[Coin-symphony] Issue with find_first_feasible

[Paul Pacheco]

When using gap_limit 1.0, I am also seeing something interesting:
The Lower bound is greater than the upper bound which is theoretically 
imposible.

I'll send the problem in ampl as soon as I change the name of the variables 
(my boss requested this).


*******************************************************
*   This is SYMPHONY Version 5.1alpha                 *
*   Copyright 2000-2005 Ted Ralphs                    *
*   All Rights Reserved.                              *
*   Distributed under the Common Public License 1.0   *
*******************************************************

SYMPHONY was called with the following arguments:
/mnt/sfw/symphony/symphony_m_tm
-f params
-F bill.mod
-D small/0.dat

============= Other Parameter Settings =============

# amount of processors in the net
max_active_nodes 10


#time_limit 300

#find_first_feasible 1

#diving_strategy 2

do_reduced_cost_fixing 1

#keep_warm_start 1
#logging 1

#tree_log_file_name tree.log
#cut_log_file_name cut.log


#generate_cgl_rounding_cuts 1

gap_limit 1.0

#do_logical_fixing 1

#do_primal_heuristic 1
====================================================

Reading model section from bill.mod...
283 lines were read
Reading data section from small/0.dat...
471 lines were read
Generating OnePlanPerUser...
Generating CalcPeakOverageMinutes...
Generating CalcDCOverageMinutes...
Generating RestrictNightstoPlan...
Generating CalculateGroupOverage...
Generating CalculateDCGroupOverage...
Generating cost...
Model has been successfully generated
[t401a1] BEGIN
[t401a2] BEGIN
[t280087] BEGIN
[t280088] BEGIN
[t800fb] BEGIN
[t800fc] BEGIN
[t800fd] BEGIN
[t3c0025] BEGIN
[t3c0026] BEGIN
[t3c0027] BEGIN
[t800fc]
[t800fc] ****** Found Better Feasible Solution !
[t800fc] ****** Cost: 91785.000000
[t800fc]
[t3c0026] EOF
[t800fc] EOF
[t800fb] EOF
[t800fd] EOF
[t401a1] EOF
[t280088] EOF

****************************************************
* Target Gap Achieved                              *
* Now displaying stats and best solution found...  *
****************************************************

====================== Misc Timing =========================
  Problem IO        0.098
======================= CP Timing ===========================
  Cut Pool                  0.000
====================== LP/CG Timing =========================
  LP Solution Time          1.550
  Variable Fixing           0.005
  Pricing                   0.000
  Strong Branching          4.919
  Separation                1.148
=================== Parallel Overhead ======================
  Communication         0.101
  Ramp Up Time (TM)     1.741
  Ramp Up Time (LP)     2.944
  Ramp Down Time        0.019
  Idle Time (Node Pack) 0.020
  Idle Time (Nodes)     0.367
  Idle Time (Names)     0.161
  Idle Time (Diving)    2.350
  Idle Time (Cuts)      0.000
  Total User Time              7.821
  Total Wallclock Time         22.973

====================== Statistics =========================
Number of created nodes :       819
Number of analyzed nodes:       419
Depth of tree:                  40
Size of the tree:               819
Number of Chains:               70
Number of Diving Halts:         70
Number of cuts in cut pool:     0
Lower Bound in Root:            91422.088

Current Upper Bound:         91785.000
Current Lower Bound:         91791.000
Gap Percentage:              -0.01

No Solution Found