COIN-OR::LEMON - Graph Library

About

Downloads

Documentation

Development

Issue Tracker

Browse Tickets

Custom Query

Roadmap

Browse Source

Local Info

Mailing Lists

Search

Ticket #219: ns-geq-cf49ce98b6a4.patch

File ns-geq-cf49ce98b6a4.patch, 20.0 KB (added by Peter Kovacs, 15 years ago)

  • doc/groups.dox 

    # HG changeset patch
# User Peter Kovacs <kpeter@inf.elte.hu>
# Date 1238030755 -3600
# Node ID cf49ce98b6a4e25e6dfb263baf306d76a837324e
# Parent  c7d160f73d52e2d3f1ab1236cb0b440daf5f7a85
Support <= supply constraints in NetworkSimplex (#219 and #234)

The same inequality constraints are supported as by Circulation.
The documentation of the modules and the classes NetworkSimplex,
Circulation are also improved and extended with important notes and
explanations.

diff --git a/doc/groups.dox b/doc/groups.dox
    a b  
    317317
    318318The \e maximum \e flow \e problem is to find a flow of maximum value between
    319319a single source and a single target. Formally, there is a \f$G=(V,A)\f$
    320 digraph, a \f$cap:A\rightarrow\mathbf{R}^+_0\f$ capacity function and
     320digraph, a \f$cap: A\rightarrow\mathbf{R}^+_0\f$ capacity function and
    321321\f$s, t \in V\f$ source and target nodes.
    322 A maximum flow is an \f$f:A\rightarrow\mathbf{R}^+_0\f$ solution of the
     322A maximum flow is an \f$f: A\rightarrow\mathbf{R}^+_0\f$ solution of the
    323323following optimization problem.
    324324
    325 \f[ \max\sum_{a\in\delta_{out}(s)}f(a) - \sum_{a\in\delta_{in}(s)}f(a) \f]
    326 \f[ \sum_{a\in\delta_{out}(v)} f(a) = \sum_{a\in\delta_{in}(v)} f(a)
    327     \qquad \forall v\in V\setminus\{s,t\} \f]
    328 \f[ 0 \leq f(a) \leq cap(a) \qquad \forall a\in A \f]
     325\f[ \max\sum_{sv\in A} f(sv) - \sum_{vs\in A} f(vs) \f]
     326\f[ \sum_{uv\in A} f(uv) = \sum_{vu\in A} f(vu)
     327    \quad \forall u\in V\setminus\{s,t\} \f]
     328\f[ 0 \leq f(uv) \leq cap(uv) \quad \forall uv\in A \f]
    329329
    330330LEMON contains several algorithms for solving maximum flow problems:
    331331- \ref EdmondsKarp Edmonds-Karp algorithm.
    … …  
    350350
    351351The \e minimum \e cost \e flow \e problem is to find a feasible flow of
    352352minimum total cost from a set of supply nodes to a set of demand nodes
    353 in a network with capacity constraints and arc costs.
     353in a network with capacity constraints (lower and upper bounds)
     354and arc costs.
    354355Formally, let \f$G=(V,A)\f$ be a digraph,
    355356\f$lower, upper: A\rightarrow\mathbf{Z}^+_0\f$ denote the lower and
    356 upper bounds for the flow values on the arcs,
     357upper bounds for the flow values on the arcs, for which
     358\f$0 \leq lower(uv) \leq upper(uv)\f$ holds for all \f$uv\in A\f$.
    357359\f$cost: A\rightarrow\mathbf{Z}^+_0\f$ denotes the cost per unit flow
    358 on the arcs, and
    359 \f$supply: V\rightarrow\mathbf{Z}\f$ denotes the supply/demand values
    360 of the nodes.
    361 A minimum cost flow is an \f$f:A\rightarrow\mathbf{R}^+_0\f$ solution of
    362 the following optimization problem.
     360on the arcs, and \f$sup: V\rightarrow\mathbf{Z}\f$ denotes the
     361signed supply values of the nodes.
     362If \f$sup(u)>0\f$, then \f$u\f$ is a supply node with \f$sup(u)\f$
     363supply, if \f$sup(u)<0\f$, then \f$u\f$ is a demand node with
     364\f$-sup(u)\f$ demand.
     365A minimum cost flow is an \f$f: A\rightarrow\mathbf{Z}^+_0\f$ solution
     366of the following optimization problem.
    363367
    364 \f[ \min\sum_{a\in A} f(a) cost(a) \f]
    365 \f[ \sum_{a\in\delta_{out}(v)} f(a) - \sum_{a\in\delta_{in}(v)} f(a) =
    366     supply(v) \qquad \forall v\in V \f]
    367 \f[ lower(a) \leq f(a) \leq upper(a) \qquad \forall a\in A \f]
     368\f[ \min\sum_{uv\in A} f(uv) \cdot cost(uv) \f]
     369\f[ \sum_{uv\in A} f(uv) - \sum_{vu\in A} f(vu) =
     370    sup(u) \quad \forall u\in V \f]
     371\f[ lower(uv) \leq f(uv) \leq upper(uv) \quad \forall uv\in A \f]
    368372
    369 LEMON contains several algorithms for solving minimum cost flow problems:
    370  - \ref CycleCanceling Cycle-canceling algorithms.
    371  - \ref CapacityScaling Successive shortest path algorithm with optional
     373Note that \f$\sum_{u\in V} sup(u)\f$ must be zero in order to have a
     374feasible solution.
     375
     376NetworkSimplex, which provides the most efficient solution method,
     377supports a more general form of the problem, in which the supply/demand
     378constraints (mass balance constraints) are inequalities
     379(just like the \ref Circulation "network circulation problem").
     380
     381A feasible solution for the minimum cost flow problem can be found
     382using \ref Circulation.
     383
     384LEMON contains several algorithms for solving minimum cost flow problems.
     385 - \ref CycleCanceling Cycle-Canceling algorithms.
     386 - \ref CancelAndTighten The Cancel and Tighten algorithm.
     387 - \ref CapacityScaling Successive Shortest %Path algorithm with optional
    372388   capacity scaling.
    373  - \ref CostScaling Push-relabel and augment-relabel algorithms based on
     389 - \ref CostScaling Push-Relabel and Augment-Relabel algorithms based on
    374390   cost scaling.
    375  - \ref NetworkSimplex Primal network simplex algorithm with various
     391 - \ref NetworkSimplex Primal Network Simplex algorithm with various
    376392   pivot strategies.
     393
     394In general NetworkSimplex is the most efficient implementation,
     395but in special cases other algorithms could be faster.
     396For example, if the capacities are very small, CapacityScaling is
     397usually the fastest algorithm (without effective scaling).
    377398*/
    378399
    379400/**

