Beyond Prediction: Solving the Multiple Knapsack Problem at Scale: How Uber Optimizes Incentives

What this post added

This post details the evolution of Uber's incentive optimization platform, Tarot, by framing it as a large-scale Multiple Knapsack Problem (MKP). It introduces the concept of a multi-lever optimizer, utilizing Google OR-Tools CP-SAT solver, and a two-layer ROI calculation that incorporates strategic weighting (w) and predicted metric uplifts (x) to handle complex trade-offs like cross-vertical cannibalization and long-term strategic bets. The Budget Pacer's role as a control loop for managing spend velocity against deterministic financial budgets is further elaborated.