Using CPLEX python API for linear programming with continuous inputs

Question

I am new to CPLEX Python API. I wish to solve a Linear Programming problem in python which I have already done in the CPLEX OPL IDE by taking a .mod and .dat files as inputs. I want to use it in python since I wish to vary my inputs continuously. My mod file for the problem is given below. Can someone help me on how to use this for the python API.

int n = ...; 
int m = ...; 

int c = ...; 
int s = ...; 

range v = 1..n;
range p = 1..m;

int c_req[v] = ...;
int s_req[v] = ...;

int trust[v][v] = ...;


// decision variables

dvar boolean assign[p][v]; 

// expressions

dexpr int used[pi in p] = max(vi in v) assign[pi][v]; // used[i] = 1     iff pi is used
dexpr int totalUsed = sum(pi in p) used[pi];

execute {
  cplex.tilim = 60; // Time limit 60 seconds
}

// model

minimize totalUsed;

subject to {
  forall(pi in p) 
    c_cap:
    sum(vi in v) c_req[vi] * assign[pi][vi] <= c;

  forall(pi in p)
    s_cap:
    sum(vi in v) s_req[vi] * assign[pi][vi] <= s;

  forall(vi in v)
    v_all:
    sum(pi in p) assign[pi][vi] == 1;

  forall(pi in p, v1 in v, v2 in v) if (v1 < v2) if (trust[v1][v2] ==  0)
    trust_constraint:
    assign[p][v1] + assign[p][v2] <= 1;
}

Alex Fleischer Alex Fleischer · Accepted Answer · 2017-03-01T08:22:09

you could write

subprocess.check_call(["C:/CPLEXStudio127/opl/bin/x64_win64/oplrun", "diet.mod", "diet.dat"])

in order to call OPL from python. And you would generate diet.dat beforehand.

Full example at https://www.ibm.com/developerworks/community/forums/html/threadTopic?id=0b6cacbe-4dda-4da9-9282-f527c3464f47

Then you do not have to migrate your model from OPL to Python.

You may also translate your model to Python and then I recommend DOCPLEX : https://developer.ibm.com/docloud/documentation/optimization-modeling/modeling-for-python/

regards

Using CPLEX python API for linear programming with continuous inputs

1 Answers