Why data science will help Instacart succeed where others failed - Interview with Jeremy Stanley
summary
Instacart describes a fulfillment system that only works at scale with data science: predictive time-and-space models estimate per-shopper travel and shopping time and their variance, then a real‑time routing engine solves a traveling‑salesman–style assignment blended with heuristics. Results claim 20% faster fulfillment and 20% fewer late deliveries. Future work targets in‑store shopper routing and tighter integration of machine‑learning predictions with optimization rather than separate components.
AI-written summary · about
chapters (6)
- 0:00Start
- 1:42A Specific Example of How Data Science Has Improved Efficiency
- 5:05What Do You See Happening with Data Science and Logistics over the Next Three Years
- 5:50Robotics
- 6:55Hadley Wickham
- 7:47How Do You Organize the Data Science Team
related talks
31mFrom stockouts to happy customers: Proven solutions for time series forecasting in retail
22mReal-world applications of machine learning: Beyond the hype
24mData Science: the best path to align business objectives
13mEnterprise-Level Data Science Success
41mWhy is Data Science still not a mainstream in Corporations
25m"Data Science & Agile: Strategies for Working Together"
32mLean Experimentation and Decision Science with Humans in Warehouses
19mThe Data Science behind Cruise’s Robotaxi Fleet