A randomized concave programming method for choice network revenue management

Models incorporating more realistic models of customer behavior, as customers choosing froman offer set, have recently become popular in assortment optimization and revenue management.The dynamic program for these models is intractable and approximated by a deterministiclinear program called the CDLP which has an exponential number of columns. However, whenthe segment consideration sets overlap, the CDLP is difficult to solve. Column generationhas been proposed but finding an entering column has been shown to be NP-hard. In thispaper we propose a new approach called SDCP to solving CDLP based on segments and theirconsideration sets. SDCP is a relaxation of CDLP and hence forms a looser upper bound onthe dynamic program but coincides with CDLP for the case of non-overlapping segments. Ifthe number of elements in a consideration set for a segment is not very large (SDCP) can beapplied to any discrete-choice model of consumer behavior. We tighten the SDCP bound by(i) simulations, called the randomized concave programming (RCP) method, and (ii) by addingcuts to a recent compact formulation of the problem for a latent multinomial-choice model ofdemand (SBLP+). This latter approach turns out to be very effective, essentially obtainingCDLP value, and excellent revenue performance in simulations, even for overlapping segments.By formulating the problem as a separation problem, we give insight into why CDLP is easyfor the MNL with non-overlapping considerations sets and why generalizations of MNL posedifficulties. We perform numerical simulations to determine the revenue performance of all themethods on reference data sets in the literature.

A new compact linear programming formulation for choice network revenue management

The choice network revenue management model incorporates customer purchase behavioras a function of the offered products, and is the appropriate model for airline and hotel networkrevenue management, dynamic sales of bundles, and dynamic assortment optimization.The optimization problem is a stochastic dynamic program and is intractable. A certainty-equivalencerelaxation of the dynamic program, called the choice deterministic linear program(CDLP) is usually used to generate dyamic controls. Recently, a compact linear programmingformulation of this linear program was given for the multi-segment multinomial-logit (MNL)model of customer choice with non-overlapping consideration sets. Our objective is to obtaina tighter bound than this formulation while retaining the appealing properties of a compactlinear programming representation. To this end, it is natural to consider the affine relaxationof the dynamic program. We first show that the affine relaxation is NP-complete even for asingle-segment MNL model. Nevertheless, by analyzing the affine relaxation we derive a newcompact linear program that approximates the dynamic programming value function betterthan CDLP, provably between the CDLP value and the affine relaxation, and often comingclose to the latter in our numerical experiments. When the segment consideration sets overlap,we show that some strong equalities called product cuts developed for the CDLP remain validfor our new formulation. Finally we perform extensive numerical comparisons on the variousbounds to evaluate their performance.

On the throughput-WIP trade-off in queueing systems, diminishing returns and the threshold property: A linear programming approach

We present a new unifying framework for investigating throughput-WIP(Work-in-Process) optimal control problems in queueing systems,based on reformulating them as linear programming (LP) problems withspecial structure: We show that if a throughput-WIP performance pairin a stochastic system satisfies the Threshold Property we introducein this paper, then we can reformulate the problem of optimizing alinear objective of throughput-WIP performance as a (semi-infinite)LP problem over a polygon with special structure (a thresholdpolygon). The strong structural properties of such polygones explainthe optimality of threshold policies for optimizing linearperformance objectives: their vertices correspond to the performancepairs of threshold policies. We analyze in this framework theversatile input-output queueing intensity control model introduced byChen and Yao (1990), obtaining a variety of new results, including (a)an exact reformulation of the control problem as an LP problem over athreshold polygon; (b) an analytical characterization of the Min WIPfunction (giving the minimum WIP level required to attain a targetthroughput level); (c) an LP Value Decomposition Theorem that relatesthe objective value under an arbitrary policy with that of a giventhreshold policy (thus revealing the LP interpretation of Chen andYao's optimality conditions); (d) diminishing returns and invarianceproperties of throughput-WIP performance, which underlie thresholdoptimality; (e) a unified treatment of the time-discounted andtime-average cases.

Implementing TURF analysis through binary linear programming

This paper introduces the approach of using Total Unduplicated Reach and Frequency analysis (TURF) to design a product line through a binary linear programming model. This improves the efficiency of the search for the solution to the problem compared to the algorithms that have been used to date. The results obtained through our exact algorithm are presented, and this method shows to be extremely efficient both in obtaining optimal solutions and in computing time for very large instances of the problem at hand. Furthermore, the proposed technique enables the model to be improved in order to overcome the main drawbacks presented by TURF analysis in practice.

Restless bandits, linear programming relaxations and a primal-dual index heuristic

We develop a mathematical programming approach for the classicalPSPACE - hard restless bandit problem in stochastic optimization.We introduce a hierarchy of n (where n is the number of bandits)increasingly stronger linear programming relaxations, the lastof which is exact and corresponds to the (exponential size)formulation of the problem as a Markov decision chain, while theother relaxations provide bounds and are efficiently computed. Wealso propose a priority-index heuristic scheduling policy fromthe solution to the first-order relaxation, where the indices aredefined in terms of optimal dual variables. In this way wepropose a policy and a suboptimality guarantee. We report resultsof computational experiments that suggest that the proposedheuristic policy is nearly optimal. Moreover, the second-orderrelaxation is found to provide strong bounds on the optimalvalue.

