BookmarkSubscribeRSS Feed
☑ This topic is solved. Need further help from the community? Please sign in and ask a new question.
Santha
Pyrite | Level 9

Rob

Thanks for your support as always. I am learning a lot of stuff thanks to you. 

I am creating a new thread, based on this, (earlier thread), just to be clear. I think we are almost done, just left with 4 questions/help. I am trying to replicate the toy model to the full model by adding indices for T1 (days, T2(weeks) and Org, Des because rates and specs are function of org, des, box, mol, t1 and t2 as you know.  Based on your earlier comment, I realized that we should go for the case where "vendor cooperates with each other to reduce overall network cost", in which case we can make the model simpler by not having the fix/unfix and the second solve. The $3,622 is what we need as it is the true optimum. When I remove those fix/unfixes , rational constraint and second solve, the model gives $3,622 which is what we want. I want the same $3,622 when I do the full model. That is my goal. Now I have 4 questions when we translate this to full model . My full model code is at the bottom of these 4 questions. 

(1) Without i in ISN: For the full model,  I have no problems when I have i in ISN because it has Org, DES, T1 and T2 in it and I can cull out using [i], example Org[i]. But when it comes to CFS constraints,  impvars, I am not able to do it correctly as there are no i in ISN. 

(2) Proportion: There is no proportion decision variable and a constraint related to it in the code. But I tested few different numbers of Volume in ISNs and the code seems to be doing it correctly. I just wanted to make sure that the proportion is accounted for, i.e., for a given ISN, the proportion in each box adds up to 1 for CYS. And for CFS, for a given week, the proportion of Total Volume adds up to 1. The following code is what we had earlier where there was no  MOL, day and week, that was done a year ago. I tried in the full model but was not successful. If we do not need these proportion variables at all, it then it is fine. 

 

var Proportion {ISN,BOX} >= 0 <= 1; impvar VolumeInsideBox {i in ISN, b in BOX} = Volume_ISN[i] * Proportion[i,b]; 
con Volume_Constraint {i in ISN, b in BOX}: Volume_Capacity[Org_ISN[i],Des_ISN[i],b] * BoxesNeeded[i,b] >= VolumeInsideBox[i,b];

3) Create Data step: For the output, I would like to have it in a format like the one that I have attached in this spreadsheet (adding on to the Toy.xlsx) from cells  A25 to X29. Again, I tried (just for CYS) in the full model but was facing syntax issues. 

(4) Rational Vendor error: This is the last preference as we are not using the constraint anymore, now that we said we will let the vendor cooperate. This will be just for my understanding. Not an urgent one. 

126     con RationalVendor {i in ISN}:
127         IsMol[i,'CFS'] = 1 implies
                               -------
                               22
                               76
ERROR 22-322: Syntax error, expecting one of the following: ;, !!, *, **, +, ',', -, .., /, <>, ><, BY, CROSS, DIFF, ELSE, INTER, 
              SUFFIXES, SYMDIFF, TO, UNION, ^, ||.  
ERROR 76-322: Syntax error, statement will be ignored.
128         vol_ISN[i] * (BoxBasedCostCFS + sum {j in ISN} (volRate['CFS'] * vol_ISN[j] + shipmentRate['CFS']) * IsMol[j,'CFS'])
129         <= optCYSCost[i] * TotalVol['CFS'];

 My full code is here below. 

proc optmodel;

set <str> ORG;
read data CASUSER.Unique_ORG into ORG = [ORG];

set <str> DES; 
read data CASUSER.Unique_DES into DES = [DES];

set <str> BOX; 
read data CASUSER.Unique_BOX into BOX = [BOX];

set <str> MOL; 
read data CASUSER.Unique_MOL into MOL = [MOL];

set <str> T1; 
read data CASUSER.Unique_T1 into T1 = [T1];

set <str> T2; 
read data CASUSER.Unique_T2 into T2 = [T2];