  • lemon/circulation.h 

    diff --git a/lemon/circulation.h b/lemon/circulation.h
    a b  
    2424
    2525///\ingroup max_flow
    2626///\file
    27 ///\brief Push-relabel algorithm for finding a feasible circulation.
     27///\brief Push-relabel algorithm for the network circulation problem.
    2828///
    2929namespace lemon {
    3030
    … …  
    119119
    120120     The exact formulation of this problem is the following.
    121121     Let \f$G=(V,A)\f$ be a digraph,
    122      \f$lower, upper: A\rightarrow\mathbf{R}^+_0\f$,
    123      \f$delta: V\rightarrow\mathbf{R}\f$. Find a feasible circulation
     122     \f$lower, upper: A\rightarrow\mathbf{R}^+_0\f$ denote the lower and
     123     upper bounds for the flow values on the arcs, for which
     124     \f$0 \leq lower(uv) \leq upper(uv)\f$ holds for all \f$uv\in A\f$, and
     125     \f$delta: V\rightarrow\mathbf{R}\f$ denotes signed lower bounds for the
     126     actual supply of the nodes. Find a feasible circulation
    124127     \f$f: A\rightarrow\mathbf{R}^+_0\f$ so that
    125      \f[ \sum_{a\in\delta_{out}(v)} f(a) - \sum_{a\in\delta_{in}(v)} f(a)
    126      \geq delta(v) \quad \forall v\in V, \f]
    127      \f[ lower(a)\leq f(a) \leq upper(a) \quad \forall a\in A. \f]
    128      \note \f$delta(v)\f$ specifies a lower bound for the supply of node
    129      \f$v\f$. It can be either positive or negative, however note that
    130      \f$\sum_{v\in V}delta(v)\f$ should be zero or negative in order to
    131      have a feasible solution.
    132128
    133      \note A special case of this problem is when
    134      \f$\sum_{v\in V}delta(v) = 0\f$. Then the supply of each node \f$v\f$
    135      will be \e equal \e to \f$delta(v)\f$, if a circulation can be found.
    136      Thus a feasible solution for the
    137      \ref min_cost_flow "minimum cost flow" problem can be calculated
    138      in this way.
     129     \f[ \sum_{uv\in A} f(uv) - \sum_{vu\in A} f(vu)
     130     \geq delta(u) \quad \forall u\in V, \f]
     131     \f[ lower(uv) \leq f(uv) \leq upper(uv) \quad \forall uv\in A. \f]
     132
     133     Note that \f$\sum_{u\in V} delta(u)\f$ must be zero or negative in
     134     order to have a feasible solution (since the sum of the expressions
     135     on the left side of the inequalities are zero).
     136
     137     A special case of this problem is when the sum of the supply
     138     values (\f$\sum_{u\in V} delta(u)\f$) is zero.
     139     Then all the inequalities will be equality, i.e. the supply of each
     140     node \f$u\f$ will be \e equal \e to \f$delta(u)\f$ if a feasible flow
     141     can be found.
     142     Thus a feasible solution for the \ref min_cost_flow
     143     "minimum cost flow problem" can be calculated with this algorithm.
     144
     145     \note If you need the opposite inequality in the supply/demand
     146     constraints (i.e. the total supply is greater than the total demand),
     147     then you could easily transform the problem to this form by reversing
     148     the direction of the arcs and taking the negative of the supply values
     149     (e.g. using \ref ReverseDigraph and \ref NegMap adaptors).
    139150
    140151     \tparam GR The type of the digraph the algorithm runs on.
    141152     \tparam LM The type of the lower bound capacity map. The default

