 |
Optimization in Forestry Logistics and Scheduling
Members
J. Christopher Beck
Amir Sepasi
|
Project description
The tactical forest logistics problem of supplying timber to wood mills
requires the integration of inventory management (to store the right
amounts
of the right types of timber at wood mills and forest areas), routing
(to plan the routes of trucks between wood mills and forest areas), and
scheduling (to schedule the use of loaders at wood mills and forest
areas to load and unload the trucks).
One of major challenges of this problem lies in the need to synchronize
the trucks with the loading and unloading equipment available at the
forests
sites and mills. Underlying the need for such integrated optimization is
the inherent uncertainty in the system: both supply and demand are
uncertain in
quality and time; loading and unloading activities are subject to
congestion at the forest site and wood mill; and travel times vary due
to weather and
road conditions. Integrating transportation planning, inventory, and
loader scheduling therefore requires modeling and solving a complex,
integrated,
and stochastic problem. In an environment where distances can be large,
the variability of the driving time can be the same order of magnitude
as the service time. It is thus important to incorporate, as much as
possible, these uncertain aspects into the schedule.
We aim to integrate the approaches developed at École Polytechnique and
University of Toronto to model and solve problems in a real-time
environment where the uncertainty in travel time, service time and
inventory level are taken into account. To achieve this goal we will
pursue research on the following topics:
- A fast re-optimization algorithm, which can be applied in a
real-time environment, in order to provide assistance whenever the
current plan/schedule can no longer be followed due to unforeseen
events.
- Optimization models that take into account the uncertainty
which lies in the scheduling and routing aspects of the problems, such
as loading and unloading service time, machines failures, congestion at
the mills (e.g., due to arrivals of trucks from another company).
Date
October 2011 - February 2012
| |
