I have a dataset containing the walking times it takes to walk from one postal bin to another postal bin in a mail distribution center. The features I have, include the total mail workers on the floor (indicates how busy it is), the seniority of the mail workers, the rack number where the worker is now, the rack number of where the worker needs to go to pick up his/her container with letters. The rack numbers reflect the physical order of the racks so it indicates how far they are apart. I also know whether the container with letters is on the floor or up and whether the container is small/medium/large.

I have tried several basic sklearn regression models and did data cleaning/feature engineering (standardization, taking out outliers, etc.). The results are okay but not amazing. The models also don't perform so well when the containers are small, it overestimates the time taken. What else seems worth trying? I know it's possible to get better results because others on the leaderboard have gotten way better results.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.