This method of examination aims to manipulate only one variable at the time, by keeping other variables as the constants.Therefore, the results of this study do not guarantee the global optimal design of the process and by this method a nearly optimal design of process can be achieved. The problem can be tackled only with the coupling of an external sub-routine/software for optimization with process simulators
in process design you must follow a process to achieve a point, so you must consider anything about your point, but in process optimization you can keep other varied as the constants and work on a variable which can be optimize your process.
You may want to consider Plackett–Burman method: https://en.wikipedia.org/wiki/Plackett%E2%80%93Burman_design https://www.itl.nist.gov/div898/handbook/pri/section3/pri335.htm