4 ms·
The article overstates the use of ML at least in fulfillment we use a lot of math models (eg xpress solver) and optimization algorithms (traveling salesmen type
by beastcoast 8y ago
The article overstates the use of ML at least in fulfillment we use a lot of math models (eg xpress solver) and optimization algorithms (traveling salesmen type problems) but not so much in the ML space.
- mabbo 8y agoMy team does ML in Fulfillment. Search my username on phone tool to figure out my team.
- east2west 8y agoUsually optimization takes forecasts as part of inputs. It is actually a lot like reinforcement learning, where the sales forecasts are part of the environment. This is standard process for supply-chain management, not just retail industry or at Amazon.