Estimated Delivery Model

3. Model

Features engineering

Features	Feature engineering	Description
Order features: subtotal, cuisine
Item features: price and type
Order type: group, catering
Merchant details
Store ID	Store Embedding
Realtime feature	Number of orders, number of dashers, traffic, travel estimates
Time feature	Time of day (lunch/dinner), day of week, weekend, holiday
Historical Aggregates	Past X weeks average delivery time for: Store/City/market/TimeOfDay
Similarity	Average parking times, variance in historical times
Latitude/longitude	Measure estimated driving time between delivery of order(to consumer) & restaurants

Training data

• We can use historical deliveries for the last 6 months as training data. Historical deliveries include delivery data and actual total delivery time, store data, order data, customers data, location, and parking data.

Model

Gradient Boosted Decision Tree

• Gradient Boosted Decision Tree sample