Title from PDF of title page. / Document formatted into pages; contains 124 pages. / Thesis (Ph.D.)--University of South Florida, 2003. / Includes bibliographical references. / Text (Electronic thesis) in PDF format. / ABSTRACT: As technology scales down and CMOS circuits are powered by lower supply voltages, standby leakage current becomes significant. A behavioral level framework for the synthesis of data-paths with low leakage power is presented. There has been minimal work done on the behavioral synthesis of low leakage datapaths. We present a fast architectural simulator for leakage (FASL) to estimate the leakage power dissipated by a system described hierarchically in VHDL. FASL uses a leakage power model embedded into VHDL leafcells. These leafcells are characterized for leakage accurately using HSPICE. We present results which show that FASL measures leakage power significantly faster than HSPICE, with less than a 5% loss in accuracy, compared to HSPICE. We present a comprehensive framework for synthesizing low leakage power data-paths using a parameterized Multi-threshold CMOS (MTCMOS) component library. / ABSTRACT: The component library has been characterized for leakage power and delay as a function of sleep transistor width. We propose four techniques for minimization of leakage power during behavioral synthesis: (1) leakage power management using MTCMOS modules; (2) an allocation and binding algorithm for low leakage based on clique partitioning; (3) selective binding to MTCMOS technology, allowing the designer to have control over the area overhead; and (4) a performance recovery technique based on multi-cycling and introduction of slack, to alleviate the loss in performance attributed to the introduction of MTCMOS modules in the data-path. Finally, we propose two iterative search based techniques, based on Tabu search, to synthesize low leakage data-paths. The first technique searches for low leakage scheduling options. The second technique simultaneously searches for a low leakage schedule and binding. It is shown that the latter technique of unified search is more robust. / ABSTRACT: The quality of results generated bytabu-based technique are superior to those generated by simulated annealing (SA) search technique. / System requirements: World Wide Web browser and PDF reader. / Mode of access: World Wide Web.

Commuter rail systems, operating on unused or under-used railroad rights-of-way, are being introduced into many urban transportation systems. Since locations of available rail rights-of-way were typically chosen long ago to serve the needs of rail freight customers, these locations are not optimal for commuter rail users. The majority of commuter rail users do not live or work within walking distance of potential commuter rail stations, so provision of quick, convenient access to and from stations is a critical part of overall commuter decisions to use commuter rail. Minimizing access time to rail stations and final destinations is crucial if commuter rail is to be a viable option for commuters. Well-designed feeder routes or circulator systems are regarded as potential solutions to provide train station to ultimate destination access. Transit planning for main line or feeder routes relies upon static demand estimates describing a typical day. Daily and peak-hour demands change in response to the state of the transport system, as influenced by weather, incidents, holiday schedules and many other factors. Recent marketing successes of “smart phones” might provide an innovative means of obtaining real time data that could be used to identify optimal paths and stop locations for commuter rail circulator systems. Such advanced technology could allow commuter rail users to provide real-time final destination information that would enable real time optimization of feeder routes. This dissertation focuses on real time optimization of the Commuter Rail Circulator Route Network Design Problem (CRCNDP). The route configuration of the circulator system – where to stop and the route among the stops – is determined on a real-time basis by employing adaptive Tabu Search to timely solve an MIP problem with an objective to minimize total cost incurred to both transit users and transit operators. Numerical experiments are executed to find the threshold for the minimum fraction of travelers that would need to report their destinations via smart phone to guarantee the practical value of optimization based on real-time collected demand against a base case defined as the average performance of all possible routes. The adaptive Tabu Search Algorithm is also applied to three real-size networks abstracted from the Martin Luther King (MLK) station of the new MetroRail system in Austin, Texas. / text

