topic Re: Sas Studio Proc Optmodel: Multi-part constraint in Mathematical Optimization, Discrete-Event Simulation, and OR

Sas Studio Proc Optmodel: Multi-part constraint

ccockrum — Wed, 14 Dec 2016 22:10:38 GMT

Hey, I am needing some help figuring out how to do a multi-part constraint for the proc optmodel. Here is the data I am working with and trying to structure correctly. The specific part I am struggling with is how to do setup the minimum and maximum production constraints per product per plant. I've attached an Excel spreadsheet showing the data I am working with and below is my attempt at writing the SAS code correctly.

Proc optmodel;
set Plant = {'Osaka','Nagoya','Yokohama','Nara','Fukuoka'};
set Product = {'Haihai','Sensei','Hiro','Kirito'};

number Supply {Plant} = [673 580 659 662 623] ;
number Demand {Product} = [514 527 495 493];
number VariaCost {Plant,Product} = [
25.25	24.4	25.45	24.6
21.15	22.3	24.8	23.15
22.5	25.35	24.85	23.35
24.05	24.4	21.65	22.35
21.5	23.5	23.5	21.85];
number FixedCost {Plant} = [1434 1249 2527];
number M = 1000000;
number MinUse {Plant,Product} = [
60	98	139	133
71	85	80	55
70	61	56	57
136	145	134	144
109	95	118	145];
number MaxUse {Plant,Product} = [
354	343	369	346
299	324	358	319
347	338	290	340
327	325	280	323
308	300	370	354];

var X {Plant,Product} integer >= 0;
var Y {Plant} binary;

minimize TotalCost = sum {i in Plant} ( sum {j in Product} ( X[i,j]*VariaCost[i,j] ) + Y[i]*FixedCost[i] );

con SupplyConst {i in Plant} : sum {j in Product} X[i,j]<=Supply[i];
con DemandConst {j in Product} : sum {i in Plant} X[i,j]>=Demand[j];
con Producing {i in Plant} : sum {j in Product} X[i,j] <= M * Y[i];
con UseMinimum {i in Plant} : sum {j in Product} X[i,j] >= MinUse[i] * Y[i];




solve;

print TotalCost X;

quit;

Thanks in advance for helping a rookie,

Chad

Re: Sas Studio Proc Optmodel: Multi-part constraint

RobPratt — Thu, 15 Dec 2016 02:21:09 GMT

From the data in the spreadsheet, it looks like you want:

number FixedCost {Plant,Product} = [

And:

var Y {Plant,Product} binary;

With corresponding changes in the objective declaration. And then:

/*con Producing {i in Plant} : sum {j in Product} X[i,j] <= M * Y[i];*/
/*con UseMinimum {i in Plant} : sum {j in Product} X[i,j] >= MinUse[i] * Y[i];*/
con UseMinimum {i in Plant, j in Product}: X[i,j] >= MinUse[i,j] * Y[i,j];
con UseMaximum {i in Plant, j in Product}: X[i,j] <= MaxUse[i,j] * Y[i,j];

You don't need M.

Re: Sas Studio Proc Optmodel: Multi-part constraint

ccockrum — Thu, 15 Dec 2016 18:05:14 GMT

Thanks for the help Rob, I really appreciate it. I thought I was tracking with you on the changes, but I've hit some errors I'm not sure how to debug.

at line 91:

minimize TotalCost = sum {i in Plant} ( sum {j in Product} ( X[i,j]*VariaCost[i,j] ) + Y[i]*FixedCost[i] );

ERROR 618-782: The subscript count does not match array 'Y', 1 NE 2.

ERROR 618-782: The subscript count does not match array 'FixedCost', 1 NE 2.

at line 96:

con Producing {i in Plant} : sum {j in Product} X[i,j] <= M * Y[i];

ERROR 618-782: The subscript count does not match array 'Y', 1 NE 2.

at line 97:

con UseMinimum {i in Plant} : sum {j in Product} X[i,j] >= MinUse[i] * Y[i];

ERROR 618-782: The subscript count does not match array 'MinUse', 1 NE 2.

ERROR 618-782: The subscript count does not match array 'Y', 1 NE 2.

and finally:

ERROR: The constraint 'Producing' has an incomplete declaration.

Am I correct in assuming that one has to M?

latest code

Proc optmodel;
set Plant = {'Osaka','Nagoya','Yokohama','Nara','Fukuoka'};
set Product = {'Haihai','Sensei','Hiro','Kirito'};

number Supply {Plant} = [673 580 659 662 623] ;
number Demand {Product} = [514 527 495 493];
number VariaCost {Plant,Product} = [
25.25	24.4	25.45	24.6
21.15	22.3	24.8	23.15
22.5	25.35	24.85	23.35
24.05	24.4	21.65	22.35
21.5	23.5	23.5	21.85];
number FixedCost {Plant,Product} = [
1415	1392	803	618
321	755	942	706
1252	1178	1391	1362
1408	1349	901	1386
619	702	1473	1493];
number M = 1000000;
number MinUse {Plant,Product} = [
60	98	139	133
71	85	80	55
70	61	56	57
136	145	134	144
109	95	118	145];
number MaxUse {Plant,Product} = [
354	343	369	346
299	324	358	319
347	338	290	340
327	325	280	323
308	300	370	354];

