MATLAB实现-整数规划之分支定点法

代码实现

%main.m
%分支定界法

%求解模型：
%min z = f * x
%A * x <= b
%x >= 0,且为整数

clear gloal

global result;%存储所有整数解
global lowerBound;%下界
global upperBound;%上界
global count; %用来判断是否为第一次分支

f = [-40, -90];
A = [9, 7;
    7, 20;];
b = [56;70];
Aeq = [];
beq = [];
lb = [0; 0];
ub = [inf; inf]

[x, fval] = linprog(f, A, b, Aeq, beq, lb, ub);
x
abs(fval)
maxz = (-f) * x


%对x1进行分支 x1<=4
c = [40, 90];
A = [9, 7; 7, 20];
b = [56, 70];
Aeq = [];beq = [];
lb = [0;0];
ub = [4;inf];
[x, fval] = linprog(-c, A, b, Aeq, beq, lb, ub);
x
maxz = abs(fval)

%x1分支+x2分支 x1<=4,x2<=2
c = [40, 90];
A = [9, 7; 7, 20];
b = [56, 70];
Aeq = [];beq = [];
lb = [0;0];
ub = [4;2];
[x, fval] = linprog(-c, A, b, Aeq, beq, lb, ub);
x
maxz = abs(fval)

%x1+x2分支 x1 <=4, x2>=3
c = [40, 90];
A = [9, 7; 7, 20];
b = [56, 70];
Aeq = [];beq = [];
lb = [0;3];
ub = [4;inf];
[x, fval] = linprog(-c, A, b, Aeq, beq, lb, ub);
x
maxz = abs(fval)

%对x1分支 x1>=5
c = [40, 90];
A = [9, 7; 7, 20];
b = [56, 70];
Aeq = [];beq = [];
lb = [5;0];
ub = [inf;inf];
[x, fval] = linprog(-c, A, b, Aeq, beq, lb, ub);
x
maxz = abs(fval)

%x1+x2分支 x1>=5+x2<=2
c = [40, 90];
A = [9, 7; 7, 20];
b = [56, 70];
Aeq = [];beq = [];
lb = [5;0];
ub = [inf;2];
[x, fval] = linprog(-c, A, b, Aeq, beq, lb, ub);
x
maxz = abs(fval)

%x1+x2分支，x1>=5+x2>=3
c = [40, 90];
A = [9, 7; 7, 20];
b = [56, 70];
Aeq = [];beq = [];
lb = [5;3];
ub = [inf;inf];
[x, fval] = linprog(-c, A, b, Aeq, beq, lb, ub);
x
maxz = abs(fval)

运行结果

线性规划运行结果

然后对变量x1,x2进行分支计算

x1分支1（x1<=4）

x1,x2分支（x1<=4， x2<=2）

x1,x2分支（x1<=4， x2>=3）

x1分支2（x1>=5）

x1,x2分支(x1>=5， x2<=2)

x1,x2分支（x1>=5， x2>=3）

所以这个题目整数规划结果为