  • lemon/network_simplex.h 

    diff --git a/lemon/network_simplex.h b/lemon/network_simplex.h
    a b  
    3030
    3131#include <lemon/core.h>
    3232#include <lemon/math.h>
     33#include <lemon/circulation.h>
    3334
    3435namespace lemon {
    3536
    … …  
    4243  /// \ref NetworkSimplex implements the primal Network Simplex algorithm
    4344  /// for finding a \ref min_cost_flow "minimum cost flow".
    4445  /// This algorithm is a specialized version of the linear programming
    45   /// simplex method directly for the minimum cost flow problem.
    46   /// It is one of the most efficient solution methods.
     46  /// simplex method directly for the minimum cost flow problem,
     47  /// and it is one of the most efficient solution methods.
    4748  ///
    4849  /// In general this class is the fastest implementation available
    4950  /// in LEMON for the minimum cost flow problem.
    5051  ///
     52  /// In fact, this algorithm supports a more general form of the
     53  /// \ref min_cost_flow "minimum cost flow problem", in which the
     54  /// supply/demand constraints are inequalities, just like the
     55  /// \ref Circulation "network circulation problem".
     56  /// The exact formulation of this problem is the following
     57  /// (using the same notations as in the minimum cost flow problem).
     58  /**
     59      \f[ \min\sum_{uv\in A} f(uv) \cdot cost(uv) \f]
     60      \f[ \sum_{uv\in A} f(uv) - \sum_{vu\in A} f(vu) \geq
     61          sup(u) \quad \forall u\in V \f]
     62      \f[ lower(uv) \leq f(uv) \leq upper(uv) \quad \forall uv\in A \f]
     63  */
     64  /// In this form \f$sup(u)\f$ specifies a lower bound for the actual
     65  /// supply of node \f$u\f$.
     66  /// It can be either positive or negative, however note that
     67  /// \f$\sum_{u\in V} sup(u)\f$ must be zero or negative in order to
     68  /// have a feasible solution (since the sum of the expressions on the
     69  /// left side of the inequalities are zero).
     70  ///
     71  /// A special case of this problem is when the sum of the supply
     72  /// values (\f$\sum_{u\in V} sup(u)\f$) is zero.
     73  /// Then all the inequalities will be equality, i.e. the supply of each
     74  /// node \f$u\f$ will be \e equal \e to \f$sup(u)\f$ if a feasible flow
     75  /// can be found.
     76  /// Thus this special case is exactly the same as the \ref min_cost_flow
     77  /// "minimum cost flow problem".
     78  ///
     79  /// \note If you need the opposite inequality in the supply/demand
     80  /// constraints (i.e. the total supply is greater than the total demand),
     81  /// then you could easily transform the problem to this form by reversing
     82  /// the direction of the arcs and taking the negative of the supply values
     83  /// (e.g. using \ref ReverseDigraph and \ref NegMap adaptors).
     84  ///
    5185  /// \tparam GR The digraph type the algorithm runs on.
    5286  /// \tparam V The value type used in the algorithm.
    5387  /// By default it is \c int.
    … …  
    142176
    143177    // Parameters of the problem
    144178    ValueArcMap *_plower;
    145     ValueArcMap *_pupper;
     179    ValueArcMap *_pcap;
    146180    ValueArcMap *_pcost;
    147181    ValueNodeMap *_psupply;
    148182    bool _pstsup;
    … …  
    584618    /// \param graph The digraph the algorithm runs on.
    585619    NetworkSimplex(const GR& graph) :
    586620      _graph(graph),
    587       _plower(NULL), _pupper(NULL), _pcost(NULL),
