Browsing by Author "Dr. Subhashis Ghosal, Committee Member"

Filter results by typing the first few letters
Now showing 1 - 3 of 3
  • No Thumbnail Available
    An Algorithm for Determining Optimal Resource Allocation in Stochastic Activity Networks
    (2008-04-24) Rudolph, Adam J; Dr. Salah E. Elmaghraby, Committee Chair; Dr. Subhashis Ghosal, Committee Member; Dr Julie Ivy, Committee Member
  • No Thumbnail Available
    Optimal Dynamic Resource Allocation in Activity Networks
    (2006-08-17) Ramachandra, Girish; Dr. Abdelhakim Artiba, Committee Member; Dr. James R. Wilson, Committee Member; Dr. Subhashis Ghosal, Committee Member; Dr. Matthias F. (Matt) Stallmann, Committee Member; Dr. Salah E. Elmaghraby, Committee Chair
    We treat the problem of optimally allocating resources of limited availability under uncertainty to the various activities of a project to minimize a certain economic objective composed of resource cost and tardiness cost. Traditional project scheduling methods assume that the uncertainty resides in the duration of the activities. Our research assumes that the work content (or 'effort') of an activity is the source of uncertainty and the duration is the result of the amount of resource allocated to the activity, which then becomes the decision variable. The functional relationship between the work content (w), the resource allocation (x), and the duration of the activity (y) is arbitrary, though we assume that the relationship obeys the 'power law.' In other words, y = f(w,xˆ[gamma]), where the exponent, gamma, is some constant. As preliminary, we first treat the problem assuming that the work content is known deterministically. We develop two new models, a nonlinear programming model, which can be used when resource availabilities are continuous, and an integer program that handles the case when resource availabilities are discrete. When the work content is known only in probability, we first treat the special case when the work content is exponentially distributed. This results in a continuous-time Markov chain with a single absorbing state. We establish convexity of the cost function and develop a Policy Iteration--like approach that achieves the optimum in a finite number of steps. In case of arbitrary probability distribution of the work content, we develop a simulation-cum optimization method that incorporates sampling optimization and variance reduction techniques, and which can be used for the purposes of estimation of total project cost, resource consumption levels, etc.
  • No Thumbnail Available
    P-Coffee: A New Divide-and-conquer Method for Multiple Sequence Alignment
    (2005-01-19) Choi, Kwangbom; Dr. Subhashis Ghosal, Committee Member; Dr. Dennis R. Bahler, Committee Chair; Dr. Jon Doyle, Committee Member
    We describe a new divide-and-conquer method, P-Coffee, for alignment of multiple sequences. P-Coffee first identifies candidate alignment columns using a position-specific substitution matrix (the T-Coffee extended library), tests those columns, and accepts only qualified ones. Accepted columns do not only constitute a final alignment solution, but also divide a given sequence set into partitions. The same procedure is recursively applied to each partition until all the alignment columns are collected. In P-Coffee, we minimized the source of bias by aligning all the sequences simultaneously without requiring any heuristic function to optmize, phylogenetic tree, nor gap cost scheme. In this research, we show the performance of our approach by comparing our results with that of T-Coffee using the 144 test sets provided in BAliBASE v1.0. P-Coffee outperformed T-Coffee in accuracy especially for more complicated test sets.