General

The objective is to maximize the expected present value of harvesting in different stands at a forest property.
This analysis is NOT based on the assumption that all relevant future conditions are perfectly known. The optimal sequential harvest decisions based on sequential net price information are calculated. The "classical" deterministic optimal solutions (the optimal deterministic harvest years and the optimal deterministic present values) are also calculated.

How to use this software:

1. First, the user creates a text file (a stand data list) using some standard text editor, such as Wordpad.

One example of such a stand data list is available if you, from the main page, click on "Stand data list (sample file)".
 

n	Stand number
V0	Volume per hectare right now (when t = 0). The usual unit is cubic metre per hectare but any unit can be used as long as the same unit is found everywhere in the list. Example from Sweden: m3sk/hectare
V1	Volume growth per hectare and year (during each year until the final harvest takes place). Units: Compare V0. Example from Sweden: m3sk/hectare/year
P0	Expected (during an average business period) net price per volume unit right now (when t = 0). This is the price of the timber (including pulpwood, fuelwood etc) subtracted by the relevant costs per volume unit. These costs may include different things in different countries and regions. Example from Sweden: SEK/m3sk
P1	The growth of the net price per volume unit and year (during each year until the final harvest takes place). Units: Compare P0. Example from Sweden: SEK/m3sk/year
L	Land value per hectare (of the bare land after the final harvest). Example from Sweden: SEK/hectare
A	Area of the stand. Example from Sweden: hectares
Y0	The first year when harvesting of the stand is possible (because of constraints in the forest act or other reasons specified by the user.)
Ymax	The last year when harvesting of the stand is possible (because of constraints in the forest act or other reasons specified by the user.)
PD	Maximum random deviation of the true net price from the expected net price (each five year period).

2. The user inputs the number of stands in the first box. (Example: 8)

3. The user inputs the rate of interest in the capital market (Example: 3). Note that the rate of interest is defined in the unit % and that all calculations within the software are based on discounting in continuous time. Hence, the discount factor is EXP(-r/100*t) when r is the rate of interest and t is the number of years until some cost or revenue occurs.

4. The user should use the commands "copy" and "paste" in order to move the stand data list from the editor (or from the sample file) to the stand data list window in the ASPRotp1 software. Note: It is very important that the format of the stand data list exactly corresponds to the format of the sample stand data list. Exactly the same number of positions etc. have to be used. There should be no empty spaces or similar deviations from the sample file format. The present version of the software is constrained to a maximum of 20 stands. Most private forest owners have less than 20 stands available. Larger versions of the software can easily be constructed but then the costs of calculations may increase. Such costs have to be covered by the user in one way or another. The present software version can be used for free.

5. The ASP-software ASPRotp1 maximizes the expected present value when the user presses the button "Optimize!". The optimal (deterministic) harvest years and the optimal (deterministic) present values per hectare for each stand are printed. The expected present value if optimal adaptive decisions are used, W, is reported for eacch stand. The lowest price which is sufficiently high to motivate instant harvesting (the reservation price), q, is printed for each stand.

The total expected present value of all stands, for deterministic planning and for adaptive decisions,  are reported.

Note that the expected present value is higher when we make adaptive harvest decisions, taking the true sequential market price information into account, than if we use the "stiff" market insensitive planning approach. This is found in the output of this software.  

The user should be aware that, in some countries, special kinds of restrictions in forest acts, influence the structure of the decision problem. In such cases, the user is suggested to consult the author of this software. Special solution methods may then be needed in order to optimize the total present value of the forest property.
 

An example:

Please look at the stand data list (sample file).

Stand number 6 presently contains 160 cubic metres per hectare (=V0).

The growth is 4.0 cubic metres per hectare and year (= V1). (Note that the unit in the table is 0.1 cubic metres for V1.)

If you harvest the stand right now, the expected net price per cubic metre is 190 SEK (= P0).

Each year, the expected net price increases by 1.0 SEK ( = P1). (Note that the unit in the table is 0.1 SEK per cubic metre and year for P1.)

The value of the bare land is 2000 SEK/hectare. (Note that the unit in the table is 1000 SEK for L.)

The area of the stand is 10 hectares (= A).

You are allowed to harvest the stand no earlier than zero years from now (= Y0) and no later than twenty years from now (=Ymax). (In the adaptive calculations (stochastic dynamic programming), it is however assumed that the Y0 value is zero and that instant harvesting is possible, irrespective of which value you find in the Y0 column.)

The maximum "random" deviation from the expected net price level is 100 SEK (= PD).

In case you use the rate of interest "3" and the number of stands "8" and input the stand data list (sample file), then you will find that the optimal deterministic harvest year of stand 6 is zero years from now (instantly!). The deterministic present value per hectare will be 32 400 SEK in this stand.

The expected present value of the same stand is W = 39 669 SEK per hectare if optimal adaptive decisions are made. Then, the stand should be instantly harvested only in case the true (stochastic) net price offer is higher than 224.81 SEK per cubic metre. In case the net price offer is lower, you should not harvest yet. You should wait longer.

Note that the stands 7 and 8 differ from stand 6 only with respect to the value of PD. The degree of random price variation is lower for those stands, for one reason or another. As a result, the expected present values are lower and the reservation prices are lower.

(If the rate of interest changes, the results also change.)