Winning solution for the orderbatching problem from the third Relaxdays Hackathon.
Every problem instance consists articles and orders. Every article has a volume and is located in a warehouses aisle. The goal is to distribute orders on waves and to separate the articles from each wave into batches under the following constraints:
BatchVolume <= 10,000
(each article in a wave has a volume)WaveSize <= 250
(maximal amount of articles in a wave)
The quality of a solution is measured in the following way:
$TourCost = \sum_{b \in Batches} CountWarehouses_b * 10 + CountAisles_b*5$ $RestCost = CountWaves * 10 + CountBatches*5$