I frequently see references to the use of Markowitz optimisation (the efficient frontier) or the Sharpe ratio in the context of optimisation modelling in oil and gas applications. To me, these generally seem like inappropriate frameworks in practical cases.
Markowitz optimisation and the Sharpe ratio are used to optimise financial portfolios (e.g. stocks or equities); these are standard tools in finance and in academic textbooks. In particular, their power is that only two parameters are required to do define the optimisation solution (i.e. the mean and standard deviation of any trial portfolio).
On the other hand, in practical oil and gas portfolio optimisation, almost always the criteria applied are usually quite different. For example, one may wish to achieve some future production level (with at least a given probability, such as 90%) by placing limits on total expenditure over a given future period (such as three or five years) and also whilst not exceeding a particular annual threshold.
Such practical cases require different optimisation models. Of course, the underlying principles, as well as the tools to search for optimum solutions may be the same (or similar in each case), but the models and objectives are different, so that these financial-asset-oriented frameworks may provide a reference point for basic optimisation concepts, but are generally not at all sufficient to real-life applications to the oil and gas or many other sectors.