ClassDescriptionAn Assignment is a variable -> domains mapping, used
to report solutions to the user.This is the base class for building an Lns operator.A BaseObject is the root of all reversibly allocated objects.
A DebugString method and the associated << operator are implemented
as a convenience.Cast constraints are special channeling constraints designed
to keep a variable in sync with an expression.Defines operators which change the value of variables;
each neighbor corresponds to *one* modified variable.
Sub-classes have to define ModifyValue which determines what the new
variable value is going to be (given the current value and the variable).A constraint is the main modeling object.Solver parameters.Solver parameters.Internal parameters of the solver.Statistics on the search in the constraint solver.Statistics on the search in the constraint solver.A Decision represents a choice point in the search tree.A DecisionBuilder is responsible for creating the search tree.A DecisionVisitor is used to inspect a decision.
It contains virtual methods for all type of 'declared' decisions.This struct holds all parameters for the default search.
DefaultPhaseParameters is only used by Solver::MakeDefaultPhase methods.
Note this is for advanced users only.A Demon is the base element of a propagation queue.First solution strategies, used as starting point of local search.First solution strategies, used as starting point of local search.Protobuf enum
operations_research.FirstSolutionStrategy.ValueGlobalVehicleBreaksConstraint ensures breaks constraints are enforced on
all vehicles in the dimension passed to its constructor.
It is intended to be used for dimensions representing time.
A break constraint ensures break intervals fit on the route of a vehicle.
For a given vehicle, it forces break intervals to be disjoint from visit
intervals, where visit intervals start at CumulVar(node) and last for
node_visit_transit[node].Interval variables are often used in scheduling.The class IntExpr is the base of all integer expressions in
It contains the basic protocol for an expression:
- setting and modifying its bound
- querying if it is bound
- listening to events modifying its bounds
- casting it into a variable (instance of IntVar)Represents a function that accepts two int-valued arguments and produces a long-valued result.The class IntVar is a subset of IntExpr.The class Iterator has two direct subclasses.Base operator class for operators manipulating variables.This class acts as a intermediate step between a c++ decision builder and a java one.Classes to which this template function can be applied to as of 04/2014.
Usage: LocalSearchOperator* op = MakeLocalSearchOperator<Relocate>(...);
Local Search Filters are used for fast neighbor pruning.
Filtering a move is done in several phases:
- in the Relax phase, filters determine which parts of their internals
will be changed by the candidate, and modify intermediary State
- in the Accept phase, filters check that the candidate is feasible,
- if the Accept phase succeeds, the solver may decide to trigger a
Synchronize phase that makes filters change their internal representation
to the last candidate,
- otherwise (Accept fails or the solver does not want to synchronize),
a Revert phase makes filters erase any intermediary State generated by the
Relax and Accept phases.
A given filter has phases called with the following pattern:
(Relax.Accept.Synchronize | Relax.Accept.Revert | Relax.Revert)*.
Filters's Revert() is always called in the reverse order their Accept() was
called, to allow late filters to use state done/undone by early filters'
Accept()/Revert().Filter manager: when a move is made, filters are executed to decide whether
the solution is feasible and compute parts of the new cost.Local search metaheuristics used to guide the search.Local search metaheuristics used to guide the search.Protobuf enum
operations_research.LocalSearchMetaheuristic.ValueThis class represent a reversible FIFO structure.
The main difference w.r.t a standard FIFO structure is that a Solver is
given as parameter to the modifiers such that the solver can store the
Iterator's traversing order should not be changed, as some algorithm
depend on it to be consistent.
It's main use is to store a list of demons in the various classes of
The base class for all local search operators.
A local search operator is an object that defines the neighborhood of a
solution.Statistics on local search.Statistics on local search.First solution statistics collected during search.First solution statistics collected during search.Statistics on local search filters called during the search.Statistics on local search filters called during the search.Statistics on local search operators called during the search.Statistics on local search operators called during the search.Represents an operation upon three
long-valued operands and producing a
long-valued result.Represents a predicate (boolean-valued function) uppon three
long-valued operands.Implements a complete cache for model elements: expressions and
constraints.Model visitor.This class encapsulates an objective.Base class of the local search operators dedicated to path modifications
(a path is a set of nodes linked together by arcs).