/*Read ISN Wt and Volume for a given ISN* - START */
set <str> ISN;
str Org_ISN {ISN};
str Des_ISN {ISN};
str T1_ISN {ISN};
str T2_ISN {ISN};
num Volume_ISN {ISN} init 0;
num Weight_ISN {ISN} init 0;

read data CASUSER.InputData_AssetMix into ISN=[ISN] Org_ISN=ORG Des_ISN=DES Volume_ISN=Volume Weight_ISN=Weight T1_ISN=T1 T2_ISN=T2;

num T2_Volume{T2} init 0;
read data CASUSER.Unique_T2 into [T2] T2_Volume=T2_Volume;

num T2_Weight{T2} init 0;
read data CASUSER.Unique_T2 into [T2] T2_Weight=T2_Weight;
/*Read ISN Wt and Volume for a given ISN* - END*/

/*Read Wt and Volume Capacity for a given ORG, DES, BOX - START */
num Volume_Capacity {ORG,DES,BOX,MOL,T1,T2} init 0;  num Volume_Min {ORG,DES,BOX,MOL,T1,T2} init 0;
num Weight_Capacity {ORG,DES,BOX,MOL,T1,T2} init 0;  num Weight_Min {ORG,DES,BOX,MOL,T1,T2} init 0;
read data CASUSER.BOXSPECS into [ORG DES BOX MOL T1 T2] 
Volume_Capacity=Volume_Capacity
Weight_Capacity=Weight_Capacity
Volume_Min=Volume_Min
Weight_Min=Weight_Min;
/*Read Wt and Volume Capacity for a given ORG, DES, BOX - END */


/* A Binary variable that is 0 or 1 for a given asset for a given lane, 0 indicated not available, 1- available */
num Is_BoxAvalable_for_a_lane {ORG,DES,BOX,MOL,T1,T2};
read data CASUSER.BOXSpecs into [ORG Des Box Mol T1 T2] Is_BoxAvalable_for_a_lane = BoxAvailability;

/* Decision Variable - START */
var BoxesNeeded_CYS {ISN,BOX,T1,T2} >= 0 integer;
var BoxesNeeded_CFS {BOX,T2} >= 0 integer;
var Is_ISN_MOL{ISN,MOL} binary;
/* Decision Variable and Associated Impvars - END */

/*Define Rates and ImplicitVariables START */
/*Box Based Costs START */

set <str,str,str,str,str,str> PerBox_Based_Rate_NoZeroes;
num PerBox_Based_Rate {PerBox_Based_Rate_NoZeroes}; 
read data CASUSER.BOXRATE (where=(RateBasis="PerBox" and Ratetype='Linehaul' and Rate>0)) 
into PerBox_Based_Rate_NoZeroes=[ORG DES BOX MOL T1 T2] PerBox_Based_Rate=Rate;

impvar PerBox_Based_Costs_CYS {i in ISN} = sum {b in BOX,m in {'CYS'},d in T1,w in T2: <Org_ISN[i],Des_ISN[i],b,m,d,w> in PerBox_Based_Rate_NoZeroes}
PerBox_Based_Rate[Org_ISN[i],Des_ISN[i],b,m,d,w] * BoxesNeeded_CYS[i,b,d,w];


impvar PerBox_Based_Costs_CFS {w in T2} = sum {b in BOX} PerBox_Based_Rate[b,w] * BoxesNeeded_CFS[b,w];

impvar TotalBoxBasedCost = sum {i in ISN} PerBox_Based_Costs_CYS[i] + sum {w in T2} PerBox_Based_Costs_CFS[w];