     621      _plower(NULL), _pcap(NULL), _pcost(NULL),
    588622      _psupply(NULL), _pstsup(false),
    589623      _flow_map(NULL), _potential_map(NULL),
    590624      _local_flow(false), _local_potential(false),
    … …  
    638672    /// \return <tt>(*this)</tt>
    639673    template<typename UPPER>
    640674    NetworkSimplex& upperMap(const UPPER& map) {
    641       delete _pupper;
    642       _pupper = new ValueArcMap(_graph);
     675      delete _pcap;
     676      _pcap = new ValueArcMap(_graph);
    643677      for (ArcIt a(_graph); a != INVALID; ++a) {
    644         (*_pupper)[a] = map[a];
     678        (*_pcap)[a] = map[a];
    645679      }
    646680      return *this;
    647681    }
    … …  
    851885    /// \return <tt>(*this)</tt>
    852886    NetworkSimplex& reset() {
    853887      delete _plower;
    854       delete _pupper;
     888      delete _pcap;
    855889      delete _pcost;
    856890      delete _psupply;
    857891      _plower = NULL;
    858       _pupper = NULL;
     892      _pcap = NULL;
    859893      _pcost = NULL;
    860894      _psupply = NULL;
    861895      _pstsup = false;
    … …  
    9901024
    9911025      // Initialize node related data
    9921026      bool valid_supply = true;
    993       if (!_pstsup && !_psupply) {
     1027      Value total_supply = 0;
     1028      if (!_psupply && !_pstsup) {
    9941029        _pstsup = true;
    9951030        _psource = _ptarget = NodeIt(_graph);
    9961031        _pstflow = 0;
    9971032      }
    9981033      if (_psupply) {
    999         Value sum = 0;
    10001034        int i = 0;
    10011035        for (NodeIt n(_graph); n != INVALID; ++n, ++i) {
    10021036          _node_id[n] = i;
    10031037          _supply[i] = (*_psupply)[n];
    1004           sum += _supply[i];
     1038          total_supply += _supply[i];
    10051039        }
    1006         valid_supply = (sum == 0);
     1040        valid_supply = (total_supply <= 0);
    10071041      } else {
    10081042        int i = 0;
    10091043        for (NodeIt n(_graph); n != INVALID; ++n, ++i) {
    … …  
    10111045          _supply[i] = 0;
    10121046        }
    10131047        _supply[_node_id[_psource]] =  _pstflow;
    1014         _supply[_node_id[_ptarget]]   = -_pstflow;
     1048        _supply[_node_id[_ptarget]] = -_pstflow;
    10151049      }
    10161050      if (!valid_supply) return false;
    10171051
     1052      const Value max_cap = std::numeric_limits<Value>::max();
     1053      const Value max_cost = std::numeric_limits<Value>::max() / 4;
     1054
     1055      // Run Circulation to check if a feasible solution exists
     1056      typedef ConstMap<Arc, Value> ConstArcMap;
     1057      ValueNodeMap *csup = NULL;
     1058      bool local_csup = false;
     1059      if (_psupply) {
     1060        csup = _psupply;
     1061      } else {
     1062        csup = new ValueNodeMap(_graph, 0);
     1063        (*csup)[_psource] =  _pstflow;
     1064        (*csup)[_ptarget] = -_pstflow;
     1065        local_csup = true;
     1066      }
     1067      bool circ_result = false;
     1068      if (_plower) {
     1069        if (_pcap) {
     1070          Circulation<GR, ValueArcMap, ValueArcMap, ValueNodeMap>
     1071            circ(_graph, *_plower, *_pcap, *csup);
     1072          circ_result = circ.run();
     1073        } else {
     1074          Circulation<GR, ValueArcMap, ConstArcMap, ValueNodeMap>
     1075            circ(_graph, *_plower, ConstArcMap(max_cap), *csup);
