====== The space dimension in the decision LP matrix ======
{{model:agentsbehaviour:dtm_30m_valleesina.png|DTM Colli Esini}}
A great difference with AgriPoliS is that the MIP is spatially explicit. So it will be composed of the following activities:
* n spacial explicit activities * m plot numbers
* i spacial explicit investments * j_i suitable plots
* z other spacial implicit activities
* q other spacial implicit investments
The spacial explicitness has the disadvantage of a much larger matrix, especially in case of farmers with many plots, but it has the following advantages:
* transport costs are associated directly with the individual pixels and already included in the Gross Margin;
* pixel heterogeneity (e.g. different slopes) can be included straightforward, just changing the gross margins;
* an explicit map of the region can be draw to show the different productions
* binds on individual pixels can be easily inserted on the map (e.g. if the plot_4 is already employed for perennial crops, the matrix will not have the activity option "maize on plot_4" till the perennial crop investment will dead).
An example of matrix following such approach is given:
^ Resources ^ ^Prod. activity 1\\ on plot1 ^Prod. activity 1\\ on plot2 ^Prod. activity 1\\ on plot3 ^Prod. activity 2\\ on plot1 ^Prod. activity 2\\ on plot2 ^Prod. activity 2\\ on plot3 ^Inv. activity 3\\ on plot3 ^ ^Used recourses ^Available resources ^
| Gross margin | |300 |295 |305 |60 |58 |62 |140 | | | |
| ... | | | | | | | | | | | |
| Plot1 | |1 | | |1 | | | | |0 |1 |
| Plot2 | | |1 | | |1 | | | |0 |1 |
| Plot3 | | | |1 | | |1 |1 | |0 |1 |
| ... | | | | | | | | | | | |
| Other resources | |... |... |... |... |... |... |... | |... |... |
| ... | | | | | | | | | | | |
| Quantities | |0 |0 |0 |0 |0 |0 |0 | | | |
Gross margin already take into account that plot1 is closer than plot2 and that plot3 is even more close to the farm.\\
If the farmer decide to invest activity3 //(a perennial crop)// on plot3, the following years, "Prod. activity 1 on plot3" and "Prod. activity 2 on plot3" will obviously not be available.
This appoach let us fully differentiate between plots, e.g. on their elevation or even on their slope or exposure.