9.2 Eplex Instances
In this chapter, the problem passed to the external solver will be referred
to as an eplex problem. An eplex problem consists of a set of linear
arithmetic constraints, whose variables have bounds and may possibly have
integrality constraints. The external solver will solve such a problem by optimising
these constraints with respect to an objective function.
With the eplex library, it is possible to have more than one eplex problem
within one program. The simplest way to write such programs with the
library is through Eplex Instances. An eplex instance is an instance
of the eplex solver, to which an eplex problem can be sent. An external
solver state can be associated with each eplex instance, which can be
invoked to solve the eplex problem. Declaratively, an eplex instance can be
seen as a compound constraint consisting of all the variables, their bounds, and
constraints of the eplex problem.
Like other solvers, each eplex instance has its own module. To use an eplex
instance, it must first be declared, so that the module can be created.
This is done by:
This predicate will initialise an eplex instance Name. Once
initialised, a Name module will exist, to which the user can post the
constraints for the eplex problem and setup and use the external solver
state to solve the eplex problem. Normally, this predicate should be issued
as a directive in the user's program, so that the program code can refer to
the instance directly in their code. For example:
For convenience, the eplex library declares eplex as an eplex instance
when the library is loaded.
9.2.1 Linear Constraints
The constraints provided are equalities and inequalities over
Their operational behaviour is as follows:
As with all predicates defined for an eplex instance, these constraints
should be module-qualified with the name of the eplex instance. In the following
they are shown qualified with the eplex instance. Other instances can
be used if they have been declared using eplex_instance/1.
When they contain no variables, they simply succeed or fail.
- When they contain exactly one variable, they are translated into
a bound update on that variable, which may in turn fail, succeed,
or even instantiate the variable.
Note that if the variable's type is integer, the bound will be adjusted
to the next suitable integral value.
- Otherwise, the constraint is transferred to the external solver state
if the state has been setup. If it has not, the constraint
delays and is transferred to the external solver state when it is setup.
This mechanism makes it possible to interface to a non-incremental
black-box solver that requires all constraints at once,
or to send constraints to the solver in batches
X is equal to Y. X and Y are linear expressions.
X is greater or equal to Y. X and Y are linear expressions.
X is less or equal to Y. X and Y are linear expressions.
9.2.2 Linear Expressions
The following arithmetic expression can be used inside the constraints:
Bounds for variables can be given to an eplex instance via the
Variables. If X is not yet a problem variable, it is turned into one
via an implicit declaration X $:: -1.0Inf..1.0Inf.
- 123, 3.4
Integer or floating point constants.
Equivalent to the sum of all list elements.
Scalar product: The sum of the products of the corresponding
elements in the two lists. The lists must be of equal length.
EplexIntance: Vars $:: Lo..Hi
Restrict the external solver to assign solution values for the eplex
problem within the bounds specified by Lo..Hi.
Passes to the external solver the bounds for the variables in Vars.
Lo, Hi are the lower and upper bounds, respectively. Note that the
bounds are only passed to the external solver if they would narrow the
current bounds, and failure will occur if the resulting interval is empty.
Note also that the external solver does not do any bound propagation
and will thus not change the bounds on its own. The default bounds for
variables are notionally -1.0Inf..1.0Inf (where infinity is actually
defined as the solver's notion of infinity).
The difference between using an LP vs. an MIP solver is made by
declaring integrality to the solver via the integers/1 constraint:
Note that all the above constraints are local to the eplex instance; they
do not place any restrictions on the variables for other eplex instances or
solvers. Failure will occur only when inconsistency is detected within the
same eplex instance, unless the user explicitly try to merge the constraints
from different solvers/eplex instance.
Inform the external solver to treat the variables Vars as integral.
It does not impose the integer type on Vars. However, when a
typed_solution is retrieved (via lp_get/3 or
lp_var_get/3), this will be rounded to the nearest integer.
Note that unless eplex:integers/1 (or lp_add/3, see
section 9.4.2) is invoked, any invocation
of the eplex external solver (via lp_solve/2, lp_probe/3 or
lp_demon_setup/5) will only solve a continuous relaxation, even
when problem variables have been declared as integers in other
solvers (e.g. ic).
A counterpart, reals/1 `constraint' also exists – this simply declares the
variables specified are problem variables, and does not actually place any
other constraint on the variables.
9.2.5 Solving Simple Eplex Problems
In order to solve an eplex problem, the eplex instance must be set up
for an external solver state. The solver state can then be invoked to
solve the problem. The simplest way to do this is to use:
Here is a simple linear program, handled by the predefined eplex instance 'eplex':
This predicate creates a new external solver state and associates it
with the eplex instance. Any arithmetic, integrality and bound
constraints posted for this eplex instance are collected to create the
external solver state. After this, the solver state can be invoked to
solve the eplex problem.
Objective is either min(Expr) or max(Expr) where Expr is a linear
expression (or quadratic, if supported by the external solver).
Explicitly invokes the external solver state. Any new constraints
posted are taken into account. If the external solver can find an
optimal solution to the eplex problem, then the predicate succeeds and Cost is
instantiated to the optimal value. If the problem is infeasible (has no
solution) or unbounded (Cost is not bounded by the constraints), then
the predicate fails.
eplex: (X+Y $>= 3),
eplex: (X-Y $= 0),
The same example using a user-defined eplex instance:
my_instance: (X+Y $>= 3),
my_instance: (X-Y $= 0),
Running the program gives the optimal value for Cost:
[eclipse 2]: lp_example(Cost).
Cost = 1.5
Note that if the eplex eplex instance is used instead of my_instance, then the eplex_instance/1 declaration is not
By declaring one variable as integer, we obtain a Mixed Integer Problem:
my_instance: (X+Y $>= 3),
my_instance: (X-Y $= 0),
[eclipse 2]: mip_example(Cost).
Cost = 2.0
The cost is now higher because X is constrained to be an integer. Note also
that in this example, we posted the constraints before setting up the
external solver, whereas in the previous example we set up the solver
first. The solver set up and constraint posting can be done in
any order. If integers/1 constraints are only posted after problem
setup, the problem will be automatically converted from an LP to a MIP
This section has introduced the most basic ways to use the eplex library.
We will discuss more advanced methods of using the eplex instances in