/*Box Based Costs END */

/*Vol Based Costs START*/
set <str,str,str,str,str,str> PerVol_Based_Rate_NoZeroes;
num PerVol_Based_Rate {PerVol_Based_Rate_NoZeroes}; 
read data CASUSER.BOXRATE (where=(RateBasis="PerVolUOM" and Ratetype='Carrier_BOL' and Rate>0)) 
into PerVol_Based_Rate_NoZeroes=[ORG DES MOL BOX T1 T2] PerVol_Based_Rate=Rate;

impvar VolBasedCost {i in ISN} = sum {b in BOX,m in MOL,d in T1,w in T2: <Org_ISN[i],Des_ISN[i],b,m,T1_ISN[i],T2_ISN[i]> in PerVol_Based_Rate_NoZeroes}
Volume_ISN[i] * PerVol_Based_Rate[Org_ISN[i],Des_ISN[i],b,m,T1_ISN[i],T2_ISN[i]] * Is_ISN_Mol[i,m];

impvar TotalVolBasedCost = sum {i in ISN} VolBasedCost[i];

/*Vol Based Costs END*/

/*Shipment Based Costs START*/
set <str,str,str,str,str,str> PerShp_Based_Rate_NoZeroes;
num PerShp_Based_Rate {PerShp_Based_Rate_NoZeroes}; 
read data CASUSER.BOXRATE (where=(RateBasis="PerShipment" and Ratetype='Export_Declaration' and Rate>0)) 
into PerShp_Based_Rate_NoZeroes=[ORG DES MOL BOX T1 T2] PerShp_Based_Rate=Rate;

impvar ShipmentBasedCost {i in ISN} = sum {b in BOX,m in MOL,d in T1,w in T2: <Org_ISN[i],Des_ISN[i],b,m,T1_ISN[i],T2_ISN[i]> in PerShp_Based_Rate_NoZeroes}
PerShp_Based_Rate[Org_ISN[i],Des_ISN[i],b,m,T1_ISN[i],T2_ISN[i]] * Is_ISN_Mol[i,m];

impvar TotalShipmentBasedCost = sum {i in ISN} ShipmentBasedCost[i];

/*Shipment Based Costs END*/
/*Define Rates and ImplicitVariables END*/


Min TotalCost = TotalBoxBasedCost + TotalVolBasedCost + TotalShipmentBasedCost;

/* Constraints START*/

for {i in ISN, b in BOX,m in MOL,d in T1,w in T2: Is_BoxAvalable_for_a_lane[Org_ISN[i],Des_ISN[i],b,m,d,w] = 0}
      fix BoxesNeeded_CYS[i,b,d,w] = 0;

for {i in ISN, b in BOX,m in MOL,d in T1,w in T2: Is_BoxAvalable_for_a_lane[Org_ISN[i],Des_ISN[i],b,m,d,w] = 0}
      fix BoxesNeeded_CFS[b,w] = 0;

con OneMOL{i in ISN}:
    sum{M in MOL} Is_ISN_MOL[i,m] = 1; /* This is to make sure one ISN can go either CYS or CFS */

con BoxConCYS {i in ISN}:
    sum {b in BOX} Volume_Capacity[Org_ISN[i],Des_ISN[i],b,'CYS',T1_ISN[i],T2_ISN[i]] * BoxesNeeded_CYS[i,b,T1_ISN[i],T2_ISN[i]] >= Volume_ISN[i] * Is_ISN_Mol[i,'CYS'];

impvar TotalVol {m in MOL} = sum {i in ISN} Volume_ISN[i] * Is_ISN_Mol[i,m];

con BoxConCFS: {i in ISN}:
      sum {b in BOX} Volume_Capacity[Org_ISN[i],Des_ISN[i],b,'CYS',T1_ISN[i],T2_ISN[i]] * BoxesNeeded_CFS[b,T2_ISN[i]] >= TotalVol['CFS'];


solve with milp / decomp=(method=concomp);

num optCYSCost {ISN};

for {i in ISN} optCYSCost[i] = PerBox_Based_Costs_CYS[i] + VolBasedCost[i] + ShipmentBasedCost[i];
print BoxBasedCostCYS VolBasedCost ShipmentBasedCost optCYSCost;

   num volShareCFS {i in ISN} = vol_ISN[i] * Is_ISN_Mol[i,'CFS'].sol / TotalVol['CFS'].sol;

   print IsMol;
   print NumBoxesCYS NumBoxesCFS;
   print BoxBasedCostCYS BoxBasedCostCFS;
   print volShareCFS VolBasedCost TotalVolBasedCost TotalVol;
   print ShipmentBasedCost;
quit;

