Capacity and efficiency problems in real-time logistics - and how data at Swiggy solves this
summary
Presents how Swiggy uses historical and real-time data to decide capacity (when to stop accepting orders) and to maximise efficiency (orders delivered per delivery-executive per unit time). Covers data accuracy issues, high variance in inputs, and the trade-offs involved in designing optimal real-time logistics systems.
AI-written summary · about
chapters (9)
- 0:00Start
- 6:44Capacity - Aggregated
- 10:01Order Limiter - True, Dynamic, Automated
- 11:15Capacity - Order Level
- 18:08Trade-offs
- 28:07Challenges. High Variance
- 32:07Running Experiments in Logistics
- 35:37Experiment Strategies
- 40:08Summary - Learnings
related talks
44mHow Swiggy builds reliable agentic data co-pilots at scale
29mPredicting order batchability
27mOur experiments with food recommendations
08mAI-Driven Innovation for Enhanced Food Delivery Operations
23mOptimising organized transport at scale
30mBuild on-demand logistics apps with Python, OR-Tools, and DecisionOps
36mSmarter Tools for (Citi)Bike-Sharing
27mOptimizing Omnichannel Order Fulfillment with AI and Advanced Analytics