|
 Printer Friendly |
Rollover analyses using Woodstock and Stanley
In a rollover analysis, you are simulating the process of creating, implementing and updating or adapting a management plan based on the outcomes of previous periods' activities. The management schedule is periodically updated to allow the model to respond to the outcomes of management and forest growth in previous periods. In effect, you are creating a series of management schedules that follow a particular management strategy – in terms of forest dynamics, actions and outcomes, goals and objectives – that has been periodically updated to account for the decisions made in the previous plans.
To conduct a rollover analysis, you need to decide (a) how far into the future you wish to develop a schedule and (b) how frequently the schedule should be updated – e.g., after every period, after every second period, etc. Combined, these two factors determine how many times the software must be run. For example, to project a model that uses 5-year planning periods over 100 years – 20 periods – with updates after every period, would require 20 model runs. If updates occur every 5th period, then only 4 model runs are necessary to project over the 20 periods.
Rollovers incorporate both schedule playback and optimization; the activity sequence is played-back until the rollover period, at which point a new management schedule is generated. Using the example above, an optimal management schedule is generated for periods 1–20, the first period in the schedule is played-back and the outcomes realized and then a new management schedule is generated for periods 2–21. This is done 20 times in succession to produce updated schedules for every period in the planning horizon, or 4 times – in periods 5, 10, 15 and 20 – to produce updated schedules every 5th period.
You can use rollovers to test the robustness of management strategies and assumptions, to test how the system responds to unplanned-for events such as catastrophic disturbance, to project long-term forest change spatially and to demonstrate the range of effects that management policies can have on future forest patterns and composition.
Ugo Feunekes, Vice President, Remsoft Inc., will be presenting a paper titled 'Rolling Planning Horizons & Uncertainty in the Forest' at the INFORMS conference in Miami, November 4th-7th 2001.
|
|
Stora Enso Predicts a 2.5% Savings
“We asked, ‘If we had made the decision and optimized the problems, how much money would we have saved?’” The answer was approximately 2.5 percent.”
- Continue reading...
- View all case studies...
|
 |
|
|
|
