The Treatment Optimization Model (TOM) calculations rely on the Minimum Travel Time (MTT) calculations to identify major fire travel routes and attempt to block them efficiently with fuel treatments. For the target weather conditions, this translates to a specific optimization objective to find the specific treatment areas that reduce fire growth rates the greatest for the given treatment amount. In other words, it is attempting to maximize the minimum travel time for fire moving across the landscape. There are a number of assumptions of this analysis that are discussed in the Technical Topics.
TOM can be very computer intensive and can take a long time to complete runs.
The landscape resolution is the most significant way to control the computational intensity (i.e. run time) of your simulation. It is best to start with a "coarse" landscape. For example, doubling the resolution from 30m to 60m would decrease run time by approximately 75% since there would be only 25% of the nodes to preform calculations on. However setting the resolution greater than the Landscape (.LCP) File cell size will cause some of the cells to be "skipped over" for the calculations.
The Resolution distance units (meters, feet, or kilometers) are the Grid Distance Units: of the landscape file.
To change the landscape resolution select the Utilities > Resample Landscape command and create a new landscape file.
Before getting started, you will need to identify the appropriate landscape size and extent that constitutes a “fire shed” or the spatial domain where large fires can spread under your “target” weather conditions. Your fire shed can be the entire landscape or a portion you define as a separate Analysis Area. Information for this decision comes from local fire records and “fire atlases” or neighboring areas that depict the location and extent of past large fires. You will need weather records (fuel moistures, wind speed and direction) for those types of large fires that you consider to be a “problem” for fire management or for ecological and other management impacts. This gives you the basic information on the fire scenario and fire behavior for treated and untreated fuel types. That is, you can see how your stand-level fuel treatment prescriptions change the fire behavior.
TOM runs utilize all the same fuel moisture, wind, custom fuel model, etc Inputs Tab parameters as other types of FlamMap runs. Wind and Fuel Moisture Conditioning Period options are also appropriate for TOM runs. TOM will work best if the wind direction is from one of the four major cardinal directions (N,E,S,W).
Remember that several settings on the Minimum Travel Time tab will affect TOM modeling runs, Spot Probability, Spotting Delay, and the Vertical/Lateral Search Depths.
This is a line ignition located at the upwind edge of a landscape that defines the location where a large fire spreads from (winds are entered on the Inputs tab). For example, if winds are specified as blowing from the south, the ignition would extend across the southern edge of the landscape.
WARNING: If the ignition is not perpendicular to the wind direction or not on the "up wind" side of the landscape strange results can be expected.
For a line ignition source you can define current ignitions on the display pane or utilize an existing shapefile created outside of FlamMap. Typically for a TOM line ignition you would use just two points, beginning and ending at the corners of the landscape. To create a line ignition in FlamMap use the Create Lines 
toolbar button to locate the ignition before you open the "Runs" dialog.
Click the 
button to display three Ignition File options,
The Resolution of Calculations is controlled by the landscape file resolution.
To change the landscape resolution select the Utilities > Resample Landscape command and create a new landscape file.
This is a complete second landscape that depicts the treatment characteristics (surface fuels, canopy fuels etc.) everywhere that treatments can be placed. In other words, you must create a new landscape that contains changes to fuels (that reflect treatment prescriptions) in every polygon or pixel where it is possible to do treatments. TOM will pick those areas for treatment where treatments change fire behavior along major pathways. TOM will only choose treatment locations that reduce the rate of spread – thus, TOM will not locate fuel treatments where fire behavior increases with treatment or the ideal and existing landscape contain identical fuel information.
The ideal landscape is a difficult concept to comprehend at first since most land managers are familiar with identifying treatments by eliminating areas with resource conflicts. The perfect ideal landscape should propose a treatment for every vegetated cell. It is not the areas that will be treated, only the area that can be treated (and the fire behavior effects of those treatments). TOM will decide the optimum areas to treat based on this and the other input settings. The key to designing a ideal landscape for TOM to use is to design treatments that are commensurate with the resource values to be protected from undesirable fire effects.
Click the 
button to display the Ideal Landscape options,
This is the maximum distance dimension of the treatment units TOM identifies, in landscape units. Many times you will have to calculate this based on
the maximum area allowed for a treatment unit (use the square-root of the maximum treatment area) and
the maximum distance that a large fire can travel before it ideally encounters a treatment unit.
This is a factor that increases the number iterations and solution time, but has shown only minor benefits in terms of improved output.
This is the maximum proportion of the Analysis Area that can be treated. For example, a setting of 0.15 will attempt to find the 15% of the landscape that makes the most difference to fire growth. The treatment areas will be selected from the areas where the ideal landscape shows a reduction in fire growth compared to the current landscape.
Experience to date give some indication of useful values,
this setting should be between 0.10 and 0.30,
there appears to be little additional benefit gained with values above 0.30,
settings below 0.05 show no effective results,
values above 0.50 give results similar to random or other treatment prioritization methods.
The TOM procedure produces the same outputs generated by MTT and additional grids specific to fuel treatments:
Treatment Opportunities Grid - Displays the results of comparing the fire spread rates of the current and ideal landscapes. A positive cell value indicates the ideal landscape has a lower fire spread rate, zero indicates no difference between the two landscape, and a negative value points out where the ideal landscape has a higher spread rate. TOM will identify treatments only in cells with a value of +1.
Treatment Grid - TOM selected cells with a value of “1” for treatment. Cells with zero values were not treated.
Treated landscape - right-click the TOM run item and select Create Treated Landscape from the shortcut menu.
Four default grids are included in every FlamMap run, three of Elliptical Dimensions and one for Maximum Spread Direction. These are intermediate products of the MTT model and made available for research and creating unique output products.
The FlamMap "Run:" dialog box/tabs have a status bar and functional buttons at the bottom to help you keep track of where you are at with the set up process.
The fuel treatments identified by TOM are designed to retard fire growth rates for the landscape as a whole. The effect of any one treatment unit is difficult to discern.
The difference between the post-treatment and pre-treatment fire growth rates can be demonstrated by running MTT on the pre-treatment landscape and comparing fire arrival times with the same point on the arrival time grid produced by TOM.
The treatment grid produced by TOM can be exported to a GIS and compared to existing stand-polygon coverages to understand the practical implications of the treatment optimization on actual polygons. The treatment grid tends to clump treated cells together, because neighboring cells often have similar fire behavior, but does not follow any stand or vegetation polygon boundaries.