... into a mathematical function to be maximised under constraints (like a linear or non-linear regression (Operations Research, OR) model). Some problems can, due to size of parameters. But most do not. Eg for my MS thesis I used a combination of optimisation and "heuristic" technique/approach to find as optimal as possible solution to an aircraft assignment problem.
Plus most problems, be they personal, family/household, policy or business, do not have full data available to plug them in a get a solution. Usually logic, esp. deductive reasoning, trade-off analysis, as well as a philosophical mentality or even wisdom are the necessary ingredients.
As one of my professors in Forecasting used to say, if the world could be "represented" by a model, then that model would become the world.
Any lessons re the "financial world" and its certainties?