝑇) times. We show under a mild regularity condition that it attains a regret against the hindsight optimal egalitarian welfare of order 𝑂 (1) when all agents have linear utilities and 𝑂 (log 𝑇) otherwise. We further propose the Backward Infrequent Re-solving with Thresholding (𝖡𝖨𝖱𝖳) policy, which enhances the (𝖡𝖨𝖱𝖳) policy by thresholding adjustments and performs similarly without any assumption whatsoever. More specifically, we prove the (𝖡𝖨𝖱𝖳) policy attains an 𝑂 (1) regret independently of the horizon length 𝑇 when all agents have linear utilities and 𝑂 (log²⁺^𝜀) otherwise. We conclude by presenting numerical experiments to corroborate our theoretical claims and to illustrate the significant performance improvement against several benchmark policies. The performance of ML models degrades when the training population is different from that seen under operation. Towards assessing distributional robustness, we study the worst-case performance of a model over 𝒂𝒍𝒍 subpopulations of a given size, defined with respect to core attributes 𝑍. This notion of robustness can consider arbitrary (continuous) attributes 𝑍, and automatically accounts for complex intersectionality in disadvantaged groups. We develop a scalable yet principled two-stage estimation procedure that can evaluate the robustness of state-of-the-art models. We prove that our procedure enjoys several finite-sample convergence guarantees, including 𝒅𝒊𝒎𝒆𝒏𝒔𝒊𝒐𝒏-𝒇𝒓𝒆𝒆 convergence. Instead of overly conservative notions based on Rademacher complexities, our evaluation error depends on the dimension of 𝑍 only through the out-of-sample error in estimating the performance conditional on 𝑍. On real datasets, we demonstrate that our method certifies the robustness of a model and prevents deployment of unreliable models.

Thesis (MBA)--Stellenbosch University, 2002. / ENGLISH ABSTRACT: Changes in the global salt market have presented Walvis Bay Salt Refiners with an opportunity to increase its current sales by approximately 40%. Following several pre-feasibility studies, the expansion project plan was created. The construction of new ponds, canals and sluices were to be performed by a subcontractor as selected through a tender process. The scope of the work comprised approximately 70% of the total project cost and it also represented the most critical part of the expansion project. Thomas Saaty’s Analytic Hierarchy Process, was used as a group decision support system for the selection of the most suitable subcontractor. The weighted average mean method was used to aggregate individual scores. A sensitivity analysis was performed following the final outcome to gain a deeper understanding of the problem, obtain a measure of margin between subcontractor scores and to check for the correctness of numbers. / AFRIKAANSE OPSOMMING: Veranderinge in die wêreld soutmark het vir Walvis Bay Salt Refiners 'n geleentheid gebied om sy verkope met ʼn beraamde 40% te verhoog. Na verskeie voorondersoeke is 'n volledige projekplan opgestel. Die vervaardiging van damwalle, kanale en sluise sou deur 'n kontrakteur gedoen word wat deur ʼn tenderprosedure gekeur sou word. Die omvang van hierdie gedeelte van die uitbreidingsprojek verteenwoordig ongeveer 70% van die totale projekkostes en is terselfdelyk die mees sensitiewe gedeelte van die projek. Thomas Saaty se Analytic Hierarchy Process is gebruik as die groepbesluitnemingsondersteuningstelsel om die mees geskikte kontrakteur te kies. Die geweegde gemiddelde is gebruik om die individuele oordele saam te voeg. Sensitiwiteits analise is uitgevoer nadat die finale uitslag bepaal is om sodoende beter insig in die probleem te ontwikkel, om ʼn beter onderskeiding tussen die kontrakteur puntetellings te kry en om die juistheid van die syfers na te gaan.

This research assisted the Yuma, Arizona citrus orchard manager in his strategic planning for achieving a low-cost position in a focused segment of the citrus industry. Citrus growers in the Yuma district are faced with major changes in their competitive environment and must adopt new strategic plans in order to continue to compete effectively in what has recently become a global industry. Since the planning horizon for new citrus orchards is in excess of 20 years, a long range planning model was developed to aid in evaluating alternative operating strategies. This research established the interrelatedness of water, nitrogen, and phosphorous relative to the yields of Valenica Oranges, Lisbon Lemons, and Redblush Grapefruit on Rough Lemon, Sour Orange, and Troyer rootstocks. A computer simulation model was used to evaluate optimal operating policies for a variety of resource prices and market conditions. The methodology utilized in development of the simulation model was unique in that it emulates individual tree performance from the time of planting until maturation. Four operating strategies were investigated and the profit maximizing and cost minimizing strategies were found to be significant. Evaluation of market selling prices indicated that the profit maximizing strategy was optimal except at very low market prices where the cost minimization strategy was optimal. Price sensitivity for water and fertilizer resources was investigated. Operating strategies were not affected by water price increases over the foreseeable future, however, price changes in nitrogen and phosphorous were found to affect the optimal operating strategy primarily through the substitution of manure in the system. Existing horticultural practices in the Yuma growing area were confirmed by the research. Additional optimal operating strategies were suggested relative to market prices. The long run policy decision making process for orchard managers was enhanced.

