W e addressed the problem of developing a model to simulate at a high level of detail the movements of over 6,000 drivers for Schneider National, the largest truckload motor carrier in the United States. The goal of the model was not to obtain a better solution but rather to closely match a number of operational statistics. In addition to the need to capture a wide range of operational issues, the model had to match the performance of a highly skilled group of dispatchers while also returning the marginal value of drivers domiciled at different locations. These requirements dictated that it was not enough to optimize at each point in time (something that could be easily handled by a simulation model) but also over time. The project required bringing together years of research in approximate dynamic programming, merging math programming with machine learning, to solve dynamic programs with extremely high-dimensional state variables. The result was a model that closely calibrated against real-world operations and produced accurate estimates of the marginal value of 300 different types of drivers.

We address the problem of determining optimal stepsizes for estimating parameters in the context of approximate dynamic programming. The sufficient conditions for convergence of the stepsize rules have been known for 50 years, but practical computational work tends to use formulas with parameters that have to be tuned for specific applications. The problem is that in most applications in dynamic programming, observations for estimating a value function typically come from a data series that can be initially highly transient. The degree of transience affects the choice of stepsize parameters that produce the fastest convergence. In addition, the degree of initial transience can vary widely among the value function parameters for the same dynamic program. This paper reviews the literature on deterministic and stochastic stepsize rules, and derives formulas for optimal stepsizes for minimizing estimation error. This formula assumes certain parameters are known, and an approximation is proposed for the case where the parameters are unknown. Experimental work shows that the approximation provides faster convergence than other popular formulas.

We address the problem of modeling long-term energy policy and investment decisions while retaining the important ability to capture fine-grained variations in intermittent energy and demand, as well as storage. In addition, we wish to capture sources of uncertainty such as future energy policies, climate, and technological advances, in addition to the variability as well as uncertainty in wind energy, demands, prices and rainfall. Accurately modeling the value of all investments such as wind and solar requires handling fine-grained temporal variability and uncertainty in wind and solar, as well as the use of storage. We propose a modeling and algorithmic strategy based on the framework of approximate dynamic programming (ADP) that can model these problems at hourly time increments over a multidecade horizon, while still capturing different types of uncertainty. This paper describes initial proof of concept experiments for an ADP-based model, called SMART, by describing the modeling and algorithmic strategy, and providing comparisons against a deterministic benchmark as well as initial experiments on stochastic datasets.

In this paper, the problem of clustering machines into cells and components into part-families with the consideration of ratio-level and ordinal-level data is dealt with. The ratio-level data is characterized by the use of workload information obtained both from per-unitprocess times and production quantity of components, and from machine capacity. In the case of ordinal-level data, we consider the sequence of operations for every component. These data sets are used in place of conventional binary data for arriving at clusters of cells and part-families. We propose a new approach to cell formation by viewing machines, and subsequently components, as 'points' in multi-dimensional space, with their coordinates defined by the corresponding elements in a Machine-Component Incidence Matrix (MCIM). An iterative algorithm that improves upon the seed solution is developed. The seed solution is obtained by formulating the given clustering problem as a Traveling Salesman Problem (TSP). The solutions yielded by the proposed clustering algorithm are found to be good and comparable to those reported in the literature.

Objective:The objective of this study was to identify the common etiological pathogens causing community acquired pneumonia (CAP) in our hospital and sensitivity patterns to the common antibiotics used.Materials and Methods:This study was undertaken in a 750 bedded multi-specialty referral hospital in Kerala catering to both urban and semi-urban populations. It is a prospective study of patients who attended the medical out-patient department and those admitted with a clinical diagnosis of CAP, during the year 2009. Data were collected based on detailed patient interview, clinical examination and laboratory investigations. The latter included sputum culture and sensitivity pattern. These were tabulated and percentage incidence of etiological pathogens calculated. The antimicrobial sensitivity pattern was also classified by percentage and expressed as bar diagram.Results:The study showed Streptococcus pneumoniae to be the most common etiological agent for CAP, in our hospital setting. The other organisms isolated in order of frequency were Klebsiella pneumoniae, Pseudomonas aeruginosa, Alpha hemolytic streptococci, Escherichia coli, Beta hemolytic streptococci and atypical coli. S. pneumoniae was most sensitive to linezolid, followed by amoxicillin-clavulanate (augmentin), cloxacillin and ceftriaxone. Overall, the common pathogens causing CAP showed highest sensitivity to amikacin, followed by ofloxacin, gentamycin, amoxicillin-clavulanate (augmentin), ceftriaxone and linezolid. The least sensitivity rates were shown to amoxicillin and cefoperazone.Conclusion:In a hospital setting, empirical management for cases of CAP is not advisable. The present study has shown S. pneumoniae as the most likely pathogen and either linezolid or amikacin as the most likely effective antimicrobial in cases of CAP, in our setting.