The research reported in this thesis considers the classical combinatorial optimization problem known as the Generalized Assignment Problem (GAP). Since the mid 1970's researchers have been developing solution approaches for this particular type of problem due to its importance both in practical and theoretical terms. Early attempts at solving GAP tended to use exact integer programming techniques such as Branch and Bound. Although these tended to be reasonably successful on small problem instances they struggle to cope with the increase in computational effort required to solve larger instances. The increase in available computing power during the 1980's and 1990's coincided with the development of some highly efficient heuristic approaches such as Tabu Search (TS), Genetic Algorithms (GA) and Simulated Annealing (SA). Heuristic approaches were subsequently developed that were able to obtain high quality solutions to larger and more complex instances of GAP. Most of these heuristic approaches were able to outperform highly sophisticated commercial mathematical programming software since the heuristics tend to be tailored to the problem and therefore exploit its structure. A new approach for solving GAP has been developed during this research that combines the exact Branch and Bound approach and the heuristic strategy of Tabu Search to produce a hybrid algorithm for solving GAP. This approach utilizes the mathematical programming software Xpress-MP as a Branch and Bound solver in order to solve sub-problems that are generated by the Tabu Search guiding heuristic. Tabu Search makes use of memory structures that record information about attributes of solutions visited during the search. This information is used to guide the search and in the case of the hybrid algorithm to generate sub problems to pass to the Branch and Bound solver. The new algorithm has been developed, imp lemented and tested on benchmark test problems that are extremely challenging and a comprehensive report and analysis of the experimentation is reported in this thesis.

During an emergency evacuation, most people will use their vehicles to evacuate. However, there is a group of people who do not have access to reliable transportation or for some reason cannot drive, even if they have their own automobiles - the carless. There are different groups of carless (disabled, medically homebound, poor or immigrant populations, etc.) who require different forms of transportation assistance during an emergency evacuation. In this study we focus on those carless who are physically intact and able to walk to a set of designated locations for transportation during an emergency, and we propose using public transit and school buses to evacuate this carless group. A model has been developed to accommodate the use of public transit and school buses to efficiently and effectively evacuate the carless. The model has two parts. Part 1 is a location problem which aims at congregating the carless at some specific locations called evacuation sites inside the affected area. To achieve this goal, the affected area is partitioned into zones and this congregating of the carless has been formulated as a Single Source Capacitated Facility Location Problem. Changes in the demand of the carless in zones over different periods of a day and over different days of the week have been considered and included in the model. A walking time constraint is explicitly considered in the model. A heuristic developed by Klincewicz and Luss (1986) has been used to solve this location model.Part 2 is a routing problem which aims at obtaining itineraries of buses to pick up the carless at evacuation sites and transport them to safe locations outside the affected area, such that the total number of carless evacuated with the given time limit is maximized. A Tabu search heuristic has been developed for solving the routing problem. Computational results show that the Tabu search heuristic efficiently and effectively solves the routing problem; in particular, the initial heuristic produces a high quality initial solution in very short time. This study has also made slight contribution to the development of the Tabu search technique.

Tabu search based algorithms are among the widely used heuristic algorithms for combinatorial optimization problems. In this project, we propose an improved enhanced tabu search algorithm for the well-known combinatorial optimization problem, the quadratic assignment problem (QAP). The new algorithm was tested on a number of instances from the library of the QAP instances  QAPLIB. The results obtained from the experiments show that the proposed algorithm appears to be superior to the earlier "pure" tabu search algorithms on many instances of the QAP.