     1076          circ_result = circ.run();
     1077        }
     1078      } else {
     1079        if (_pcap) {
     1080          Circulation<GR, ConstArcMap, ValueArcMap, ValueNodeMap>
     1081            circ(_graph, ConstArcMap(0), *_pcap, *csup);
     1082          circ_result = circ.run();
     1083        } else {
     1084          Circulation<GR, ConstArcMap, ConstArcMap, ValueNodeMap>
     1085            circ(_graph, ConstArcMap(0), ConstArcMap(max_cap), *csup);
     1086          circ_result = circ.run();
     1087        }
     1088      }
     1089      if (local_csup) delete csup;
     1090      if (!circ_result) return false;
     1091
    10181092      // Set data for the artificial root node
    10191093      _root = _node_num;
    10201094      _parent[_root] = -1;
    … …  
    10231097      _rev_thread[0] = _root;
    10241098      _succ_num[_root] = all_node_num;
    10251099      _last_succ[_root] = _root - 1;
    1026       _supply[_root] = 0;
     1100      _supply[_root] = -total_supply;
    10271101      _pi[_root] = 0;
    10281102
    10291103      // Store the arcs in a mixed order
    … …  
    10351109      }
    10361110
    10371111      // Initialize arc maps
    1038       if (_pupper && _pcost) {
     1112      if (_pcap && _pcost) {
    10391113        for (int i = 0; i != _arc_num; ++i) {
    10401114          Arc e = _arc_ref[i];
    10411115          _source[i] = _node_id[_graph.source(e)];
    10421116          _target[i] = _node_id[_graph.target(e)];
    1043           _cap[i] = (*_pupper)[e];
     1117          _cap[i] = (*_pcap)[e];
    10441118          _cost[i] = (*_pcost)[e];
    10451119          _flow[i] = 0;
    10461120          _state[i] = STATE_LOWER;
    … …  
    10531127          _flow[i] = 0;
    10541128          _state[i] = STATE_LOWER;
    10551129        }
    1056         if (_pupper) {
     1130        if (_pcap) {
    10571131          for (int i = 0; i != _arc_num; ++i)
    1058             _cap[i] = (*_pupper)[_arc_ref[i]];
     1132            _cap[i] = (*_pcap)[_arc_ref[i]];
    10591133        } else {
    1060           Value val = std::numeric_limits<Value>::max();
     1134          Value val = max_cap;
    10611135          for (int i = 0; i != _arc_num; ++i)
    10621136            _cap[i] = val;
    10631137        }
    … …  
    10831157      }
    10841158
    10851159      // Add artificial arcs and initialize the spanning tree data structure
    1086       Value max_cap = std::numeric_limits<Value>::max();
    1087       Value max_cost = std::numeric_limits<Value>::max() / 4;
    10881160      for (int u = 0, e = _arc_num; u != _node_num; ++u, ++e) {
    10891161        _thread[u] = u + 1;
    10901162        _rev_thread[u + 1] = u;
    … …  
    10951167        _cost[e] = max_cost;
    10961168        _cap[e] = max_cap;
    10971169        _state[e] = STATE_TREE;
    1098         if (_supply[u] >= 0) {
     1170        if (_supply[u] > 0 || (_supply[u] == 0 && total_supply == 0)) {
    10991171          _flow[e] = _supply[u];
    11001172          _forward[u] = true;
    11011173          _pi[u] = -max_cost;
    … …  
    13711443        }
    13721444      }
    13731445
    1374       // Check if the flow amount equals zero on all the artificial arcs
    1375       for (int e = _arc_num; e != _arc_num + _node_num; ++e) {
    1376         if (_flow[e] > 0) return false;
    1377       }
    1378 
    13791446      // Copy flow values to _flow_map
    13801447      if (_plower) {
    13811448        for (int i = 0; i != _arc_num; ++i) {