 

 

 

1 ACCEPTED SOLUTION

Accepted Solutions
RobPratt
SAS Super FREQ

Except for BoxRate.xlsx, it looks like you didn't provide the latest data tables you are using.

 

To avoid the duplicate key warnings, make sure the columns that appear within the square brackets [] in the READ DATA statement uniquely identify the observation.

 

The latest code you sent yields errors related to two IMPVAR declarations.  Here are two corrections that should help you get further:

 

   impvar VolBasedCost {i in ISN} = Volume_ISN[i] * 
/*      sum {m in MOL} PerVol_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]]*Is_ISN_MOL[i,m];*/
      sum {<Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]> in PerVol_Based_Rate_NoZeroes} PerVol_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]] * Is_ISN_MOL[i,m];
   impvar ShipmentBasedCost {i in ISN} = 
/*      sum {m in MOL} PerShp_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]] * Is_ISN_MOL[i,m];*/
      sum {<Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]> in PerShp_Based_Rate_NoZeroes} PerShp_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]] * Is_ISN_MOL[i,m];

 

View solution in original post

12 REPLIES 12
RobPratt
SAS Super FREQ

1.  Please see my response here: https://communities.sas.com/t5/Mathematical-Optimization/AssetOptimization-Contd/m-p/885516/highligh...

2. Do you still need to account for capacity and minThreshold for both volume and weight?  Your latest code uses only volume capacity.

3. If the solution uses more than one box type per shipment, do you want multiple rows for that shipment?

4. That syntax error about IMPLIES means that you are not running a version that supports indicator constraints, which are available only in SAS Viya 4.

Santha
Pyrite | Level 9

Rob

First of all, I can't thank you enough for your support and patience on this one. I tried lot of things to convert to full model. The model is picking up only CFS for a total cost of $692 (173 * $4). So I am not writing something correct. My code is below. Requesting you to see where i am missing and what needs to be done to get to the optimum $3,622. The Box based costs are not getting calculated correctly for some reason that I am not able to figure it out.  I am right now feeling low on confidence for I could not convert it successfully and not getting the syntaxes right and going in circles sometimes.  If this is fixed, I will expand the model for many weeks and many vendors , which I have a Monday deadline. 

  1. In this model,  please  ignore weight, though there are weight variables. . I can add weight later on. Only focus is on "Volume". 
  2. Please ignore minimum thresholds , that can be added later on.  Only focus is on Volume Capacity as constraint. So far this is what the model does and you have correctly used in your code as well. So far so good. 
  3. The syntax of IMPLIES - I got it. not a worry for me right now as we are now in the "all vendors cooperate" mode.
  4. I did use the response in this one to expand to the full model. I had to add variables like "PerBox_Based_Rate_CFS" and var Is_ISN_MOL_T2{ISN,MOL,T2} binary; 
  5. In the toy model, you had the below table for VolRate and ShipmentRates. 

 

 

 

 

num volRate {MOL} = [0 4];
   num shipmentRate {MOL} = [50 0];​

 

 

 

 

But in the full model, when I use the read data only CFS Mol has Volbased rates and only CYS Mol has shipment based rates. When I ran just like, that it threw an error saying "for vol rates, no CYS combo was found"  and so I manually added "0" for CYS Volrates and "0" for CFS ShipmentBased rates. I shall attach this new file in the next comment. When I add this it is ok For now this is fine and I can go with this approach if u say ok. I would like to know if there is an efficient way to do this without adding "0" rates artificially for some rateytpes.  For PerBox_Based_Rate, we had this code below, but that did not pose any problem because all combinations of ORG DES BOX MOL T1 and T2 was there. But if it is not there , will there be a problem?

