Decision Optimization

Decision Optimization

Delivers prescriptive analytics capabilities and decision intelligence to improve decision-making.


#Analytics
#DecisionOptimization
#DecisionOptimization
 View Only
  • 1.  Unknown OPL error halts CPLEX, version 12.2.0.2 -b13

    Posted 03/23/11 03:58 AM

    Originally posted by: Kagan


    I have the new CPLEX Optimization Studio version 12.2.0.2 -b13 running on Windows 7 Professional x64, with 6GB of RAM. I am trying to run the attached .mod file, which runs fine in version 12.1 (OPL version 6.3), but I am getting the following engine log,
    Tried aggregator 2 times.
    MIP Presolve eliminated 85 rows and 1 columns.
    MIP Presolve modified 10 coefficients.
    Aggregator did 17 substitutions.
    Reduced MIP has 1831 rows, 1356 columns, and 13700 nonzeros.
    Reduced MIP has 615 binaries, 0 generals, 0 SOSs, and 0 indicators.
    Probing fixed 0 vars, tightened 12 bounds.
    Probing time = 0.03 sec.
    Tried aggregator 1 time.
    Presolve time = 0.06 sec.
    Probing time = 0.00 sec.
    Clique table members: 2506.
    MIP emphasis: balance optimality and feasibility.
    MIP search method: dynamic search.
    Parallel mode: deterministic, using up to 4 threads.
    Root relaxation solution time = 0.09 sec.

    Nodes Cuts/
    Node Left Objective IInf Best Integer Best Node ItCnt Gap
    0+ 0 0.0000 1485 ---
    0 0 1.7544 118 0.0000 1.7544 1485 ---
    0 0 1.7544 93 0.0000 Cuts: 47 1587 ---
    0 0 1.7544 155 0.0000 Cuts: 111 1801 ---
    0 0 1.7544 123 0.0000 Cuts: 10 1901 ---
    0 0 1.7544 109 0.0000 Cuts: 38 1975 ---
    0 0 1.7544 135 0.0000 Cuts: 54 2051 ---
    0 2 1.7544 123 0.0000 1.7544 2051 ---
    Elapsed real time = 1.45 sec. (tree size = 0.01 MB, solutions = 1)
    30 32 1.5383 65 0.0000 1.7544 8642 ---
    76 76 1.5998 70 0.0000 1.7544 13754 ---
    240 194 1.7544 117 0.0000 1.7544 25345 ---
    353 266 1.7543 112 0.0000 1.7544 31730 ---
    543 426 0.8887 73 0.0000 1.7544 39061 ---
    548 431 1.7544 135 0.0000 1.7544 61046 ---
    551 432 1.7544 154 0.0000 1.7544 61904 ---
    557 436 1.7544 251 0.0000 1.7544 68526 ---
    561 439 1.7544 333 0.0000 1.7544 73363 ---
    570 418 1.7544 247 0.0000 1.7544 81920 ---
    Elapsed real time = 29.98 sec. (tree size = 8.41 MB, solutions = 1)
    631 274 1.7544 261 0.0000 1.7544 90532 ---
    732 216 1.7544 247 0.0000 1.7544 103177 ---
    782 202 1.7544 261 0.0000 1.7544 113983 ---
    850 198 1.7544 283 0.0000 1.7544 134723 ---
    913 204 1.7544 280 0.0000 1.7544 145253 ---
    1030 284 1.7544 298 0.0000 1.7544 167275 ---
    1132 358 1.7544 224 0.0000 1.7544 186906 ---
    1281 505 1.7544 289 0.0000 1.7544 198000 ---
    1441 615 1.7544 281 0.0000 1.7544 218525 ---
    1690 823 1.7544 286 0.0000 1.7544 242802 ---
    Elapsed real time = 71.84 sec. (tree size = 2.43 MB, solutions = 36)
    2113 1220 1.7544 241 0.0000 1.7544 283954 ---
    2178 1271 1.7544 240 0.0000 1.7544 321718 ---
    2225 1316 1.7544 245 0.0000 1.7544 341253 ---
    2347 1421 1.7544 255 0.0000 1.7544 375297 ---
    2633 1699 1.7544 289 0.0000 1.7544 413471 ---
    2837 1891 1.7544 259 0.0000 1.7544 430832 ---
    2841 1895 1.7544 260 0.0000 1.7544 438507 ---
    3193 2209 1.7544 242 0.0000 1.7544 477581 ---
    3313 2317 1.7544 237 0.0000 1.7544 504852 ---

    There may be further error information in the clone logs.

    Clique cuts applied: 32
    Implied bound cuts applied: 189
    Flow cuts applied: 8
    Mixed integer rounding cuts applied: 20
    Gomory fractional cuts applied: 11

    Root node processing (before b&c):
    Real time = 1.39
    Parallel b&c, 4 threads:
    Real time = 100.70
    Sync time (average) = 14.11
    Wait time (average) = 0.00


    Total (root+branch&cut) = 102.09 sec.

    with the following problem.
    Description Resource Path Location Type
    fl in EPeriods; dvar float+ finNodes; dvar float+ foutNodesPeriods; dvar float+ qNodesPeriods; dvar float+ m; maximize m-0.000001*sum(i in Nodes, t in Periods) q[i][t]; subject to { sum(i in Nodes)y[i]==1; forall (l in E,t in Periods) linkcapacity: f[l][t] <=l.c/T*x[l][t]; forall (i in Nodes) inputtraffic: fin[i]==p[i]*m/T; forall(i in Nodes, t in Periods) outputlimit: fout[i][t]<=a/T*y[i]; forall(i in Nodes,t in 2..T) flowbalance: q[i]t-1fin[i] sum(<s,i,c> in E) f<s,i,c>[t]==fout[i][t]+sum(<i,d,c> in E) f<i,d,c>[t]+q[i][t]; forall(i in Nodes) flowbalanceend: q[i][T]+fin[i]+ sum(<s,i,c> in E) f<s,i,c>[1]==fout[i][1]+sum(<i,d,c> in E) f<i,d,c>[1]+q[i][1]; forall(i in Nodes, t in Periods) queuecapacity: q[i][t]<=beta[i]; forall(i in Nodes) singleroute: sum(<i,d,c> in E) z<i,d,c>== 1-y[i]; forall(t in Periods, l in E) singleroute2: x[l][t]<=z[l]; forall (i in Nodes, t in Periods) singletask: sum(<s,i,c> in E) x<s,i,c>[t]+sum(<i,d, WMN3 Unknown OPL Problem Marker
    )

    Is there a fix for this problem?
    #DecisionOptimization
    #OPLusingCPLEXOptimizer


  • 2.  Re: Unknown OPL error halts CPLEX, version 12.2.0.2 -b13

    Posted 03/25/11 01:08 AM

    Originally posted by: SystemAdmin


    Using "dummy" data, I don't reproduce the error you reported. Could you either attach a reduced data file that reproduces the issue, or if your data is confidential , contact IBM Support so that they can have a look at the issue?
    #DecisionOptimization
    #OPLusingCPLEXOptimizer


  • 3.  Re: Unknown OPL error halts CPLEX, version 12.2.0.2 -b13

    Posted 03/26/11 02:20 PM

    Originally posted by: Kagan


    Thanks for your answer. I am attaching the data file that goes along with it.
    By the way, even though the latest update was supposed to fix the memory problems, apparently it did not. For those models, I am lucky to have OPL 6.3 on some of my computers...
    #DecisionOptimization
    #OPLusingCPLEXOptimizer