Control d'una bobinadora automàtica

Un dels problemes típics de regulació en el camp de l’automatització industrial és el control de velocitat lineal d’entrada del fil a les bobines, ja que com més gruix acumulem a igual velocitat de rotació de la bobina s’augmenta notablement la velocitat lineal d’entrada del fil, aquest desajust s’ha de poder compensar de forma automàtica per aconseguir una velocitat d’entrada constant. Aquest problema de regulació de velocitats és molt freqüent i de difícil control a la indústria on intervé el bobinat d’algun tipus de material com cablejat, fil, paper, làmines de planxa, tubs, etc... Els dos reptes i objectius principals són, primer, la regulació de la velocitat de rotació de la bobina per aconseguir una velocitat lineal del fil d’entrada, i segon, mitjançant el guiatge de l’alimentació de fil a la bobina, aconseguir un repartiment uniforme de cada capa de fil. El desenvolupament consisteix amb l’automatització i control d’una bobinadora automàtica mitjançant la configuració i programació de PLC’s, servomotors i encoders. Finalment es farà el muntatge pràctic sobre una bancada per verificar i simular el seu correcte funcionament que ha de donar solució a aquests problemes de regulació de velocitats. Com a conclusions finals s’han aconseguit els objectius i una metodologia per fer una regulació de velocitats de rotació per bobines, amb accionaments de servomotors amb polsos, i a nivell de coneixements he aconseguit dominar les aplicacions d’aquest tipus d’accionaments aplicats a construccions mecàniques.

Programming and analysing measurement experiments using the PicoCricket system

The aim of this project is to get used to another kind of programming. Since now, I used very complex programming languages to develop applications or even to program microcontrollers, but PicoCricket system is the evidence that we don’t need so complex development tools to get functional devices. PicoCricket system is the clear example of simple programming to make devices work the way we programmed it. There’s an easy but effective way to program small, devices just saying what we want them to do. We cannot do complex algorithms and mathematical operations but we can program them in a short time. Nowadays, the easier and faster we produce, the more we earn. So the tendency is to develop fast, cheap and easy, and PicoCricket system can do it.

Report to the Governor and General Assembly on the Status of Out-of School Arts Programming for Youth, December 2006

In the 2006 Iowa General Assembly, House File 2797 called for a study on the status of afterschool arts programs and appropriated $5,000 for the study. In accordance with the legislation, the Iowa Arts Council, who received the charge, contracted with the Iowa Afterschool Alliance to form a Resource Group of out-of-school arts providers and experts to develop and oversee the study, review its results, and make recommendations for the expansion of arts programs that operate outside the normal school day. As a part of its charge in HF 2797, the Iowa Arts Council also documented a sampling of out-of-school arts programs statewide. Five are featured in this report.

Improving Offender Programming, Data Download, Iowa Department of Corrections, June 2012

In 2008, the Iowa Department of Corrections embarked on a comprehensive audit of all institutional andcommunity corrections offender programs that receive corrections funding. The audit focused on the degree towhich each program adheres to evidence-based practices principles (what works best). The audit areas are listedbelow along with some examples of the questions asked about each program:

New Violence Risk Assessment Aids Parole and Offender Programming Decisions, Data Download, Department of Corrections

In recent years, the Department of Corrections has made major strides in assessing offenders’ risk to reoffend, particularly in measuring changes in that risk over time. Earlier this year, the DOC worked with the Board of Parole to develop a risk assessment that focuses on assessing offenders’ risk to commit violent crimes.

A bio-inspired agent-based programming environment for pervasive platforms

Abstract in English : Ubiquitous Computing is the emerging trend in computing systems. Based on this observation this thesis proposes an analysis of the hardware and environmental constraints that rule pervasive platforms. These constraints have a strong impact on the programming of such platforms. Therefore solutions are proposed to facilitate this programming both at the platform and node levels. The first contribution presented in this document proposes a combination of agentoriented programming with the principles of bio-inspiration (Phylogenesys, Ontogenesys and Epigenesys) to program pervasive platforms such as the PERvasive computing framework for modeling comPLEX virtually Unbounded Systems platform. The second contribution proposes a method to program efficiently parallelizable applications on each computing node of this platform. Résumé en Français : Basée sur le constat que les calculs ubiquitaires vont devenir le paradigme de programmation dans les années à venir, cette thèse propose une analyse des contraintes matérielles et environnementale auxquelles sont soumises les plateformes pervasives. Ces contraintes ayant un impact fort sur la programmation des plateformes. Des solutions sont donc proposées pour faciliter cette programmation tant au niveau de l'ensemble des noeuds qu'au niveau de chacun des noeuds de la plateforme. La première contribution présentée dans ce document propose d'utiliser une alliance de programmation orientée agent avec les grands principes de la bio-inspiration (Phylogénèse, Ontogénèse et Épigénèse). Ceci pour répondres aux contraintes de programmation de plateformes pervasives comme la plateforme PERvasive computing framework for modeling comPLEX virtually Unbounded Systems . La seconde contribution propose quant à elle une méthode permettant de programmer efficacement des applications parallélisable sur chaque noeud de calcul de la plateforme