 

set <str,str,str,str,str,str> PerBox_Based_Rate_NoZeroes;
num PerBox_Based_Rate {PerBox_Based_Rate_NoZeroes};
read data CASUSER.BOXRATE (where=(RateBasis="PerBox" and Ratetype='Linehaul' and Rate>0))
into PerBox_Based_Rate_NoZeroes=[ORG DES BOX MOL T1 T2] PerBox_Based_Rate=Rate;df

6) Duplicates : if u look below,

num Volume_Capacity {BOX};
223  read data CASUSER.BOXSPECS into [BOX] Volume_Capacity=Volume_Capacity;
WARNING: Duplicate key <'20F'> was read at observation 4.I

i have given one instance of duplicates but there are many. The volume capacity is indexed over BOX. But I am extracting from BOXSPECS which has combinations of MOL, T1 and T2. Ideally, the rates and specs were prepared for all combinations of ORG, DES, BOX, MOL, T1, T2 so that we can have different rates wherever applicable. Even for the same lane, SHA-ATL the 20F rate may be $800 in Week 1 and could be $850 the next week just as an example. So , want to learn the efficient way to handling  this "PerVol_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]]".

 

7. About the create data output. yes, if the solution uses more than one box type per shipment, I would like to have multiple rows for that shipment. 

😎My code below:

proc optmodel;

set <str> ORG;
read data CASUSER.Unique_ORG into ORG = [ORG];

set <str> DES; 
read data CASUSER.Unique_DES into DES = [DES];

set <str> BOX; 
read data CASUSER.Unique_BOX into BOX = [BOX];

set <str> MOL; 
read data CASUSER.Unique_MOL into MOL = [MOL];

set <str> T1; 
read data CASUSER.Unique_T1 into T1 = [T1];

set <str> T2; 
read data CASUSER.Unique_T2 into T2 = [T2];

/*Read ISN Wt and Volume for a given ISN* - START */
set <str> ISN;
str Org_ISN {ISN};
str Des_ISN {ISN};
str T1_ISN {ISN};
str T2_ISN {ISN};
num Volume_ISN {ISN};
num Weight_ISN {ISN};

read data CASUSER.InputData_AssetMix into ISN=[ISN] Org_ISN=ORG Des_ISN=DES Volume_ISN=Volume Weight_ISN=Weight T1_ISN=T1 T2_ISN=T2;

num T2_Volume{T2} init 0;
read data CASUSER.Unique_T2 into [T2] T2_Volume=T2_Volume;

num T2_Weight{T2} init 0;
read data CASUSER.Unique_T2 into [T2] T2_Weight=T2_Weight;
/*Read ISN Wt and Volume for a given ISN* - END*/

/*Read Wt and Volume Capacity for a given ORG, DES, BOX - START */
num Volume_Min {ORG,DES,BOX,MOL,T1,T2} init 0;
num Weight_Capacity {ORG,DES,BOX,MOL,T1,T2} init 0;  num Weight_Min {ORG,DES,BOX,MOL,T1,T2} init 0;
read data CASUSER.BOXSPECS into [ORG DES BOX MOL T1 T2] 
Weight_Capacity=Weight_Capacity
Volume_Min=Volume_Min
Weight_Min=Weight_Min;

num Volume_Capacity {BOX}; 
read data CASUSER.BOXSPECS into [BOX] Volume_Capacity=Volume_Capacity;

num Volume_Capacity_CFS{BOX,T2};
read data CASUSER.BOXSPECS into [BOX T2] Volume_Capacity_CFS=Volume_Capacity;
/*Read Wt and Volume Capacity for a given ORG, DES, BOX - END */


