Specifying Searches

Top  Previous  Next



Quick Start Instructions


1.Open or launch the market simulator (the Choice Simulator).

2.Add a simulation scenario that (typically) includes multiple competitive products.

3.For searched products, rather than selecting specific attribute levels, you may type multiple values or ranges into the attribute entry grid.  Examples:


1-3 indicates that search should examine levels 1, 2, and 3 for this attribute

1-3,5   indicates that search should examine levels 1, 2, 3, and 5 for this attribute

100, 125, 145.75, 175, 200, 240, 280, 300 indicates that search should examine levels of 100, 125, 145.75, etc. through 300 (for a continuous attribute, where you have associated values of 100 to 300 with the levels included in your attribute list)

=Range(100,300)   indicates that search should examine all level values (levels included in the attribute level list) from 100 to 300

=Range(100,300,10) indicates that search should examine all values from 100 to 300 using a step size of 10 units (100, 110, 120, etc. through 300)




4.On the My Scenario Settings tab, change the Range Behavior to one of the search methods (e.g. "Search (Grid)").

5.Click the gear_blue icon next to the Range Behavior option to choose a goal (objective) for the search. For example, Share of Preference can be the goal to maximize.  (If you are maximizing profits, specify costs associated with different attribute levels back at the Home tab, Revenues & Costs option on the ribbon.)

6.Click Simulate.




To conduct product searches in the ASM, the user specifies several items:


One or more products for which optimal attribute levels are to be searched.  Rather than specifying fixed attribute levels for searched products in the product specification grid, variable levels are specified using the various syntax as described above, including using a Range function, such as =Range(1,4), which means to search across levels 1 to 4 of this attribute.  Interpolation steps or specific interpolated values to test may be specified as additional arguments in the Range function.


Competitive products that will be held constant throughout the analysis (unless searches are to maximize utility or purchase likelihood for a single offering).


The objective (or multiple objectives) to be optimized (estimated share of preference, purchase likelihood, profitability, revenue, cost minimization, or total utility) whether for a single product or across a portfolio of products.


If profitability searches or cost minimization are used, product cost information is also required. Attribute-based price information can also be included, so that the addition of certain product features is always accompanied by a set increase (or decrease) in price.

Page link: http://www.sawtoothsoftware.com/help/lighthouse-studio/manual/index.html?optimization_introduction_3.html