Else, Someone on Nostr: ...you only describe the inequalities, which doesn't even which give you an easy way ...

...you only describe the inequalities, which doesn't even which give you an easy way to construct a "feasible" solution, leave alone an "optimal" in whatever sense of optimality. So you basically end up with a MILP, which is NP hard, and can only be approximated