[Dip] optimal in doCut and infeasible in B&C&P(doPriceCut)

[Matthew Galati]

It thinks the root node Phase 1 LB is positive - which would mean it is
infeasible.

Send Ted and I the code/data to repeat the issue and one of us will debug
it when we have time.

Thanks,
Matt


On Mon, Jul 2, 2012 at 10:44 AM, Shahin Gelareh <shahin.gelareh at gmail.com>wrote:

> Hi everybody
>
> I have got a problem which is decomposed and uses DIP for testing the
> numerical behaviour under B&C&P(doPriceCut) and doCut.
>
> branch and price log is the folloinwg which declares infeasibility:
> Model core nCols= 35 nRows = 99
> ==  Welcome to the Abstract Library for Parallel Search (ALPS)
> ==  Copyright 2000-2011 Lehigh University and others
> ==  All Rights Reserved.
> ==  Distributed under the Eclipse Public License 1.0
> ==  Version: Trunk (unstable)
> ==  Build Date: Jul  2 2012
> Alps0250I Starting search ...
> Process Node 0 (algo = PRICE_AND_CUT, phaseLast = PHASE_UNKNOWN) gLB =
> -INF gUB =  INF gap =  INF time = 1.560
> phase = PHASE_PRICE1
> Processing Node   0 algo= PRICE_AND_CUT phase= PHASE_PRICE1 c=    0 p=
> 1 LB=     -1.000 UB=      9.000 nodeLB=     -1.000 gLB=     -1.000 gUB=
>    INF lpGap=     10.000 ipGap=        INF time=       1.66
> Processing Node   0 algo= PRICE_AND_CUT phase= PHASE_PRICE1 c=    0 p=
> 2 LB=      0.000 UB=      6.500 nodeLB=      0.000 gLB=      0.000 gUB=
>    INF lpGap=      6.500 ipGap=        INF time=       1.76
> Processing Node   0 algo= PRICE_AND_CUT phase= PHASE_PRICE1 c=    0 p=
> 3 LB=      1.400 UB=      4.800 nodeLB=      1.400 gLB=      1.400 gUB=
>    INF lpGap=      2.429 ipGap=        INF time=       1.85
> Processing Node   0 algo= PRICE_AND_CUT phase= PHASE_PRICE1 c=    0 p=
> 4 LB=      4.000 UB=      4.000 nodeLB=      4.000 gLB=      4.000 gUB=
>    INF lpGap=      0.000 ipGap=        INF time=       1.90
>
> Alps0202I Problem is infeasible. 1 nodes processed, 0 nodes left
> Alps0264I No solution found
> Alps0265I Number of nodes fully processed:          1
> Alps0266I Number of nodes partially processed:      0
> Alps0267I Number of nodes branched:                 0
> Alps0268I Number of nodes pruned before processing: 0
> Alps0270I Number of nodes left:                     0
> Alps0272I Tree depth: 0
> Alps0274I Search CPU time: 0.40 seconds
> Alps0278I Search wall-clock time: 0.40 seconds
>
> ========== DECOMP Statistics [BEGIN]: =========
> Total Decomp          =       0.34    100.00     1  0.34
> Total Solve Relax     =       0.31     90.00    25  0.08
> Total Solve Relax App =       0.00      0.00    25  0.00
> Total Solution Update =       0.00      1.18     4  0.00
> Total Generate Cuts   =       0.00      0.00     0  0.00
> Total Generate Vars   =       0.03      9.12     4  0.02
> Total Compress Cols   =       0.00      0.00     0  0.00
> ========== DECOMP Statistics [END  ]: =========
> Status= 5 BestLB=        -INF BestUB=        INF Nodes=      1 SetupCPU=
> 1.50 SolveCPU= 0.52 TotalCPU= 2.03 SetupReal= 1.50 SolveReal= 0.52
> TotalReal=
>  2.03
>
>
>
> but branch and cut ends to following log:
>
> ....
> Alps0208I Search completed.
> Alps0261I Best solution found had quality -16 and was found at depth 1
> Alps0265I Number of nodes fully processed:          3
> Alps0266I Number of nodes partially processed:      0
> Alps0267I Number of nodes branched:                 1
> Alps0268I Number of nodes pruned before processing: 0
> Alps0270I Number of nodes left:                     0
> Alps0272I Tree depth: 1
> Alps0274I Search CPU time: 0.44 seconds
> Alps0278I Search wall-clock time: 0.44 seconds
>
> ========== DECOMP Statistics [BEGIN]: =========
> Total Decomp          =       0.39    100.00     3  0.29
> Total Solve Relax     =       0.00      0.00     0  0.00
> Total Solve Relax App =       0.00      0.00     0  0.00
> Total Solution Update =       0.01      1.27     7  0.00
> Total Generate Cuts   =       0.12     29.52     5  0.02
> Total Generate Vars   =       0.00      0.00     0  0.00
> Total Compress Cols   =       0.00      0.00     0  0.00
> ========== DECOMP Statistics [END  ]: =========
> Status= 0 BestLB=   -16.00000 BestUB=  -16.00000 Nodes=      3 SetupCPU=
> 1.06 SolveCPU= 0.56 TotalCPU= 1.62 SetupReal= 1.06 SolveReal= 0.56
> TotalReal=
>  1.62
> Optimal Solution
>
>
>
> how to interpret this? the optimal solution of doCut is correct.
> I have difficulties to understand what goes wrong that makes the problem
> infeasible.
>
> I appreciate any comment.
>
> regards,
> Shahin
>
>
>
>
>
>
>
>
>
> _______________________________________________
> Dip mailing list
> Dip at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/dip
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/dip/attachments/20120702/b93f1256/attachment.html>