This study analyzes the change in whole-farm net revenues and nutrient reduction from the implementation of five best management practices (BMPs) on a typical beef and beef-poultry farm in Shenandoah County. Whole-farm net revenues, resource allocation, nutrient loss reductions, and the cost efficiency of reducing nutrient losses were analyzed to assess which BMPs are the most cost efficient to implement, assuming the baseline scenarios have no voluntarily applied BMPs. The effects of stacking additional BMPs, in combinations of two or more, were also assessed. No-till cropping, winter wheat cover crop, herbaceous riparian buffer, fencing, and P-based NMP were the BMPs that were analyzed. Incentive payments from state and federal governments were incorporated into the cost of BMP adoption. A brief analysis of a farmer's time value of money, with respect to incentive payments, was also conducted. Results indicated that no-till crop management was the most cost efficient BMP, and was the only BMP to increase net revenues for both farm models. Fencing and P-based NMP were the least cost efficient for the beef farm. For the beef-poultry farm, fencing was the least cost efficient. The implications of this study are that farmers that choose to adopt BMP should evaluate both their interests in maintaining (or increasing) farm net revenues along with their interest in improving water quality through the reduction of nutrient losses. There is potential for implementing multiple BMPs, while increasing net revenues from a farm's baseline scenario. For farmers and policy makers, no-till cropping can be a profitable and therefore cost efficient BMP to implement. Incentive payments are intended to encourage the adoption of BMPs by subsidizing a portion of the start-up costs. Policy makers should attempt to make cost-share payments reflect nutrient reduction goals. This can be done by analyzing both the compliance cost to farmers and the nutrient reduction effectiveness of BMPs. / Master of Science

Reinforcement learning algorithms hold promise in many complex domains, such as resource management and planning under uncertainty. Most reinforcement learning algorithms are iterative - they successively approximate the solution based on a set of samples and features. Although these iterative algorithms can achieve impressive results in some domains, they are not sufficiently reliable for wide applicability; they often require extensive parameter tweaking to work well and provide only weak guarantees of solution quality. Some of the most interesting reinforcement learning algorithms are based on approximate dynamic programming (ADP). ADP, also known as value function approximation, approximates the value of being in each state. This thesis presents new reliable algorithms for ADP that use optimization instead of iterative improvement. Because these optimization-based algorithms explicitly seek solutions with favorable properties, they are easy to analyze, offer much stronger guarantees than iterative algorithms, and have few or no parameters to tweak. In particular, we improve on approximate linear programming - an existing method - and derive approximate bilinear programming - a new robust approximate method. The strong guarantees of optimization-based algorithms not only increase confidence in the solution quality, but also make it easier to combine the algorithms with other ADP components. The other components of ADP are samples and features used to approximate the value function. Relying on the simplified analysis of optimization-based methods, we derive new bounds on the error due to missing samples. These bounds are simpler, tighter, and more practical than the existing bounds for iterative algorithms and can be used to evaluate solution quality in practical settings. Finally, we propose homotopy methods that use the sampling bounds to automatically select good approximation features for optimization-based algorithms. Automatic feature selection significantly increases the flexibility and applicability of the proposed ADP methods. The methods presented in this thesis can potentially be used in many practical applications in artificial intelligence, operations research, and engineering. Our experimental results show that optimization-based methods may perform well on resource-management problems and standard benchmark problems and therefore represent an attractive alternative to traditional iterative methods.

