CODEWITHSUNDEEP
optaplanner
Vibeeshan Mahadeva
Vibeeshan Mahadeva

Reputation: 7248

What's the better what to add constrains in OptaPlanner

I am building an OptaPlan solution for an optimization problem.

I have an List<Integer> called weights These are the only planning variable in my model, all others are shadow.

I want to make sure that sum of weights (weights[0] + weights[1] + weights[2]... = 1000) , While variables other than weights (shadow) will make sure whats the best solution (Using SimpleScore).

Using a simple EasyScoreCalculator might produce a huge set to search for my problem. I believe penalizing score every time for weights[0] + weights[1] + weights[2]... != 1000 is an un-optimized (slow) method.

Whats the best way to do this.

Upvotes: 0

Views: 93

Answers (1)

Geoffrey De Smet
Geoffrey De Smet

Reputation: 27337

You should penalize the difference between that sum and 1000. So for example, if the weight sum in 920, penalize -80. If the weight it 1019, penalize -19.

Take a look at the cloud balancing problem, I suspect you'll need to just sum all the weights of the items that have @PlanningVariable Boolean included equal to true.

Upvotes: 1

Related Questions