Simulation Strategy : Jedox

Hi,

Today in this post, I would like to share an simulation strategy using Jedox spreadsheets, where business users can design models via formulas and upon which simulations can run, building upon the case list for statistical insight.

Background:

Many times in real-world scenario, the clients of any business have a profile (meaning they have their choices, preferences, and a business can have multiple types of different profiled client), on which the business has to price products for optimal profit.

Now some times in retailing the bundling of products is also required to attract more customers. So for further taking this post, I have build a case study for an Indian takeaway business where the owner can profile his diners and design bundled menus for optimal profit.

Model Question:

With my existing Indian take away menu, and the following customer profile :

Couple

Type: Vegetarian

Categories Opted: Entrée (Any), Two different main course dish, rice, bread and dessert

With all the possible combinations of my menu has to offer, to the above diner experience.

Which combination of items on the menu will give me the  Maximum, Minimum and Average  diner spending.

Model Answer:

The model in the post runs a 1000 combinations of all the random items with the relevant filter to generate the diner spending and archives the information. Upon which the statistics on the generated list is searched for the required insights.

Notes: In the workbook, there is a simulation worksheet, which is actually not necessary as for larger list of combination generation it would dramatically slow down the system and its not actually required by the user, so in the long term for such simulated models I would stored the data in an array itself only and dump the evaluated statics on that array, thus saving the workbook size and opening the doors for larger iteration simulations.

Also higher the count of simulations, more accurate your model results.

In this post I have used a very simple simulation model to exhibit the basics for the simulation strategy, but this can open doors to more complex evaluations of Monte Carlo Simulations and similar for option pricing and more, which I will plan to come up soon as well.

http://en.wikipedia.org/wiki/Monte_Carlo_method

http://en.wikipedia.org/wiki/Monte_Carlo_methods_for_option_pricing

Hope it helps !!!

Enjoy !!!

Downloads:

http://amolpandey.com/home/downloads/blog31_download.zip