 |
 |
Optessa MLP can be applied to a wide variety of problems.
These are variously described as planning
or high level scheduling
problems in manufacturing companies.
Optessa
MLP has the ability to generate the
“guaranteed” optimal solution for planning
problems, such as:
- Evaluate and generate an order set that
maximizes capacity utilization or minimizes parts and finished goods
inventory.
- Evaluate and generate an order set that maximizes
a cost function that combines order priorities with capacity
utilization.
- Select an order set / move orders across months to
generate an order set that minimizes the need for capacity adjustment.
- Modify the order counts in an order set so as to
satisfy all capacity and inventory constraints.
- Minimize capacity addition (e.g., overtime) so as
to satisfy all due date / inventory requirements for an order set.
- Evaluate and generate an order set across multiple
weeks to smooth capacity usage.
Following are some examples of problems that Optessa MLP has been
designed to solve.
This
is an important problem for manufacturers of
configured products in volumes. An obvious example is the auto
industry; other industries including power / farm / industrial
equipment, semiconductor face the same problem.
The objective is to select, from a large pool of
actual or
“fixed” orders, a subset of orders
while satisfying
a large number of constraints and required allocations.
Optessa
MLP will maximize
the number of orders
selected.
Given
a demand forecast
and a set of critical
capacity and marketing
constraints, assign the forecast to multiple plants /
lines and to
multiple future time bins. The time
bins can range from days, weeks,
months to years.
The critical capacity constraints can include
material, plant and
labor, MPL,
related constraints from:
- Assembly / finished product plants
- In-house facilities
- Supplier or outsourced facilities
Marketing constraints include:
- Regional and dealer allocations
- Desirable “equipment” or
“feature” mixes
Optessa
MLP will provide an optimal balance between
demand and capacity
utilization at all levels, while minimizing constraint violations.
Forecast order generation, FOG, is a key input
step
to the planning
process in the auto and other industries. The FOG process has to
consider:
- Order templates
- Products and options offered: configuration
rules
- Critical MPL constraints
The number of valid product combinations can be
very large. For
example, some models of cars can have millions of combinations. An
effective FOG
must be able to generate order sets, in the tens or
hundreds of thousands, given the above restrictions.
Optessa
MLP has been demonstrated to be a highly effective FOG tool
with rapid execution times. By using Optessa MLP for FOG, a smaller and
more effective set of forecast orders can be generated as input to the
planning. In comparison, a conventional demand planning system would be
required to generate a considerably larger set of forecast orders to be
equally effective for planning.
|
|
|
|
|