  • test/min_cost_flow_test.cc 

    diff --git a/test/min_cost_flow_test.cc b/test/min_cost_flow_test.cc
    a b  
    3333
    3434char test_lgf[] =
    3535  "@nodes\n"
    36   "label  sup1 sup2 sup3\n"
    37   "    1    20   27    0\n"
    38   "    2    -4    0    0\n"
    39   "    3     0    0    0\n"
    40   "    4     0    0    0\n"
    41   "    5     9    0    0\n"
    42   "    6    -6    0    0\n"
    43   "    7     0    0    0\n"
    44   "    8     0    0    0\n"
    45   "    9     3    0    0\n"
    46   "   10    -2    0    0\n"
    47   "   11     0    0    0\n"
    48   "   12   -20  -27    0\n"
     36  "label  sup1 sup2 sup3 sup4\n"
     37  "    1    20   27    0   20\n"
     38  "    2    -4    0    0   -8\n"
     39  "    3     0    0    0    0\n"
     40  "    4     0    0    0    0\n"
     41  "    5     9    0    0    6\n"
     42  "    6    -6    0    0   -5\n"
     43  "    7     0    0    0    0\n"
     44  "    8     0    0    0    0\n"
     45  "    9     3    0    0    0\n"
     46  "   10    -2    0    0   -7\n"
     47  "   11     0    0    0  -10\n"
     48  "   12   -20  -27    0  -30\n"
    4949  "\n"
    5050  "@arcs\n"
    5151  "       cost  cap low1 low2\n"
    … …  
    141141template < typename GR, typename LM, typename UM,
    142142           typename SM, typename FM >
    143143bool checkFlow( const GR& gr, const LM& lower, const UM& upper,
    144                 const SM& supply, const FM& flow )
     144                const SM& supply, const FM& flow, bool geq = false )
    145145{
    146146  TEMPLATE_DIGRAPH_TYPEDEFS(GR);
    147147
    … …  
    155155      sum += flow[e];
    156156    for (InArcIt e(gr, n); e != INVALID; ++e)
    157157      sum -= flow[e];
    158     if (sum != supply[n]) return false;
     158    if ( (!geq && sum != supply[n]) || (geq && sum < supply[n]) )
     159      return false;
    159160  }
    160161
    161162  return true;
    … …  
    189190               const GR& gr, const LM& lower, const UM& upper,
    190191               const CM& cost, const SM& supply,
    191192               bool result, typename CM::Value total,
    192                const std::string &test_id = "" )
     193               const std::string &test_id = "", bool geq = false )
    193194{
    194195  check(mcf_result == result, "Wrong result " + test_id);
    195196  if (result) {
    196     check(checkFlow(gr, lower, upper, supply, mcf.flowMap()),
     197    check(checkFlow(gr, lower, upper, supply, mcf.flowMap(), geq),
    197198          "The flow is not feasible " + test_id);
    198199    check(mcf.totalCost() == total, "The flow is not optimal " + test_id);
    199200    check(checkPotential(gr, lower, upper, cost, mcf.flowMap(),
    … …  
    224225  // Read the test digraph
    225226  Digraph gr;
    226227  Digraph::ArcMap<int> c(gr), l1(gr), l2(gr), u(gr);
    227   Digraph::NodeMap<int> s1(gr), s2(gr), s3(gr);
     228  Digraph::NodeMap<int> s1(gr), s2(gr), s3(gr), s4(gr);
    228229  ConstMap<Arc, int> cc(1), cu(std::numeric_limits<int>::max());
    229230  Node v, w;
    230231
    … …  
    237238    .nodeMap("sup1", s1)
    238239    .nodeMap("sup2", s2)
    239240    .nodeMap("sup3", s3)
     241    .nodeMap("sup4", s4)
    240242    .node("source", v)
    241243    .node("target", w)
    242244    .run();
    … …  
    245247  {
    246248    NetworkSimplex<Digraph> mcf(gr);
    247249
     250    // Check the equality form
    248251    mcf.upperMap(u).costMap(c);
    249252    checkMcf(mcf, mcf.supplyMap(s1).run(),
    250253             gr, l1, u, c, s1, true,  5240, "#A1");
    … …  
    265268             gr, l1, cu, cc, s3, true,   0, "#A7");
    266269    checkMcf(mcf, mcf.boundMaps(l2, u).run(),
    267270             gr, l2, u, cc, s3, false,   0, "#A8");
     271
     272    // Check the inequality form
     273    mcf.reset().upperMap(u).costMap(c).supplyMap(s4);
     274    checkMcf(mcf, mcf.run(),
     275             gr, l1, u, c, s4, true,  3530, "#A9", true);
     276    checkMcf(mcf, mcf.lowerMap(l2).run(),
     277             gr, l2, u, c, s4, true,  4540, "#A10", true);
    268278  }
    269279
    270280  // B. Test NetworkSimplex with each pivot rule