We present an implementation of a very recent approximation algorithm for scheduling jobs on a single machine with precedence constraints, minimising the total weighted completion time. We also evaluate the performance of this implementation. The algorithm was published by Shi Li in 2021 and is a (6+ε)-approximation algorithm for the multiprocessor problem P|prec|∑j wjCj. We have implemented a version which is a (2+ε)-approximation algorithm for the single processor problem 1|prec|∑j wjCj. This special case can easily be generalised to the multiprocessor case, as the two algorithms are based on the same LP relaxation of the problem. Unlike other approximation algorithms for this and similar problems, for example, those published by Hall, Schulz, Shmoys and Wein in 1997, and by Li in 2020, this algorithm has been developed with a focus on obtaining a good asymptotic run time guarantee, rather than obtaining the best possible guarantee on the quality of solutions. Li’s algorithm has run time O((n+κ) · polylog(n+κ) · log3 pmax · 1/ε2), where n is the number of jobs, κ is the number of precedence constraints and pmax is the largest of the processing times of the jobs. We also present a detailed explanation of the algorithm aimed at readers who do not necessarily have a background in scheduling and/or approximation algorithms, based on the paper by Li. Finally, we empirically evaluate how well (our implementation of) this algorithm performs in practice. The performance was measured on a set of 96 randomly generated instances, with the largest instance having 1024 jobs and 32 768 precedence constraints. We can find a solution for an instance with 512 jobs and 11 585 precedence constraints in 25 minutes. / Vi presenterar en praktisk implementation av en ny approximationsalgoritm för schemaläggning av jobb på en maskin med ordningsbivillkor, under minimering av den viktade summan av sluttider. Algoritmen, som publicerades av Shi Li år 2021, är en (6+ε)-approximationsalgoritm för multiprocessorproblemet P|prec|∑j wjCj. Vi har implementerat en version som är en (2+ε)-approximationsalgoritm för enprocessorproblemet 1|prec|∑j wjCj. Detta specialfall kan enkelt generaliseras till multiprocessorfallet, eftersom de två algoritmerna baseras på samma LP-relaxation av problemet. Till skillnad från andra approximationsalgoritmer för detta och liknande problem, exempelvis de från Hall, Schulz, Shmoys och Wein år 1997, och från Li år 2020, har denna algoritm utvecklats med fokus på att uppnå en bra garanti på asymptotisk körtid, istället för att försöka uppnå den bästa möjliga garantin på lösningarnas kvalité. Lis algoritm har körtid O((n+κ) · polylog(n+κ) · log3 pmax · 1/ε2), där n är antalet jobb, κ antalet ordningsbivillkor och pmax är den största körtiden bland jobben. En detaljerad beskrivning av algoritmen riktad till personer som inte nödvändigtvis har förkunskaper inom schemaläggning och/eller approximationsalgoritmer, baserad på artikeln, ges också. Slutligen utvärderar vi empiriskt hur väl (vår implementation av) denna algoritm presterar i praktiken. Implementationens egenskaper mättes på en uppsättning av 96 slumplässigt genererade instanser, där den största instansen har 1024 jobb och 32768 ordningsbivillkor. Med vår implementation kan vi hitta en lösning för en instans med 512 jobb och 11 585 precedencensbivillkor på 25 minuter.

[pt] O presente trabalho tem por objetivo desenvolver uma formulação e um sistema que seja uma ferramenta auxiliar para o engenheiro de projeto na definição de modelos de bielas e tirantes de estruturas bi-dimensionais. No sistema desenvolvido o usuário pode partir de um modelo complexo, com um número grande de barras nas mais variadas direções. Essa solução inicial pode se basear nos resultados em tensão de uma análise linear elástica por elementos finitos e na experiência do projetista. Todavia, como o usuário terá muitas dúvidas quanto ao melhor modelo as ser usado, ele pode superpor no modelo inicial todas as soluções que achar possíveis. A função do sistema é escolher entre todas as soluções admissíveis (aquelas que respeitam equilíbrio e resistência) a que for mais econômica. Define-se como solução mais econômica aquela que corresponder ao menor consumo de armadura. A formulação do problema conduz a um problema de Programação Linear (PL) e o programa LINDO [2] é utilizado para resolvê-lo. A visualização do modelo é também parte do sistema desenvolvido e é feito a partir dos resultados gerados pelo programa LINDO. O trabalho se divide basicamente em três partes. Na primeira parte é feita uma revisão sobre modelos de bielas e tirantes e sobre fundamentos de programação linear. A segunda parte consta da formulação do problema e por ultimo os resultados do programa são comparados com aqueles fornecidos pela literatura. A ferramenta desenvolvida neste trabalho pode ser de grande utilidade para a definição de modelos de bielas e tirantes mas ela nem de longe tem a pretensão de substituir a experiência, a intuição e o bom senso do engenheiro de projeto que deve sempre prevalecer, inclusive na análise crítica dos resultados do programa. / [en] The present work has the aim of developing a formulation and a computational system, which should be used by the structural engineers an auxiliary tool in the definition of strut and tie models of bidimensional structures. In the developed system the user can define initially a very complex system with a great number of bars and many different directions. This initial solution can be based on both the stress results of a linear elastic finite element analysis and the experience of the engineer. As the user may be in doubt about the best model to be used he may superimpose in the initial model all the allowable solutions he may find. The aim of the system is to choose among all the allowable solution (these ones wich satisfy equilibrium and strenght) the more economical one. As more economical solution is considered the one which correspond to the least consumption of reinforcement. The problem formulation leads to a linear programming problem (LP) and the LINDO program [2] is used for solving it. The visualization of the model and results is also part of the system. It is based on the data generated by the LINDO program. The work is basicaly divided in three parts. In the first one a revision about strut and tie models as well as the basic knowledge in linear programming is presented. In the second part the problem formulation is shown and the last the results of the program are compared with those old the technical literature. The computational tool presented here may be of great utility in the definition of strut and tie models, but it has not by far the intention and the feeling of the engineers which shall ever prevail, inclusive in the critic analysis of the program results. / [es] El presente trabajo tiene como objetivo desarrollar un sistema que sirva como herramienta auxiliar para el ingeniero de proyecto en la definición de modelos de bielas y tirantes de extructuras bidimensionales. En el sistema desarrollado, el usuario puede partir de un modelo complejo, con un número grande de barras en varias direcciones. Esa solución inicial puede basearse en los resultados en tensión de un análisis lineal elástica por elementos finitos y en la experiencia del proyectista. Como el usuario tendrá muchas dudas respecto al mejor modelo as ser usado, él puede superponer en el modelo inicial todas las soluciones que juzgue posibles. La función del sistema es elegir entre todas las soluciones admisibles (aquellas que respetan equilibrio y resistencia) la que sea más económica. Se define como solución más económica aquella que correspdonda al menor consumo de armadura. La formulación del problema conduce a un problema de Programación Lineal (PL) que es resuelto con ayuda del programa LINDO [2]. La visualización del modelo también es parte del sistema desarrollado y parte de los resultados generados por el programa LINDO. El trabajo se divide básicamente en tres partes. En la primeira parte se realiza una revisión de los modelos de bielas y tirantes y de los fundamentos de programación lineal. La segunda parte contiene la formulación del problema y la presentación de los resultados del programa se comparan con los reportados en la literatura. La herramienta desarrollada en este trabajo puede ser de gran utilidad para la definición de modelos de bielas y tirantes pero no tiene la pretensión de substituir la experiencia, la intuición y el buen senso del ingeniero de proyecto que debe siempre prevalecer, incluso en el análisis crítico de los resultados del programa.