by Lau Tze Yiu, Peter. / Thesis (Ph.D.)--Chinese University of Hong Kong, 1997. / Includes bibliographical references (p. 352-365). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web.

In this thesis we consider two comparative statics questions of changes in risk. The first question concerns situations where an individual faces some risk and has no control over the uncertain environment. In these situations we ask what kind of changes in risk will cause the individual's expected utility to increase. The second comparative statics question concerns situations where an individual faces some risk and has some control over the uncertain environment. In particular, we consider situations where the individual maximizes her expected utility with respect to some control parameter. Here we ask what kind of changes in risk will cause the individual's optimal value of the control parameter to increase. The existing literature has answered these questions for a class of individuals (for example, the class of risk averse individuals). This thesis differs from existing literature as it focuses on a given individual, and thus reveals some of the person-specific factors that affect individual?s responses to changes in risk. The aim of the thesis is to show how an order on distributions, termed single crossing likelihood ratio (SCLR) order, can intuitively answer both questions for a given individual. The main contributions of the thesis are as follows. First, the thesis presents the SCLR order and its main properties. Second, the thesis shows that the SCLR order can answer the above comparative statics questions in an intuitive way. In particular, the thesis shows that the answer to the above questions, with the use of the SCLR order, depends on a risk reference point which can be interpreted as a "certainty equivalent" point. Thus it is demonstrated that individual's responses to changes in risk are affected by her "certainty equivalent" point. Lastly, the results of the thesis can be used to provide an intuitive explanation of related existing results that were obtained for a class of individuals.

The surveillance of geographic borders and critical infrastructures using limited sensor capability has always been a challenging task in many homeland security applications. While geographic borders may be very long and may go through isolated areas, critical assets may be large and numerous and may be located in highly populated areas. As a result, it is virtually impossible to secure each and every mile of border around the country, and each and every critical infrastructure inside the country. Most often, a compromise must be made between the percentage of border or critical asset covered by surveillance systems and the induced cost. Although threats to homeland security can be conceived to take place in many forms, those regarding illegal penetration of the air, land, and maritime domains under the cover of day-to-day activities have been identified to be of particular interest. For instance, the proliferation of drug smuggling, illegal immigration, international organized crime, resource exploitation, and more recently, modern piracy, require the strengthening of land border and maritime awareness and increasingly complex and challenging national security environments. The complexity and challenges associated to the above mission and to the protection of the homeland may explain why a methodology enabling the design and optimization of distributed detection systems architectures, able to provide accurate scanning of the air, land, and maritime domains, in a specific geographic and climatic environment, is a capital concern for the defense and protection community. This thesis proposes a methodology aimed at addressing the aforementioned gaps and challenges. The methodology particularly reformulates the problem in clear terms so as to facilitate the subsequent modeling and simulation of potential operational scenarios. The needs and challenges involved in the proposed study are investigated and a detailed description of a multidisciplinary strategy for the design and optimization of detection architectures in terms of detection performance and cost is provided. This implies the creation of a framework for the modeling and simulation of notional scenarios, as well as the development of improved methods for accurate optimization of detection architectures. More precisely, the present thesis describes a new approach to determining detection architectures able to provide effective coverage of a given geographical environment at a minimum cost, by optimizing the appropriate number, types, and locations of surveillance and detection systems. The objective of the optimization is twofold. First, given the topography of the terrain under study, several promising locations are determined for each sensor system based on the percentage of terrain it is covering. Second, architectures of sensor systems able to effectively cover large percentages of the terrain at minimal costs are determined by optimizing the number, types and locations of each detection system in the architecture. To do so, a modified Genetic Algorithm and a modified Particle Swarm Optimization are investigated and their ability to provide consistent results is compared. Ultimately, the modified Particle Swarm Optimization algorithm is used to obtain a Pareto frontier of detection architectures able to satisfy varying customer preferences on coverage performance and related cost.

