When MONTE is used, the plan is developed using multi-objective quantitative approach. The first step is to simulate alternative management schedules for stand compartments. The second step is optimisation which combines the information from these simulations with the management objectives of the forest owner. The optimisation process develops a planning model, which is solved using numerical techniques. The simulations form the decision space from which the optimisation picks the single best combination of stands’ treatment schedules.
Technically, the optimisation begins with the selection of objective variables from a list that is displayed to the user. MONTE assumes that all objectives are equally important. However, the weights of the objectives may later be adjusted in the interactive phase of planning. A sub-utility function is defined for every objective based on the minimum, target and maximum values of the objective variables and the relative priorities of these three values. The objective weights and the sub-utility functions result in a utility function which is maximised by one of the following heuristics available in MONTE: Hero, simulated annealing, tabu search, and genetic algorithms. The task of the heuristic algorithm is to find such a combination of stands’ treatment schedules which maximises the utility function derived from the management objectives of the forest landowner.
The values of objective variables in the optimal solution are displayed to the user in a dialog that allows the user to continue the optimisation process in an interactive way. The dialog of interactive optimisation enables the user to change the objective weights as well as the target levels of objectives. Once the most satisfactory solution, in terms of management objectives, has been found, MONTE automatically produces the basic elements of a management plan report (harvested volumes, areas of treatments, development of total growing stock, treatment prescriptions of stands, etc.)
The optimisation is controlled through user-friendly dialogs, making the process easy to the user.