La programmmation foetale de la dysfonction vasculaire pulmonaire chez la souris : rôle des mécanismes épigénétiques = Fetal programming of pulmonary vascular dysfunction in mice : role of epigenetic mechanisms

Rapport de synthèseDes événements pathologiques survenant pendant la période foetale prédisposent la descendance aux maladies cardiovasculaires systémiques. Il existe peu de connaissances au sujet de la circulation pulmonaire et encore moins quant aux mécanismes sous-jacents. La sous-alimentation maternelle pendant la grossesse peut représenter un modèle d'investigation de ces mécanismes, parce que chez l'animal et l'homme elle est associée à une dysfonction vasculaire systémique chez la progéniture. Chez le rat, la diète restrictive pendant la grossesse induit une augmentation du stress oxydatif dans le placenta. Les dérivés de l'oxygène sont connus pour induire des altérations épigénétiques et peuvent traverser la barrière placentaire. Nous avons dès lors spéculé que chez la souris la diète restrictive pendant la grossesse induit une dysfonction vasculaire pulmonaire chez sa progéniture qui serait liée à un mécanisme épigénétique.Pour tester cette hypothèse, nous avons examiné la fonction vasculaire pulmonaire et la méthylation de l'ADN pulmonaire à la fin de 2 semaines d'exposition à l'hypoxie chez la progéniture de souris soumises à une diète restrictive pendant la grossesse et des souris contrôles. Nous avons trouvé que la vasodilatation endothélium-dépendante de l'artère pulmonaire in vitro était défectueuse, et que l'hypertension pulmonaire et l'hypertrophie ventriculaire droite induites par l'hypoxie in vivo étaient exagérées chez la progéniture de souris soumises à une diète restrictive pendant la grossesse. Cette dysfonction vasculaire pulmonaire était associée avec une altération de la méthylation de l'ADN pulmonaire. L'administration d'inhibiteurs de la déacétylase des histones, le Butyrate et la Trichostatine-A à la progéniture de souris soumises à une diète restrictive pendant la grossesse a normalisé la méthylation de l'ADN et la fonction vasculaire pulmonaire. Finalement, l'administration du nitroxyde Tempol aux mères durant la diète restrictive pendant la grossesse a prévenu la dysfonction vasculaire et la dysméthylation chez la progéniture.Ces découvertes démontrent que chez la souris la sous-alimentation pendant la gestation induit une dysfonction vasculaire chez la progéniture qui est causée par un mécanisme épigénétique. Il est possible qu'un mécanisme similaire soit impliqué dans la programmation foetale de la dysfonction vasculaire chez les humains.

Programming of marginal zone B-cell fate by basic Kruppel-like factor (BKLF/KLF3).

Splenic marginal zone (MZ) B cells are a lineage distinct from follicular and peritoneal B1 B cells. They are located next to the marginal sinus where blood is released. Here they pick up antigens and shuttle the load onto follicular dendritic cells inside the follicle. On activation, MZ B cells rapidly differentiate into plasmablasts secreting antibodies, thereby mediating humoral immune responses against blood-borne type 2 T-independent antigens. As Krüppel-like factors are implicated in cell differentiation/function in various tissues, we studied the function of basic Krüppel-like factor (BKLF/KLF3) in B cells. Whereas B-cell development in the bone marrow of KLF3-transgenic mice was unaffected, MZ B-cell numbers in spleen were increased considerably. As revealed in chimeric mice, this occurred cell autonomously, increasing both MZ and peritoneal B1 B-cell subsets. Comparing KLF3-transgenic and nontransgenic follicular B cells by RNA-microarray revealed that KLF3 regulates a subset of genes that was similarly up-regulated/down-regulated on normal MZ B-cell differentiation. Indeed, KLF3 expression overcame the lack of MZ B cells caused by different genetic alterations, such as CD19-deficiency or blockade of B-cell activating factor-receptor signaling, indicating that KLF3 may complement alternative nuclear factor-κB signaling. Thus, KLF3 is a driving force toward MZ B-cell maturation.

Maximum likelihood Linear Programming Data Fusion for Speaker Recognition

Biometric system performance can be improved by means of data fusion. Several kinds of information can be fused in order to obtain a more accurate classification (identification or verification) of an input sample. In this paper we present a method for computing the weights in a weighted sum fusion for score combinations, by means of a likelihood model. The maximum likelihood estimation is set as a linear programming problem. The scores are derived from a GMM classifier working on a different feature extractor. Our experimental results assesed the robustness of the system in front a changes on time (different sessions) and robustness in front a change of microphone. The improvements obtained were significantly better (error bars of two standard deviations) than a uniform weighted sum or a uniform weighted product or the best single classifier. The proposed method scales computationaly with the number of scores to be fussioned as the simplex method for linear programming.