/* A Binary variable that is 0 or 1 for a given asset for a given lane, 0 indicated not available, 1- available */
num Is_BoxAvalable_for_a_lane {ORG,DES,BOX,MOL,T1,T2};
read data CASUSER.BOXSpecs into [ORG Des Box Mol T1 T2] Is_BoxAvalable_for_a_lane = BoxAvailability;

/* Decision Variable - START */
var BoxesNeeded_CYS {i in ISN,b in BOX} >= 0 integer <= ceil(Volume_ISN[i] / Volume_Capacity[b]);;
var BoxesNeeded_CFS {b in BOX,w in T2} integer >=0 <= ceil(sum {i in ISN} Volume_ISN[i] / Volume_Capacity[b]);
var Is_ISN_MOL{ISN,MOL} binary;
var Is_ISN_MOL_T2{ISN,MOL,T2} binary;
/* Decision Variable and Associated Impvars - END */

/*Define Rates and ImplicitVariables START */
/*Box Based Costs START */

set <str,str,str,str,str,str> PerBox_Based_Rate_NoZeroes;
num PerBox_Based_Rate {PerBox_Based_Rate_NoZeroes}; 
read data CASUSER.BOXRATE (where=(RateBasis="PerBox" and Ratetype='Linehaul' and Rate>0)) 
into PerBox_Based_Rate_NoZeroes=[ORG DES BOX MOL T1 T2] PerBox_Based_Rate=Rate;

impvar PerBox_Based_Costs_CYS {i in ISN} = sum {b in BOX,m in {'CYS'}}
PerBox_Based_Rate[Org_ISN[i],Des_ISN[i],b,'CYS',T1_ISN[i],T2_ISN[i]] * BoxesNeeded_CYS[i,b];

set <str,str> PerBox_Based_Rate_NoZeroes_CFS;
num PerBox_Based_Rate_CFS {PerBox_Based_Rate_NoZeroes_CFS}; 
read data CASUSER.BOXRATE (where=(RateBasis="PerBox" and Ratetype='Linehaul' and MOL='CFS' and Rate>0)) 
into PerBox_Based_Rate_NoZeroes_CFS=[BOX T2] PerBox_Based_Rate_CFS=Rate;

impvar PerBox_Based_Costs_CFS {w in T2} = sum {b in BOX} PerBox_Based_Rate_CFS[b,w] * BoxesNeeded_CFS[b,w];
impvar TotalBoxBasedCost = sum {i in ISN} PerBox_Based_Costs_CYS[i] + sum {w in T2} PerBox_Based_Costs_CFS[w];

print PerBox_Based_Rate_CFS PerBox_Based_Rate;
/*Box Based Costs END */

/*Vol Based Costs START*/
set <str,str,str,str,str> PerVol_Based_Rate_NoZeroes;
num PerVol_Based_Rate {PerVol_Based_Rate_NoZeroes} init 0; 
read data CASUSER.BOXRATE (where=(RateBasis="PerVolUOM" and Ratetype='Carrier_BOL' and Rate>=0)) 
into PerVol_Based_Rate_NoZeroes=[ORG DES MOL T1 T2] PerVol_Based_Rate=Rate;

impvar VolBasedCost {i in ISN} =Volume_ISN[i]* 
sum {m in MOL} PerVol_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]]*Is_ISN_MOL[i,m] ;


impvar TotalVolBasedCost = sum {i in ISN} VolBasedCost[i];

/*Vol Based Costs END*/

/*Shipment Based Costs START*/
set <str,str,str,str,str> PerShp_Based_Rate_NoZeroes;
num PerShp_Based_Rate {PerShp_Based_Rate_NoZeroes}; 
read data CASUSER.BOXRATE (where=(RateBasis="PerShipment" and Ratetype='Export_Declaration' and Rate>=0)) 
into PerShp_Based_Rate_NoZeroes=[ORG DES MOL T1 T2] PerShp_Based_Rate=Rate;

impvar ShipmentBasedCost {i in ISN} = sum {m in MOL}
PerShp_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]] * Is_ISN_MOL[i,m];

impvar TotalShipmentBasedCost = sum {i in ISN} ShipmentBasedCost[i];

 
print PerBox_Based_Rate PerBox_Based_Rate_CFS PerVol_Based_Rate PerShp_Based_Rate ;
/*Shipment Based Costs END*/
/*Define Rates and ImplicitVariables END*/