This family of neighborhoods supposes they are handling next variables
representing the arcs (var[i] represents the node immediately after i on
Several services are provided:
- arc manipulators (SetNext(), ReverseChain(), MoveChain())
- path inspectors (Next(), Prev(), IsPathEnd())
- path iterators: operators need a given number of nodes to define a
neighbor; this class provides the iteration on a given number of (base)
nodes which can be used to define a neighbor (through the BaseNode method)
Subclasses only need to override MakeNeighbor to create neighbors using
the services above (no direct manipulation of assignments).NOLINT
The PropagationBaseObject is a subclass of BaseObject that is also
friend to the Solver class.Usual limit based on wall_time, number of explored branches and
number of failures in the search treeA search limit The default values for int64 fields is the maxima value, i.e., 2^63-1A search limit The default values for int64 fields is the maxima value, i.e., 2^63-1This class adds reversibility to a POD type.
It contains the stamp optimization.This class adds reversibility to a POD type.
It contains the stamp optimization.This class adds reversibility to a POD type.
It contains the stamp optimization.----- RevPartialSequence -----Dimensions represent quantities accumulated at nodes along the routes.Manager for any NodeIndex <-> variable index conversion.Struct used to sort and store vehicles by their type.Parameters which have to be set when creating a RoutingModel.Parameters which have to be set when creating a RoutingModel.Routing model visitor.Parameters defining the search used to solve vehicle routing problems.Parameters defining the search used to solve vehicle routing problems.Parameters required for the improvement search limit.Parameters required for the improvement search limit.Local search neighborhood operators used to build a solutions neighborhood.Local search neighborhood operators used to build a solutions neighborhood.Underlying solver to use in dimension scheduling, respectively for continuous and mixed models.Base class of all search limits.The base class of all search logs that periodically outputs information when
the search is running.A search monitor is a simple set of callbacks to monitor all search eventsSearch statistics.Search statistics.A sequence variable is a variable whose domain is a set of possible
orderings of the interval variables.The SequenceVarElement stores a partial representation of ranked
interval variables in the underlying sequence variable.
This representation consists of three vectors:
- the forward sequence.Base operator class for operators manipulating variables.This class is the root class of all solution collectors.
It implements a basic query API to be used independently
of the collector used.This class is used to manage a pool of solutions.Solver Class
A solver represents the main computation engine.This exceptions signal that a failure has been raised in the C++ world.Holds semantic information stating that the 'expression' has been
cast into 'variable' using the Var() method, and that
'maintainer' is responsible for maintaining the equality between
'variable' and 'expression'.SWIGTYPE_p_std__vectorT_std__setT_operations_research__RoutingModel__VehicleTypeContainer__VehicleClassEntry_t_tA symmetry breaker is an object that will visit a decision and
create the 'symmetrical' decision in return.
Each symmetry breaker represents one class of symmetry.Checker for type incompatibilities.The following constraint ensures that incompatibilities and requirements
between types are respected.
It verifies both "hard" and "temporal" incompatibilities.
Two nodes with hard incompatible types cannot be served by the same vehicle
at all, while with a temporal incompatibility they can't be on the same
route at the same time.
The VisitTypePolicy of a node determines how visiting it impacts the type
count on the route.
For example, for
- three temporally incompatible types T1 T2 and T3
- 2 pairs of nodes a1/r1 and a2/r2 of type T1 and T2 respectively, with
- a1 and a2 of VisitTypePolicy TYPE_ADDED_TO_VEHICLE
- r1 and r2 of policy ADDED_TYPE_REMOVED_FROM_VEHICLE
- 3 nodes A, UV and AR of type T3, respectively with type policies
TYPE_ADDED_TO_VEHICLE, TYPE_ON_VEHICLE_UP_TO_VISIT and
UV --> a1 --> r1 --> a2 --> r2, a1 --> r1 --> a2 --> r2 --> A and
a1 --> r1 --> AR --> a2 --> r2 are acceptable, whereas the configurations
a1 --> a2 --> r1 --> ..., or A --> a1 --> r1 --> ..., or
a1 --> r1 --> UV --> ...Checker for type requirements.