Using mixed integer programming to schedule experiments on the Texas A&M University nuclear science center reactor /

"skip-factor'' strategy prioritized variables in such

Bibliographic Details
Main Author: Drees, Lawrence David, 1969-
Format: Thesis eBook
Language:English
Published: [Place of publication not identified] : [publisher not identified] ; 1999.
Subjects:
Online Access:Link to OAKTrust copy
Description
Summary:"skip-factor'' strategy prioritized variables in such
a way that some tardiness is allowed initially in
and tested them to determine their effectiveness in
average run-time of less than one minute.
bowshot. Several practical aspects prevent it from
by the Nuclear Science Center in early 1997. A11
compared with the CPLEX default settings for solving
conforming to existing models in the literature. Some
Deadlines must be observed as constraints. We
defined in such a way to minimize - to the fullest
developed four optimizing strategies and one heuristic
especially well in terms of the number of test
extent possible - the number of integer variables in
have deadlines, while others have only due- dates.
higher the priority, the earlier the search branches
highest of all strategies tested) and did so with an
jobs may be preempted, while others may not. Some jobs
MIPs. One particular optimizing strategy worked
on that variable. The heuristic starts with the
optimizing strategies and the heuristic were tested
order to arrive at an initial feasible solution
problem is modeled as a time-indexed Mixed Integer
problems solved to optimality and the run-time. This
problems within a predetermined time limit (the
problems, each of which represented a ten-job
Program. Constraints were formed and variables were
quickly. This strategy optimized 80% of the test
scheduling problem. The parameters for the jobs were
solution to the LP relaxation and fixes all binary
solving the NSC scheduling problem. One optimizing
Some jobs require tooling, while others do not. This
strategies and the heuristic were tested on 25 test
strategies deal with various ways of prioritizing the
strategy exploits the Multiple Choice Sets structure
taken from a database of 103 jobs that were processed
The NSC scheduling problem involves a two-station graphics.
the problem. The objective is to minimize the total
using CPLEX 6.0. The results of these tests were
variable k, which is between 0 and 1. These optimizing
variables to 1 that are greater than or equal to some
variables to bias the branch-and-bound search. The
weighted tardiness of the jobs with due-dates.
within the constraints. The other three optimizing
Item Description:"Major subject: Industrial Engineering".
Vita.
Physical Description:xi, 81 leaves : illustrations ; 28 cm.
Also available online.
Bibliography:Includes bibliographical references (leaves 54-55).