Min TotalCost = TotalBoxBasedCost + TotalVolBasedCost + TotalShipmentBasedCost;

/* Constraints START*/

con OneMOL{i in ISN}:
    sum{m in MOL} Is_ISN_MOL[i,m] = 1; /* This is to make sure one ISN can go either CYS or CFS */

con BoxConCYS {i in ISN}:
    sum {b in BOX} Volume_Capacity[b] * BoxesNeeded_CYS[i,b] >= Volume_ISN[i] * Is_ISN_Mol[i,'CYS'];


impvar TotalVol {m in MOL, w in T2} = sum {i in ISN} Volume_ISN[i] * Is_ISN_MOL_T2[i,m,w];


con BoxConCFS {w in T2}: sum {b in BOX} Volume_Capacity_CFS[b,w] * BoxesNeeded_CFS[b,w] >= TotalVol['CFS',w];


solve with milp / decomp=(method=concomp);
expand;
num optCYSCost {ISN};

for {i in ISN} optCYSCost[i] = PerBox_Based_Costs_CYS[i] + VolBasedCost[i] + ShipmentBasedCost[i];
print PerBox_Based_Costs_CYS VolBasedCost ShipmentBasedCost optCYSCost;

   num volShareCFS {i in ISN,w in T2} = Volume_ISN[i] * Is_ISN_MOL[i,'CFS'].sol / TotalVol['CFS',w].sol;

   print Is_ISN_MOL; print Is_ISN_MOL_T2;
   print BoxesNeeded_CYS BoxesNeeded_CFS;
   print PerBox_Based_Costs_CYS PerBox_Based_Costs_CFS;
   print volShareCFS VolBasedCost TotalVolBasedCost TotalVol;
   print ShipmentBasedCost;
quit; 

 

Santha
Pyrite | Level 9

Rob -

Here is the rate file as mentioned in the earlier comment 

Thank you

RobPratt
SAS Super FREQ

Except for BoxRate.xlsx, it looks like you didn't provide the latest data tables you are using.

 

To avoid the duplicate key warnings, make sure the columns that appear within the square brackets [] in the READ DATA statement uniquely identify the observation.

 

The latest code you sent yields errors related to two IMPVAR declarations.  Here are two corrections that should help you get further:

 

   impvar VolBasedCost {i in ISN} = Volume_ISN[i] * 
/*      sum {m in MOL} PerVol_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]]*Is_ISN_MOL[i,m];*/
      sum {<Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]> in PerVol_Based_Rate_NoZeroes} PerVol_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]] * Is_ISN_MOL[i,m];
   impvar ShipmentBasedCost {i in ISN} = 
/*      sum {m in MOL} PerShp_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]] * Is_ISN_MOL[i,m];*/
      sum {<Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]> in PerShp_Based_Rate_NoZeroes} PerShp_Based_Rate[Org_ISN[i],Des_ISN[i],m,T1_ISN[i],T2_ISN[i]] * Is_ISN_MOL[i,m];

 

Santha
Pyrite | Level 9

Rob. 

Thank you. The only thing that is new is BoxRate.xlsx. Everything else is the same data from last time.  I will use the volbasedcost and Shipmentcost you mentioned and check it out. 

 

 

Santha
Pyrite | Level 9

Rob

I have incorporated the  volbased cost and shipment based cost. model runs without errors but the answer is still 173*4= $692.

RobPratt
SAS Super FREQ

Please supply the tables you are using based on the toy example.