In the last two decades, the western world witnessed a continuous rise in the health expenditure. Meanwhile, complaints from patients on excessive waiting times are also increasing. In the past, many researchers have tried to devise appointment scheduling rules to provide trade-offs between maximizing patients’ satisfaction and minimizing the costs of the health providers. For instance, this challenge appears appointment scheduling problems (ASP). Commonly used methods in ASP include analytical methods, simulation studies, and combination of simulation with heuristic approaches. Analytical methods (e.g., queuing theory and mathematical programming) face challenges of fully capturing the complexities of systems and usually make strong assumptions for tractability of problems. These methods simplify the whole system to a single-stage unit and ignore the actual system factors such as the presence of multiple stages and/or resource constraints. Simulation studies, conversely, are able to model most complexities of the actual system, but they typically lack an optimization strategy to deliver optimal appointment schedules. Also, heuristic approaches normally are based on intuitive rules and do not perform well as standalone methods. In order to reach an optimal schedule while considering complexities in actual health care systems, this thesis proposes efficient and effective methods that yield (near) optimal appointment schedules by integrating mathematical programming, a tabu search optimization algorithm and discrete event simulation. The proposed methodologies address the challenges and complexities of scheduling in real world multistage healthcare units in the presence of stochastic service durations, a mix of patient types, patients with heterogeneous service sequence, and resource constraints. Moreover, the proposed methodology is capable of finding the optimum considering simultaneously multiple performance criteria. A Pareto front (a set of optimal solutions) for the performance criteria can be obtained using the proposed methods. Healthcare management can use the Pareto front to choose the appropriate policy based on different conditions and priorities. In addition, the proposed method has been applied to two case studies of Operating Rooms departments in two major Canadian hospitals. The comparison of actual schedules and the ones yielded by the proposed method indicates that proposed method can improve the appointment scheduling in realistic clinical settings.

To understand evolution, and to discover how different species are related, gene order analysis is a useful tool. Problems in this area can usually be formulated in a combinatorial language. We regard genomes as signed, or unsigned permutations, and thus evolutionary operations like inversions (reversing the order of a segment of genes) are easy to describe combinatorially. A commonly studied problem is to determine the evolutionary distance between two species. This is estimated by several combinatorial distances between gene order permutations, for instance the inversion distance. The main objective of this work was to survey the existing algorithms for genome comparison and to present new approach for solving this problem. The work led to these results: - We have surveyed existing approaches of genome comparison, namely comparison by inversion distance in signed and unsigned cases. It appeared that sorting signed genomes by inversions is done in quadratic time, but sorting unsigned genomes by inversions is NP-hard. - We have proposed the method of how to apply heuristic algorithms for sorting unsigned genomes by inversions. - We have applied tabu search and genetic algorithm to solve the sorting unsigned genomes by inversions problem. - We have experimentally proven, that the worst case solutions to sorting unsigned genomes by inversions found by heuristics (tabu search and genetic algorithm) are better then ones expected from best known approximating algorithm used for... [to full text]

In the last two decades, the western world witnessed a continuous rise in the health expenditure. Meanwhile, complaints from patients on excessive waiting times are also increasing. In the past, many researchers have tried to devise appointment scheduling rules to provide trade-offs between maximizing patients’ satisfaction and minimizing the costs of the health providers. For instance, this challenge appears appointment scheduling problems (ASP). Commonly used methods in ASP include analytical methods, simulation studies, and combination of simulation with heuristic approaches. Analytical methods (e.g., queuing theory and mathematical programming) face challenges of fully capturing the complexities of systems and usually make strong assumptions for tractability of problems. These methods simplify the whole system to a single-stage unit and ignore the actual system factors such as the presence of multiple stages and/or resource constraints. Simulation studies, conversely, are able to model most complexities of the actual system, but they typically lack an optimization strategy to deliver optimal appointment schedules. Also, heuristic approaches normally are based on intuitive rules and do not perform well as standalone methods. In order to reach an optimal schedule while considering complexities in actual health care systems, this thesis proposes efficient and effective methods that yield (near) optimal appointment schedules by integrating mathematical programming, a tabu search optimization algorithm and discrete event simulation. The proposed methodologies address the challenges and complexities of scheduling in real world multistage healthcare units in the presence of stochastic service durations, a mix of patient types, patients with heterogeneous service sequence, and resource constraints. Moreover, the proposed methodology is capable of finding the optimum considering simultaneously multiple performance criteria. A Pareto front (a set of optimal solutions) for the performance criteria can be obtained using the proposed methods. Healthcare management can use the Pareto front to choose the appropriate policy based on different conditions and priorities. In addition, the proposed method has been applied to two case studies of Operating Rooms departments in two major Canadian hospitals. The comparison of actual schedules and the ones yielded by the proposed method indicates that proposed method can improve the appointment scheduling in realistic clinical settings.

