NLP for CP
Addressing Constraint Programming with Natural Language Processing
Home
Resources
Publications
Correct
predictions are in
blue
. If we detect only a subset of a labelled sentence, we highlight the caught part as
blue
, the missing part
light blue.
False positives
are in
green
and
false negatives
are in
red
.
Problem Trucking — Constraint detection
A
shipping
company
has
a
hub
and
spoke
system
.
The
shipments
to
be
delivered
are
specified
by
an
originating
spoke
,
a
destination
spoke
,
and
a
shipment
volume
.
The
trucks
have
different
types
defined
by
a
maximum
capacity
,
a
speed
,
and
a
cost
per
mile
.
The
model
is
to
assign
the
right
number
of
trucks
to
each
route
in
order
to
minimize
the
cost
of
transshipment
and
meet
the
volume
requirements
.
There
is
a
minimum
departure
time
and
a
maximum
return
time
for
trucks
at
a
spoke
,
and
a
load
and
unload
time
at
the
hub
.
Trucks
of
different
types
travel
in
different
speeds
.
Therefore
,
shipments
are
available
at
each
hub
in
a
timely
manner
.
Volume
availability
constraints
are
considered
,
that
is
,
the
shipments
that
will
be
carried
back
from
a
hub
to
a
spoke
by
a
truck
must
be
available
for
loading
before
the
truck
leaves
.
The
assumptions
are
:
Exactly
the
same
number
of
trucks
that
go
from
spoke
to
hub
return
from
hub
to
spoke
.
Each
truck
arrives
at
a
hub
as
early
as
possible
and
leaves
as
late
as
possible
.
The
shipments
can
be
broken
arbitrarily
into
smaller
packages
and
shipped
through
different
paths
.
Problem Trucking — Detection of the decisions and objects to be modeled
A
shipping
company
has
a
hub
and
spoke
system
.
The
shipments
to
be
delivered
are
specified
by
an
originating
spoke
,
a
destination
spoke
,
and
a
shipment
volume
.
The
trucks
have
different
types
defined
by
a
maximum
capacity
,
a
speed
,
and
a
cost
per
mile
.
The
model
is
to
assign
the
right
number
of
trucks
to
each
route
in
order
to
minimize
the
cost
of
transshipment
and
meet
the
volume
requirements
.
There
is
a
minimum
departure
time
and
a
maximum
return
time
for
trucks
at
a
spoke
,
and
a
load
and
unload
time
at
the
hub
.
Trucks
of
different
types
travel
in
different
speeds
.
Therefore
,
shipments
are
available
at
each
hub
in
a
timely
manner
.
Volume
availability
constraints
are
considered
,
that
is
,
the
shipments
that
will
be
carried
back
from
a
hub
to
a
spoke
by
a
truck
must
be
available
for
loading
before
the
truck
leaves
.
The
assumptions
are
:
Exactly
the
same
number
of
trucks
that
go
from
spoke
to
hub
return
from
hub
to
spoke
.
Each
truck
arrives
at
a
hub
as
early
as
possible
and
leaves
as
late
as
possible
.
The
shipments
can
be
broken
arbitrarily
into
smaller
packages
and
shipped
through
different
paths
.
Back to list