Santha
Pyrite | Level 9

Rob

Here are the tables in toy example. will send a couple of more in next one as max allowed is 5.

Also , I mean the full model to just include days and weeks to the three ISNs, 1 week and 2 days that were used in the toy model. 

The only difference between toy and full model is that I want to add all indices and still be able to get to $3,622. 

Santha
Pyrite | Level 9
 
RobPratt
SAS Super FREQ

Here is code to replicate the summary info for the toy example:

   create data SolutionDataCYS from [ISN=i BoxType=b]={i in ISN, b in BOXES: NumBoxesCYS[i,b] > 0.5}
      MOL='CYS' Org=Org_ISN[i] Des=Des_ISN[i] T1=T1_ISN[i] T2=T2_ISN[i] Volume=Volume_ISN[i] Weight=Weight_ISN[i]
      BoxNeeded=NumBoxesCYS Vol_InsideBox=Volume_ISN[i] VolShare=1
      BoxBasedCost=(boxRate[b]*NumBoxesCYS[i,b]) VolBasedCost=0 ShipmentBasedCost=shipmentRate['CYS']
      VolCapacity=(volCapacity[b]*NumBoxesCYS[i,b]);

   create data SolutionDataCFS(where=(VolShare>0)) from [ISN=i BoxType=b]={i in ISN, b in BOXES: NumBoxesCFS[b] > 0.5}
      MOL='CFS' Org=Org_ISN[i] Des=Des_ISN[i] T1=T1_ISN[i] T2=T2_ISN[i] Volume=Volume_ISN[i] Weight=Weight_ISN[i]
      BoxNeeded=(volShareCFS[i]*NumBoxesCFS[b]) Vol_InsideBox=Volume_ISN[i] VolShare=volShareCFS[i]
      BoxBasedCost=(volShareCFS[i]*boxRate[b]*NumBoxesCFS[b]) VolBasedCost[i] ShipmentBasedCost=0
      VolCapacity=(volCapacity[b]*NumBoxesCFS[b]);
quit;

data SolutionData;
   set SolutionDataCYS SolutionDataCFS;
   Total = BoxBasedCost + VolBasedCost + ShipmentBasedCost;
   Utilization = Volume/VolCapacity;
run;

proc print data=SolutionData;
   sum _numeric_;
run;
Obs ISN BoxType MOL Org Des T1 T2 Volume Weight BoxNeeded Vol_InsideBox VolShare BoxBasedCost VolBasedCost ShipmentBasedCost VolCapacity Total Utilization
1 V1 40H CYS SHA ATL D1 W1 130 304 2.00000 130 1.00000 2400.00 0 50 138 2450.00 0.94203
2 V2 40F CFS SHA ATL D1 W1 42 491 0.97674 42 0.97674 976.74 168 0 60 1144.74 0.70000
3 V8 40F CFS SHA ATL D2 W1 1 2 0.02326 1 0.02326 23.26 4 0 60 27.26 0.01667
                173 797 3.00000 173 2.00000 3400.00 172 50 258 3622.00 1.65870
Santha
Pyrite | Level 9

Rob. thank you this is awesome,. I will use this and change it accordingly in my ouptus. the only one thing pending is for the full model (with all the indices for the 3 ISNs in W1 and 2 days (D1 and D2) to get to $3,622 , the true optimum. 

Santha
Pyrite | Level 9

Rob I was able to figure out what was wrong . The TotalVol impvar that i had in my model was not correct. When i corrected it, everything flowed smoothly. Again, thanks a lot for all your support Rob. You are the best. 

I learned a lot , thanks to you. Have a good day

 

sas-innovate-2024.png

Available on demand!

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

 

Register now!

Multiple Linear Regression in SAS

Learn how to run multiple linear regression models with and without interactions, presented by SAS user Alex Chaplin.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 12 replies
  • 4005 views
  • 0 likes
  • 2 in conversation