Fractional aircraft ownership programs offer companies and individuals all the benefits of owning private jet, such as safety, consistency, and guaranteed availability, at a fraction of the cost of owning an aircraft. In the fractional ownership model, the partial owners of an aircraft are entitled to certain number of hours per year, and the management company is responsible for all the operational considerations and making sure an aircraft is available to the owners at the requested time and location. This thesis research proposes advance optimization techniques to help the management company to optimally operate its available resources and provides tools for strategic decision making. The contributions of this thesis are: (i) The development of optimization methodologies to assign and schedule aircraft and crews so that all flight requests are covered at the lowest possible cost. First, a simple model is developed to solve the crew pairing and aircraft routing problem with column generation assuming that a crew stays with one specific aircraft during its duty period. Secondly, this assumption is partially relaxed to improve resource utilization by revising the simple model to allow a crew to use another aircraft when its original aircraft goes under long maintenance. Thirdly, a new comprehensive model utilizing Benders decomposition technique and a fleet-station time line is proposed to completely relax the assumption that crew stays with one specific aircraft. It combines the fleet assignment, aircraft routing, and crew pairing problems. In the proposed methodologies, real world details are taken into consideration, such as crew transportation and overtime costs, scheduled and unscheduled maintenance effects, crew rules, and the presence of non-crew-compatible fleets. Scheduling with time windows is also discussed. (ii) The analysis of operational strategies to provide decision making support. Scenario analyses are performed to provide insights on improving business profitability and aircraft availability, such as impact of aircraft maintenance, crew swapping, effect of increasing demand by Jet-card and geographical business expansion, size of company owned aircraft, and strategies to deal with the stochastic feature of unscheduled maintenance and demand.

D. Tech. Industrial Engineering. / Discusses purpose of this study is to derive a new model capable of advanced diagnosis and analysis of energy usage to determine the possible energy efficiency potential through the following in a single model: Analysis of industrial historical data; Prediction of the industrial energy baseline; Computation of the industrial energy efficiency; and Optimization of the industrial energy consumption usage. In this context, the development of a new model involves: Carrying out literature survey; Carrying out Mathematical Analysis of the dynamics of energy efficiency in an industry; Critically analyzing and testing existing models; Evolve a new and novel model; Test the model using data from specific industry; Apply the model to eleven industrial sectors in South Africa. This thesis on energy efficiency potential will be a milestone for different stakeholders, policymakers and decision makers in the energy sector at national and international levels who are, or will be interested in reducing energy input and still produce the observed output levels, by becoming technically efficient. The approach adopted by the study is the integration of Index Decomposition Analysis (IDA), Artificial Neural Network (ANN) and Data Envelopment Analysis (DEA) into a single model.This methodology combines modeling, which is at the core of an energy-management technique, with a wider interpretation of activity growth, structure and efficiency changes which contribute to changes in energy consumption.

This dissertation describes the development of a requirements analysis methodology that takes into account the concept of operations and the hierarchical decomposition of aerospace systems. At the core of the methodology, the Analytic Network Process (ANP) is used to ensure the traceability between the qualitative and quantitative information present in the hierarchical model. The proposed methodology is implemented to the requirements definition of a hurricane tracker Unmanned Aerial Vehicle. Three research objectives are identified in this work; (1) improve the requirements mapping process by matching the stakeholder expectations with the concept of operations, systems and available resources; (2) reduce the epistemic uncertainty surrounding the requirements and requirements mapping; and (3) improve the requirements down-selection process by taking into account the level of importance of the criteria and the available resources. Several challenges are associated with the identification and definition of requirements. The complexity of the system implies that a large number of requirements are needed to define the systems. These requirements are defined early in the conceptual design, where the level of knowledge is relatively low and the level of uncertainty is large. The proposed methodology intends to increase the level of knowledge and reduce the level of uncertainty by guiding the design team through a structured process. To address these challenges, a new methodology is created to flow-down the requirements from the stakeholder expectations to the systems alternatives. A taxonomy of requirements is created to classify the information gathered during the problem definition. Subsequently, the operational and systems functions and measures of effectiveness are integrated to a hierarchical model to allow the traceability of the information. Monte Carlo methods are used to evaluate the variations of the hierarchical model elements and consequently reduce the epistemic uncertainty. The proposed methodology is applied to the design of a hurricane tracker Unmanned Aerial Vehicles to demonstrate the origin and impact of requirements on the concept of operations and systems alternatives. This research demonstrates that the hierarchical modeling methodology provides a traceable flow-down of the requirements from the problem definition to the systems alternatives phases of conceptual design.