1. This is an interesting program by Paul A. Jensen at the University of Texas that tracks/forecasts hurricanes using operations research algorithms.  It’s historical data is old, but it’s still worth a look.

    The Hurricane program tracks and forecasts hurricanes in and around the Gulf of Mexico. It accepts data published by the National Hurricane Center (NHC). The tracking program is implemented using VBA modules in an Excel workbook, hurricane.xls.

    The original forecast models were developed in the late 1970s by Bill Lesso, Professor Emeritus of the University of Texas. They were based on using a Markov process to model the movement of a storm, i.e. the next position only depends on the current position. To develop the probability transition matrices, several hundred historic storms tracks dating back to 1886 were used. The result was a simple, fast computer model that could be run on the newly introduced PC’s. On contract the NHC was using several different models that could be described as ‘aerosol physics’ models consisting of several hundred partial differential equations that, at that time took several hours to run.

    (via INFORMS)

     
  2. Here’s an interesting paper by Gurvich et al. that’s in the current issue of Management Science.

    Full paper here.

     
  3. INFORMS 2007 Presentations

    I’m not going to INFORMS in Seattle this year, but I am a co-author on two presentations:

    1. “Forecasting and Planning Diagnostic Techniques for Service Operations” (first author, along with V. Buraparate and A. Dhawan; to be presented in the “Operations Management” session on Tuesday November 6 from 4:30 – 6:00pm)

      Abstract: The authors propose a systematic methodology to identify problems existing in the forecasting and planning processes for service operations. This procedure enables the identification of areas where improvement is needed and also helps aid the Operations Research Analyst in the selection of appropriate tool(s) to resolve these issues. This process has been deployed in several consumer electronics service operations (e.g., contact and repair centers) and examples will be shared.
    2. “Inventory Control for Automated Drug Dispensing Machines: A Service Level Policy” (corresponding author, along with J. Kobza, S. Dunagan, G. Heath, and S. Liman; Dr. John Kobza is presenting this work I did for my MS degree because INFORMS only allows each person to be listed as first author on one paper; to be presented in the “Inventory Management” session on Wednesday November 7 from 3:30 – 5:00pm)

      Abstract:The authors present a service level inventory policy for automated drug dispensing machines. Through simulation, it is shown that there could be a significant reduction in pharmaceutical shortages for a case study hospital. A pilot study was also performed to validate the simulation results. The model performed well compared to theoretical predictions and pre-implementation data.

    Finally, my MS research is being presented. Look for the presentation here after it’s given in Seattle. The paper will probably be posted here at some point as well.

     
Related Posts Plugin for WordPress, Blogger...