Hello, 

I have a scheduling model and i have no way to export the decision variables (I cannot select all and copy paste in excel). My version 12.10.0.0 so I am not able to export the results in excel. In this connection, I am trying to use doopl for the purpose of extracting the results. But I am not sure how to convert the interval variables to a table, so that i can use get_table() in the doopl. Could anybody help me with several lines of code ? 

The extraction of my mod file is the following: 
using CP;
tuple Workcenter {string name; int cap;}
tuple Alt {string po; string wc;}

{Workcenter} WCs = ...;
{string} PONames = ...;
{Alt} Alts = ...;

int Duration [t in PONames] = ...;

tuple Precedence {
string pre;
string post;
};

{Precedence} Precedences = ...;

int ReleaseDate[t in PONames] = ...;
int DueDate [t in PONames] = ...;
float Weight [t in PONames] = ...;

//Create the production order interval variables
dvar interval itvs[t in PONames] size Duration[t];
//Create the alternative resource, i.e., workcenter interval variables
dvar interval alt[a in Alts] optional;

////Create the sequence variables
//dvar sequence wcs[w in WCs] in
//all(t in PONames: WCs[t]==w) itvs[t] ;

execute {
cp.param.FailLimit = 10000;
}

minimize max(t in PONames)
(Weight[t] * maxl(0, endOf(itvs[t])-DueDate[t]) + lengthOf(itvs[t]) );

subject to {
//Add the precendence constraints
forall(p in Precedences)
endBeforeStart(itvs[p.pre], itvs[p.post]);

//start after the release date
forall(t in PONames)
ReleaseDate[t] <= startOf(itvs[t]);

//Add the capacity constraints
forall(w in WCs)
sum(a in Alts: a.wc == w.name) pulse(alt[a],1) <= w.cap;

//alternative constraints
forall(t in PONames)
alternative(itvs[t], all(a in Alts: a.po==t) alt[a]);

}



------------------------------
Quan Long
------------------------------

#DecisionOptimization

You have to create a tupleSet from the values of the decision variables. This tupleSet will be available via the get_table.

for example with:
{string} PONames = ...;
dvar interval itvs[t in PONames]

you can for example have something like the following to query the start/end of the activity
tuple Report{
string name;
int start;
int end;
}
{Report} report = {<j, startOf(itvs[j]), endOf(itvs[j])> | j in PONames};

------------------------------
Vincent Beraudier
------------------------------

Original Message:
Sent: Wed November 09, 2022 02:45 PM
From: Quan Long
Subject: how to retrieve cplex decision variables using doopl

Hello, 

I have a scheduling model and i have no way to export the decision variables (I cannot select all and copy paste in excel). My version 12.10.0.0 so I am not able to export the results in excel. In this connection, I am trying to use doopl for the purpose of extracting the results. But I am not sure how to convert the interval variables to a table, so that i can use get_table() in the doopl. Could anybody help me with several lines of code ? 

The extraction of my mod file is the following: 
using CP;
tuple Workcenter {string name; int cap;}
tuple Alt {string po; string wc;}

{Workcenter} WCs = ...;
{string} PONames = ...;
{Alt} Alts = ...;

int Duration [t in PONames] = ...;

tuple Precedence {
string pre;
string post;
};

{Precedence} Precedences = ...;

int ReleaseDate[t in PONames] = ...;
int DueDate [t in PONames] = ...;
float Weight [t in PONames] = ...;

//Create the production order interval variables
dvar interval itvs[t in PONames] size Duration[t];
//Create the alternative resource, i.e., workcenter interval variables
dvar interval alt[a in Alts] optional;

////Create the sequence variables
//dvar sequence wcs[w in WCs] in
//all(t in PONames: WCs[t]==w) itvs[t] ;

execute {
cp.param.FailLimit = 10000;
}

minimize max(t in PONames)
(Weight[t] * maxl(0, endOf(itvs[t])-DueDate[t]) + lengthOf(itvs[t]) );

subject to {
//Add the precendence constraints
forall(p in Precedences)
endBeforeStart(itvs[p.pre], itvs[p.post]);

//start after the release date
forall(t in PONames)
ReleaseDate[t] <= startOf(itvs[t]);

//Add the capacity constraints
forall(w in WCs)
sum(a in Alts: a.wc == w.name) pulse(alt[a],1) <= w.cap;

//alternative constraints
forall(t in PONames)
alternative(itvs[t], all(a in Alts: a.po==t) alt[a]);

}



------------------------------
Quan Long
------------------------------

#DecisionOptimization

Original Message:
Sent: 11/16/2022 10:23:00 AM
From: Vincent Beraudier
Subject: RE: how to retrieve cplex decision variables using doopl


You have to create a tupleSet from the values of the decision variables. This tupleSet will be available via the get_table.

for example with:
{string} PONames = ...;
dvar interval itvs[t in PONames]

you can for example have something like the following to query the start/end of the activity
tuple Report{
string name;
int start;
int end;
}
{Report} report = {<j, startOf(itvs[j]), endOf(itvs[j])> | j in PONames};

------------------------------
Vincent Beraudier
------------------------------

Original Message:
Sent: Wed November 09, 2022 02:45 PM
From: Quan Long
Subject: how to retrieve cplex decision variables using doopl

Hello, 

I have a scheduling model and i have no way to export the decision variables (I cannot select all and copy paste in excel). My version 12.10.0.0 so I am not able to export the results in excel. In this connection, I am trying to use doopl for the purpose of extracting the results. But I am not sure how to convert the interval variables to a table, so that i can use get_table() in the doopl. Could anybody help me with several lines of code ? 

The extraction of my mod file is the following: 
using CP;
tuple Workcenter {string name; int cap;}
tuple Alt {string po; string wc;}

{Workcenter} WCs = ...;
{string} PONames = ...;
{Alt} Alts = ...;

int Duration [t in PONames] = ...;

tuple Precedence {
string pre;
string post;
};

{Precedence} Precedences = ...;

int ReleaseDate[t in PONames] = ...;
int DueDate [t in PONames] = ...;
float Weight [t in PONames] = ...;

//Create the production order interval variables
dvar interval itvs[t in PONames] size Duration[t];
//Create the alternative resource, i.e., workcenter interval variables
dvar interval alt[a in Alts] optional;

////Create the sequence variables
//dvar sequence wcs[w in WCs] in
//all(t in PONames: WCs[t]==w) itvs[t] ;

execute {
cp.param.FailLimit = 10000;
}

minimize max(t in PONames)
(Weight[t] * maxl(0, endOf(itvs[t])-DueDate[t]) + lengthOf(itvs[t]) );

subject to {
//Add the precendence constraints
forall(p in Precedences)
endBeforeStart(itvs[p.pre], itvs[p.post]);

//start after the release date
forall(t in PONames)
ReleaseDate[t] <= startOf(itvs[t]);

//Add the capacity constraints
forall(w in WCs)
sum(a in Alts: a.wc == w.name) pulse(alt[a],1) <= w.cap;

//alternative constraints
forall(t in PONames)
alternative(itvs[t], all(a in Alts: a.po==t) alt[a]);

}



------------------------------
Quan Long
------------------------------

#DecisionOptimization