An emerging trend in computing is to use distributed heterogeneous computing (HC) systems to execute a set of tasks. Cluster computer systems, grids, and Desktop Grids are three popular kinds of HC systems. An important component of an HC system is its resource management system (RMS). The main responsibility of an RMS is assigning resources to tasks in order to satisfy certain performance requirements. For cluster computer systems, we propose a new mapping heuristic which requires less state information than current heuristics. For Desktop Grids, we propose a new scheduling policy that exploits knowledge of the effective computing power delivered by the machines and the distribution of their fault times in order to improve performance. Finally, for grids, we propose a new decentralized load balancing policy which dramatically cuts down the communication overhead incurred in state information update. The proposed resource management policies utilize the solution to a linear programming problem (LP) which maximizes the system capacity. Our simulation experiments show that these policies perform very competitively, especially in highly heterogeneous systems. / Thesis / Doctor of Philosophy (PhD)

I investigated the effect of climate change in one of the few expanding markets in the world in Brazil. In the last decades the demand for pulpwood increased from zero to 11 million tons, leading to an expansion of 700 thousand hectares of Eucalyptus plantation. In 2024, a new mill will start operating increasing the consumption of wood fiber by 8.2 million tons (+49%). I used mathematical programming to investigate how different scenarios of productivity and land will affect the market. My results showed that around 946,000 hectares (+124%) of additional timberland will be necessary in the upcoming decades. The first impact will be an increase in the production cost of around US$550 million (+16.4%) in 25 years. If the rate of land cover change remains constant, the broad development of the local market would be constrained, discouraging future investments, and reducing potential positive externalities in the region.

京都大学 / 新制・課程博士 / 博士(情報学) / 甲第24938号 / 情博第849号 / 新制||情||142(附属図書館) / 京都大学大学院情報学研究科数理工学専攻 / (主査)准教授 原口 和也, 教授 山下 信雄, 教授 阿久津 達也 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