var X {Plant,Product} integer >= 0;
var Y {Plant,Product} binary;

minimize TotalCost = sum {i in Plant} ( sum {j in Product} ( X[i,j]*VariaCost[i,j] ) + Y[i]*FixedCost[i] );


con SupplyConst {i in Plant} : sum {j in Product} X[i,j]<=Supply[i];
con DemandConst {j in Product} : sum {i in Plant} X[i,j]>=Demand[j];
con Producing {i in Plant} : sum {j in Product} X[i,j] <= M * Y[i];
con UseMinimum {i in Plant} : sum {j in Product} X[i,j] >= MinUse[i] * Y[i];
con UseMinimum {i in Plant, j in Product}: X[i,j] >= MinUse[i,j] * Y[i,j];
con UseMaximum {i in Plant, j in Product}: X[i,j] <= MaxUse[i,j] * Y[i,j];


solve;

print TotalCost X;

quit;

Re: Sas Studio Proc Optmodel: Multi-part constraint

RobPratt — Thu, 15 Dec 2016 19:06:16 GMT

Those error messages are telling you that you can't use Y[i] and FixedCost[i] because those arrays now require two subscripts each. You need Y[i,j] and FixedCost[i,j] instead.

The Producing and UseMinimum constraints that I commented out are obsolete, and M is now also obsolete because it appears only in Producing.

Re: Sas Studio Proc Optmodel: Multi-part constraint

ccockrum — Fri, 23 Dec 2016 19:03:23 GMT

So I made the change to Y[i,j]

minimize TotalCost = sum {i in Plant} ( sum {j in Product} ( X[i,j]*VariaCost[i,j] ) + Y[i,j]*FixedCost[i,j] );

But I am still getting these errors on that line of code:

ERROR 525-782: The symbol 'j' is unknown.

ERROR 618-782: The subscript count does not match array 'Y', 2 NE 1.

ERROR 618-782: The subscript count does not match array 'FixedCost', 2 NE 1.

Thanks!

Re: Sas Studio Proc Optmodel: Multi-part constraint

RobPratt — Fri, 23 Dec 2016 19:26:27 GMT

Your parentheses on the summand are incorrect. You can use either one of the following:

minimize TotalCost = sum {i in Plant} ( sum {j in Product} ( X[i,j]*VariaCost[i,j]  + Y[i,j]*FixedCost[i,j] ));
minimize TotalCost = sum {i in Plant, j in Product} ( X[i,j]*VariaCost[i,j]  + Y[i,j]*FixedCost[i,j] );

Re: Sas Studio Proc Optmodel: Multi-part constraint

ccockrum — Fri, 23 Dec 2016 20:28:58 GMT

Okay, I'm following you. What about the other error, that one is this there.

 86         minimize TotalCost = sum {i in Plant} ( sum {j in Product} ( X[i,j]*VariaCost[i,j]  + Y[i,j]*FixedCost[i,j] ));
                                                                                                       _
                                                                                                       618
 ERROR 618-782: The subscript count does not match array 'Y', 2 NE 1.
 
 86       ! minimize TotalCost = sum {i in Plant} ( sum {j in Product} ( X[i,j]*VariaCost[i,j]  + Y[i,j]*FixedCost[i,j] ));
                                                                                                                      _
                                                                                                                      618
 ERROR 618-782: The subscript count does not match array 'FixedCost', 2 NE 1.

Total code now looks like this:

Proc optmodel;
set Plant = {'Osaka','Nagoya','Yokohama','Nara','Fukuoka'};
set Product = {'Haihai','Sensei','Hiro','Kirito'};

number Supply {Plant} = [673 580 659 662 623] ;
number Demand {Product} = [514 527 495 493];
number VariaCost {Plant,Product} = [
25.25	24.4	25.45	24.6
21.15	22.3	24.8	23.15
22.5	25.35	24.85	23.35
24.05	24.4	21.65	22.35
21.5	23.5	23.5	21.85];
number FixedCost {Plant} = [1434 1249 2527];
number M = 1000000;
number MinUse {Plant,Product} = [
60	98	139	133
71	85	80	55
70	61	56	57
136	145	134	144
109	95	118	145];
number MaxUse {Plant,Product} = [
354	343	369	346
299	324	358	319
347	338	290	340
327	325	280	323
308	300	370	354];

var X {Plant,Product} integer >= 0;
var Y {Plant} binary;

minimize TotalCost = sum {i in Plant} ( sum {j in Product} ( X[i,j]*VariaCost[i,j]  + Y[i,j]*FixedCost[i,j] ));

con SupplyConst {i in Plant} : sum {j in Product} X[i,j]<=Supply[i];
con DemandConst {j in Product} : sum {i in Plant} X[i,j]>=Demand[j];



solve;

print TotalCost X;

quit;

Thanks for taking another look at this.

Re: Sas Studio Proc Optmodel: Multi-part constraint

RobPratt — Fri, 23 Dec 2016 23:48:14 GMT

Those additional errors are because you switched back to the old declarations for FixedCost and Y. Also, you still need the new UseMinimum and UseMaximum constraints.