There is an increasing interest in sustainability and a growing debate about environmental policy measures aiming at the reduction of green house gas emissions across di erent economic sectors worldwide. The transportation sector is one major greenhouse gas emitter which is heavily regulated to reduce its dependance on oil. These regulations along with the growing customer awareness about global warming has led vehicle manufacturers to seek di erent technologies to improve vehicle e ciencies and reduce the green house gases emissions while at the same time meeting customer's expectation of mobility and exibility. Plug-in hybrid electric vehicles (PHEV) is one major promising solution for a smooth transition from oil dependent transportation sector to a clean electric based sector while not compromising the mobility and exibility of the drivers. In the medium term, plug-in hybrid electric vehicles (PHEV) can lead to signi cant reductions in transportation emissions. These vehicles are equipped with a larger battery than regular hybrid electric vehicles which can be recharged from the grid. For short trips, the PHEV can depend solely on the electric engine while for longer journeys the alternative fuel can assist the electric engine to achieve extended ranges. This is bene cial when the use pattern is mixed such that and short long distances needs to be covered. The plug-in hybrid electric vehicles are well-suited for logistics since they can avoid the possible disruption caused by charge depletion in case of all-electric vehicles with tight time schedules. The use of electricity and fuel gives rise to a new variant of the classical vehicle routing with time windows which we call the plug-in hybrid electric vehicle routing problem with time windows (PHEVRPTW). The objective of the PHEVRPTW is to minimize the routing costs of a eet of PHEVs by minimizing the time they run on gasoline while meeting the demand during the available time windows. As a result, the driver of the PHEV has two decisions to make at each node: (1) recharge the vehicle battery to achieve a longer range using electricity, or (2) continue to the next open time window with the option of using the alternative fuel. In this thesis, we present a mathematical formulation for the plug-in hybrid-electric vehicle routing problem with time windows. We solve this problem using a Lagrangian relaxation and we propose a new tabu search algorithm. We also present the rst results for the full adapted Solomon instances.

[pt] As técnicas de tecnologia de grupos vêm sendo largamente usadas em muitos sistemas de manufatura. Vários algoritmos têm sido propostos para o projeto otimizado de eficientes células de manufatura. O problema de formação de células deve levar em conta vários objetivos: o número de operações gargalo, o número de máquinas e/ou peças gargalo, o fluxo intercelular, os custos de subcontratação, os custos de duplicação de máquinas e a carga da máquina e/ou célula mais sobrecarregada, entre outros. Nesta tese propõe-se uma metodologia multi- critério para resolver o problema de formação de células com múltiplos objetivos. Este enforque é baseado no uso da meta-heurística busca tabu para resolver uma seqüência de problemas com objetivos simples e restrições múltiplas, onde cada objetivo é minimizado individualmente, segundo sua ordem de importância. Resultados computacionais envolvendo uma aplicação para um problema bi-critério são apresentados para casos com até 100 máquinas e 1000 peças. / [en] Group tecnology techniques are now widely used in many manufacturing systems. Severla algorithms have been proposed for the optimal design of efficient manufacturing cells. The cell formation problem must take into account several objectives: the number of bottleneck operations, the number of bottleneck machines and/or parts, the intercell flow, the intracell workload balancing, the subcontracting cost, the machine duplication costs, and the workload of the busiest machine and/or cell, among athers. In this work, we propose a multi-criteria methodology for solving the cell formation problem with multiple objectives. This approach is based on the use of the tabu search meta-heuristic for solving a sequence of single-objective, multi-contrained problems, in wich each objective is taken and optimized in turn, following their order of relative importance. Computational results concerning an application to a bi-criteria problem are reported for instances with up 100 machines and 1000 parts.