Ökad mänsklig aktivitet och global folkmängd har lett till ökat energibehov där största andelen av energin är från fossila bränslen. Vätgas ses som en effektiv energibärare som kan produceras med förnyelsebara resurser och ersätta fossila bränslen. Vätgas producerad med förnyelsebara resurser är dock fortfarande dyrt jämfört med produktion från fossila bränslen.Det finns få tidigare studier som har studerat utnyttjandet av spillvärme från vätgasanläggningar och vilken påverkan det kan ha på den ekonomiska lönsamheten.Den här studien som sker i samarbete med Research Institutes of Sweden (RISE), syftar till attoptimera en vätgasanläggning, med ett på förhand definierat vätgasbehov, utifrån ett ekonomiskt perspektiv och se om en sektorkoppling till ett fjärrvärmenät kan förbättra det ekonomiska resultatet. Sektorkoppling innebär i den här studien att spillvärme skickas från vätgasanläggningen in till fjärrvärmenätet. Systemet som studeras består utav en elektrolysör, en bränslecell och ett vätgaslager. Systemet är beskrivet så att i första hand producera en förutbestämd mängd vätgas som ska användas till fordonsbränsle. Därutöver kan systemet utnyttjas för elenergilagring via vätgaslagring, om det visar sig ekonomiskt fördelaktigt. Målen är att optimera fram ett pris för vad vätgasen behöver kosta för att anläggningen ska nå break-even, dimensionera elektrolys, vätgaslager och bränslecell utefter optimal drift ochstudera vad som påverkar den optimala driften av anläggningen.Metoden som används för att optimera vätgasanläggningen är linjärprogrammering i programmet MATLAB. I optimeringen ska den lägsta möjliga nettokostnaden (eventuella högsta nettointäkten) för systemet beräknas. Alla kostnader och intäkter beskrivas linjärt, sedan definieras alla variabler som påverkar kostnader och intäkter. Variablerna kommer i sin tur att bero på villkor, samt linjära likheter och olikheter som begränsar hur små eller stora värden de får anta för att påverka resultatet. Optimeringen sker över en tidsperiod på 1 år och med tidssteget två timmar. Historiskt elpris för 2021 SE3 användes i studien. Variationen i elpriset är svår att mäta, därför skapades även ett eget elpris som inte är så komplext för att närmare kunna studera variationens påverkan.Resultaten visar att vätgasen ska säljas för ungefär 36,6 kr/kg för att nå break-even när priset optimeras och spillvärmen inte utnyttjas. Utnyttjandet av spillvärme kan sänka priset med 1,6 - 2,5 kr/kg beroende på om spillvärmen utnyttjas delar av eller hela året. Elektrolysör och vätgaslager bör dimensioneras beroende på ett fördefinierat vätgasbehov, enligt dimensionsfaktorerna 4,9 [MW/(ton vätgas/dag)] respektive 1,4 [ton/(ton vätgas/dag)].Värmeeffekten som kan utnyttjas från vätgasanläggningen kan dimensioneras beroende på ett fördefinierat vätgasbehov, enligt dimensionsfaktorn 0,9 [MW/(ton vätgas/dag)]. Resultatet visar trender på att elpriset påverkar storleken på elektrolysör, vätgaslager och även vilka timmar som elektrolysen är aktiv. Att dimensionera anläggningen som funktion av elpriset har visat sig vara en komplex fråga, och fler studier behövs inom det området. En trend visar att utnyttjandet av spillvärme har en mindre påverkan på vätgaspriset vid högre verkningsgrad på elektrolysen, och ökad påverkan vid lägre verkningsgrad. Spillvärme som ersätter värme från fjärrvärmepanna minskar utsläppen med ungefär 38 kg CO2e/MWh beror på fjärrvärmeproducentens befintliga pannor/bränslen. För investerarna betyder resultaten en riktlinje för priset på vätgasen, uppskattad dimensionering av anläggningen, att det i praktiken aldrig är lönsamt med bränslecell och en inblick i det ekonomiska värdet av spillvärmen. För fjärrvärmebolaget betyder resultaten minskade utsläpp av koldioxidekvivalenter, en extravärmekälla och möjligen minskade kostnader. För energisystemet i stort betyder resultaten att den höga variationen på elpriset kan utnyttjas för att producera vätgas, att det gröna vätgaspriset kan sjunka från dagens nivåer och att spillvärmen kan bidra till ett effektivare energisystem. / Increased human activity and global population have led to an increased energy demand, with most of the energy coming from fossil fuels. Hydrogen is seen as an efficient energy carrier that can be produced using renewable resources and replace fossil fuels. However, hydrogen produced from renewable resources is still expensive compared to production from fossil fuels. There have been few previous studies that have examined the utilization of waste heat from hydrogen facilities and the impact it can have on economic profitability.This study, conducted in collaboration with Research Institutes of Sweden (RISE), aims to optimize a hydrogen production facility, with a predefined hydrogen demand, from an economic perspective and assess whether a sector coupling to a district heating network can improve the economic outcome. In this study, sector coupling means that waste heat is sent from the hydrogen facility to the district heating network. The hydrogen plant consists of an electrolyzer, a fuel cell, and a hydrogen storage. The system is designed to primarily produce a predetermined amount of hydrogen to be used as vehicle fuel. Additionally, the system can be used for electrical energy storage through hydrogen storage if it proves to be economically advantageous. The objectives are to optimize the price at which hydrogen needs to be sold for the facility to break even, sizing the electrolyzer, hydrogen storage, and fuel cell based on optimal operation, and study the factors that affect the optimal operation of the facility.The method used to optimize the hydrogen facility is linear programming in the MATLAB program. The optimization aims to calculate the lowest possible net cost (or highest net income) for the system. All costs and revenues are described linearly, and then all variables that affect costs and revenues are defined. These variables, in turn, depend on conditions as well as linear equalities and inequalities that restrict the values, they can take to influence the result. The optimization is carried out over a period of 1 year with a time step of two hours. Historical electricity prices for 2021 SE3 were used in the study. The variation in electricity prices is difficult to measure, so a simplified electricity price was also created to study the impact of the variation more closely.The results show that the hydrogen should be sold for approximately 36.6 SEK/kg to break even when the price is optimized, and waste heat is not utilized. The utilization of waste heat can reduce the price by 1.6-2.5 SEK/kg depending on whether the waste heat is utilized for parts or the entirety of the year. The electrolyzer and hydrogen storage should be dimensioned based on a predefined hydrogen demand, according to the factors of 4.9 MW/(ton of hydrogen/day) and 1.4 ton/(ton of hydrogen/day), respectively. The heat output that can be utilized from the hydrogen facility can be dimensioned based on a predefined hydrogen demand, according to the factor of 0.9 MW/(ton of hydrogen/day). The results show trends indicating that the electricity price affects the size of the electrolyzer, hydrogen storage, andthe hours during which the electrolyzer is active. Dimensioning the facility as a function of the electricity price has proven to be a complex question, and further studies are needed in that area. One trend indicates that the utilization of waste heat has a smaller impact on the hydrogen price at higher electrolyzer efficiency and a greater impact at lower efficiency. Waste heat that replaces heat from a district heating boiler reduces emissions by approximately 38 kg CO2e/MWh, depending on the existing boilers/fuels of the district heating producer. For investors, the results provide guidelines for the price of hydrogen, estimated dimensioning of the facility, the practical non-profitability of fuel cells, and insights into the economic value of waste heat. For the district heating company, the results mean reduced emissions of carbon dioxide equivalents, an additional heat source, and possibly reduced costs For the Swedish energy system, the results imply that the high variation in electricity prices can be utilized to produce hydrogen, leading to a potential decrease in the price of green hydrogen compared to current levels. Additionally, the utilization of waste heat can contribute to a more efficient energy system. / HyCoGen

