Browsing by Author "Stephen Roberts, Committee Member"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • No Thumbnail Available
    Exact and Heuristic Algorithms for the q-mode Problem
    (2005-05-18) Kulkarni, Girish; Stephen Roberts, Committee Member; Shu-Cherng Fang, Committee Member; Carla Savage, Committee Member; Yahya Fathi, Committee Chair
    In this dissertation we focus on the development of exact and inexact (i.e., heuristic) algorithms for the q-mode problem. The exact algorithms are based on integer programming models for the q-mode problem. We discuss the theoretical properties of an existing IP model and propose several enhancements. We also propose a new IP model for the problem and investigate these models through a comprehensive computational experiment. The experiment reveals that, in practice, the IP models are more effective for instances with strong natural clusters but less effective for instances containing weak natural clusters. We also propose exact algorithms based on the Benders decomposition for one of the IP models. The heuristic algorithm that we propose for the q-mode problem is a local improvement algorithm that is based on a very large scale neighborhood structure. We evaluate the algorithm through a computational experiment and empirically demonstrate its effectiveness.
  • No Thumbnail Available
    An Industrial Application of Time Series Forecasting of Lumber Demand.
    (2003-04-30) Alexander, Kristy Laurelle; Robert Handfield, Committee Chair; Stephen Roberts, Committee Member; Xuili Chao, Committee Member
    Forecasting lumber demand is vital for operational purposes in the Distribution Centers of Home Improvement retail chains. This paper describes econometric time series analyses applied to specific lumber skus from the largest Home Improvement chain in the United States. We propose simple univariate smoothing models and examine the causal relationship between temperature, housing starts, price and lumber demand. We find that complicated ARIMA models are not necessary; simple smoothing models are more appropriate. The results indicate that monthly seasonal models fit better that weekly moving average models and that even though the Point-of-Sale time series and Housing Starts time series show similar trends, the relationship is not strong enough for housing starts to be used as a short-term predictor. Also, the local maxima of the Point-of-Sale time series trends in the Spring, Summer and Fall result in low correlations between that series and the average monthly temperature or price series. So, temperature and price cannot be used as short-term predictors either.