Dear Kees,

First of all, thank you for your interesting question!

You know that the issue of MOPs is mainly discussed in literature of systems engineering. Probably, you are well aware of this point that there are a wide variety of methods and approaches to address the issue of MOPs and assess/measure the system performance. However, based on recent findings in systems engineering, any system can be characterized in terms of functional requirements (this point is particularly stated in Axiomatic Design (AD) theory. Due to your specific question, it is not necessary to mention that the system of interest can be characterized in terms of customer attributes in customer domain of the system, functional requirements in functional domain of the system, physical or design parameters in physical domain of the system, and process variables in process domain of the system). Functional requirements should be measurable and quantitative. Regarding each functional requirement, a respective target value with a specific tolerance interval is defined. Functional requirements are often random variable and, hence, there is some degree of uncertainties to these variables. For this reason, we need to utilize the probability theory to quantitatively measure the success level to meet the target value specified for each functional requirement within its respective designed tolerance interval. The value of these probabilities are reported as measure of performance for the system under consideration. Concerning this discussion, I have attached a paper to this answer as well.

Hops this helps!

Good luck!

Ali