1/16
Looks like no tags are added yet.
Name | Mastery | Learn | Test | Matching | Spaced |
---|
No study sessions yet.
How do you know if an LP is infeasible?
If the optimal solution of the Phase I simplex is not equal to 0
What us Phase I simplex used for?
To find a starting basis and to check infeasibility
LB and UB for a maximization branch and bound tree
LB: Best integer feasible solution
UB: Best we can hope to get
LB and UB for a minimization branch and bound tree
LB: Best we can hope to get
UB: Best integer feasible solution
Cutting planes algorithm
1). Write the LP Relaxation in standard form and solve using simples
2). Take one of the non integer constraints and move all the variables to one side
3). Floor all the variables and use less than equal to and floor the right side
4). Substitute all the non-basis variables
5). Add to IP
Max min
z <=
Min Max
z>=
Absolute value in objective
change to max
max in constraints:
max{12,12+5x}<=500
12 <= 500
12+5x <= 500
Absolute value in constraints
turn into max and do same
|x-2y|>=2
(x-2y>=2) OR (-x+2y>=2)
Primal is optimal
Dual is optimal
Primal is unbounded
Dual is infeasible
Primal is infeasible
Dual is unbounded or infeasible
Fixed cost constraint
#of units variable <= M (if fixed cost is accrued constraint)
x<=0 in standard form
x’=-x and replace x accordingy
x is free
replace x with (x’-x”)