Timeline for Method for (binary) optimization under constraints
Current License: CC BY-SA 4.0
10 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 21, 2022 at 10:12 | vote | accept | kris | ||
| Oct 21, 2022 at 10:10 | vote | accept | kris | ||
| Oct 21, 2022 at 10:10 | |||||
| Oct 7, 2022 at 21:40 | vote | accept | kris | ||
| Oct 21, 2022 at 10:09 | |||||
| Oct 7, 2022 at 21:36 | comment | added | kris | Thank you, i changed the constraint and the first test looks promising! | |
| Oct 7, 2022 at 21:35 | vote | accept | kris | ||
| Oct 7, 2022 at 21:39 | |||||
| Oct 5, 2022 at 20:59 | comment | added | RobPratt | OK, I missed the "maximize" part. I think you will need to introduce a new binary variable $y_i$, change the $=t_i$ constraint to $=t_i y_i$, and use a MILP solver. The all-or-nothing requirement destroys the total unimodularity. | |
| Oct 5, 2022 at 20:52 | comment | added | kris | I mean that in my problem case the demand $t_i$ should always be either fully satisfied or all edges to the node should be 0 (e.g. a machine can only run if the exact amount of resources that are needed, are supplied) And since i want to maximize the function, i also calculate $max = max(e_j * c_i)$and use $max - (e_j * c_i)$ as costs for the edges from one node $j$ with supply 1 to the demanding node $i$, so i can still use minimum-cost flow algorithm. Or is there a better algorithm for my case? | |
| Oct 5, 2022 at 18:58 | comment | added | RobPratt | Not sure what you mean. If the demand satisfaction is not enforced, the optimal solution is $p\equiv 0$. | |
| Oct 5, 2022 at 17:50 | comment | added | kris | Thank you, that's what i was looking for! So i tried to solve that with a minimum cost flow algorithm, but the problem is that as far as my understanding goes, the demand (the outgoing capacity of the demand node) doesn't have to be fully satisfied in that algorithm. Do you maybe know an algorithm where in the solution the demand is either fully satisfied or 0? | |
| Oct 3, 2022 at 20:12 | history | answered | RobPratt | CC BY-SA 4.0 |