I have been getting a lot of requests from clients regarding impact analysis. The problem: how can I modify my data a bit, run analytics, and then see the results. Mostly the goal of all that is trial and error until something is done right and a decision can be done at a business level to increase value to the company.
Examples:
Chemical industry - should we add another chemical reactor and will that boost our flow through the network, therefore maximizing profit?
Supply chain - if we do a few changes, will we achieve maximized production and not have excess reserves in our storage
The problem becomes more tricky the greater the risk is to the business if a wrong decision/investment has been made to resources that don't pay off.