The fixed income market is not as exploited as other markets and has a more complex structure compared with the equity market. On the other hand, it has been seen that demand for research for the fixed income market has increased, which in turn has created greater interest in studying the characteristics of holdings in the market. This work studies whether it is possible to replicate indices through requirements for credit rating, sectors and mathematical key figures such as Duration, convexity, duration time spread (DTS) and option adjusted spread (OAS). Replication is made through linear programming in the program Python. By implementing lasso regression, this study examines whether it is possible to exceed the return by reducing the requirements for key figures that are not selected efter selection of variables in the regression. The investment company Alfred Berg has provided relevant data for this report. The data consists of information on all assets included in the index EUR Investment grade (ER00) over the period 2017-2021. The result of the replication follows the index returns, with small deviations, and the lasso regression selects the key figures DTS and OAS in its model. It is difficult to excess index return by focusing only on the key figures DTS and OAS. Analysis of other key figures and variables selected by the lasso regression can possibly create better results, as a suggestion for further work. / Räntemarknaden är inte lika exploaterad som andra marknader och har en mer komplex struktur jämfört med aktiemarknaden. Däremot har man sett att efterfrågan på forskning för räntemarknaden har ökat, vilket i sin tur skapat ett större intresse att studera egenskaperna av innehaven på marknaden. Detta arbete studerar om det går att replikera index genom krav på credit rating, sektor och matematiska nyckeltal som Duration, convexity, duration times spread (DTS) och option adjusted spread (OAS). Replikeringen sker genom linjär programmering i programmet Python. Genom att implementera Lasso regression undersöker detta arbete även om det går att överträffa vakastningen genom att minska kraven på nyckeltal som inte väljts ut efter urval av variabler i regressionen. Investmentbolaget Alfred Berg har bidragit med data för denna rapport. Datan består av information om alla tillgångar som ingår i indexet EUR Investment Grade (ER00) under perioden 2017-2021. Resultatet visar att replikeringen av index är möjlig, med små avvikelser, och lasso regressionen väljer nyckeltalen DTS och OAS i sin modell. Det är svårt att överträffa index genom att endast fokusera på nyckeltalen DTS och OAS. Analys av andra nyckeltal och variabler som väljs ut av lasso regressionen kan skapa ett bättre resultat.

No description available.