Scribd Logo
Upload

Welcome to Scribd!

  • Branch and Bound: Example 1

    Uploaded by

    Maria José Andrade Padilla
    9 views4 pages
    Branch and bound is a general method for solving discrete optimization problems that takes exponential time in the worst case but can be useful for small instances of hard problems. It requires specifying a lower bounding scheme, branching rule, and subproblem selection rule. For a minimization problem at a node, it obtains a lower bound, an optional upper bound and candidate solution, prunes the node if the lower bound exceeds the best upper bound, otherwise branches into child nodes. An example schedules jobs on a single machine to minimize maximum lateness using branch and bound by fixing jobs in the partial schedule at each level and using preemptive EDD to lower bound unscheduled jobs.

    Original Description:

    Original Title

    bb

    Copyright

    © © All Rights Reserved

    Available Formats

    DOC, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document
    Branch and bound is a general method for solving discrete optimization problems that takes exponential time in the worst case but can be useful for small instances of hard problems. It requires specifying a lower bounding scheme, branching rule, and subproblem selection rule. For a minimization problem at a node, it obtains a lower bound, an optional upper bound and candidate solution, prunes the node if the lower bound exceeds the best upper bound, otherwise branches into child nodes. An example schedules jobs on a single machine to minimize maximum lateness using branch and bound by fixing jobs in the partial schedule at each level and using preemptive EDD to lower bound unscheduled jobs.

    Copyright:

    © All Rights Reserved
    Download as DOC, PDF, TXT or read online from Scribd
    Flag for inappropriate content
    9 views4 pages

    Branch and Bound: Example 1

    Uploaded by

    Maria José Andrade Padilla
    Branch and bound is a general method for solving discrete optimization problems that takes exponential time in the worst case but can be useful for small instances of hard problems. It requires specifying a lower bounding scheme, branching rule, and subproblem selection rule. For a minimization problem at a node, it obtains a lower bound, an optional upper bound and candidate solution, prunes the node if the lower bound exceeds the best upper bound, otherwise branches into child nodes. An example schedules jobs on a single machine to minimize maximum lateness using branch and bound by fixing jobs in the partial schedule at each level and using preemptive EDD to lower bound unscheduled jobs.

    Copyright:

    © All Rights Reserved
    Download as DOC, PDF, TXT or read online from Scribd
    Flag for inappropriate content
    of 4

    Branch and Bound

    General purpose method to solve discrete optimization


    problems
    Takes exponential time in the worst-case
    Useful to solve small instances of hard problems
    Can be applied to all scheduling problems
    Need to specif!
    "# $ lower bounding scheme
    %# Branching rule
    &# 'earch rule (subproblem selection rule)
    *verview!
    +or a minimization problem, at a given branch and bound tree
    (b-b) node i!
    "# *btain a lower bound, lb
    i
    %# (*ptional) *btain a candidate solution and an upper bound
    ub
    i
    &# .f (ub
    i
    / G0*B$0 UB) then set ub
    i
    as the G0*B$0 UB
    1# .f (lb
    i
    23 G0*B$0 UB) then prune the node and go to 4#
    5# Branch into subproblems (create child nodes)
    4# 6ick the next active subproblem#
    .f none exist, stop# 7lse, go to "#
    7xample "
    Problem: 'ingle machine, n 8obs arriving at different times,
    preemptions not allowed, minimize maximum lateness (0
    max
    )
    $ b-b node corresponds to a partial schedule
    $t level k of the b-b tree, the first k 8obs in the schedule
    have been fixed
    Branching
    Create a child node of a node at level k-" b fixing each
    remaining 8ob at kth position 79C76T!
    If a job c satisfies
    ) ) , (max( min
    l l
    J l
    c
    p r t r +

    ,
    do not create a child for that job
    0ower bounding!
    'chedule remaining 8obs according to the preemptive EDD
    rule!
    Every time the machine is freed or a new job is released,
    pick the uncompleted job with minimum due date
    Numeric Example:
    Jobs 1 2 3 4
    p
    j
    4 2 6 5
    d
    j
    8 12 11 10
    r
    j
    0 1 3 5
    0ower bound at node (",-) is obtained b appling preemptive
    7:: to 8obs %, & and 1 with starting time 1
    $t t31, available 8obs! %, &, pick &
    $t t35, available 8obs! %, &,1 pick 1
    $t t3";, available 8obs! %, &, pick &
    $t t3"5, available 8obs! %, pick %
    0
    "
    3 -1, 0
    %
    3 5, 0
    &
    3 1, 0
    1
    3 ;, 0
    max
    3 5
    Exercise: Calculate the lower bound at node (%,-)
    ;
    ",
    &, %,
    ",&,1,%
    1,
    ",&, ",%,
    lb3 5
    lb3 4
    lb3 <
    lb3 5
    ub3 5
    candidate
    prune prune
    0evel &
    0evel %
    0evel "
    0evel ;
    & " & 1 %
    1
    5
    ";
    ;
    "5
    "<
    6reemptive 7:: at node (",&,-) gives a non-preemptive
    schedule ", &, 1, % with 0
    max
    3 5 and it is an optimal solution to
    the problem

    You might also like