We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Ryan O'Neil - Build on-demand logistics apps w/ Python, OR-Tools, & DecisionOps | PyData Global 2023
Discover how to build on-demand logistics apps with Python, OR-Tools, and DecisionOps, solving optimization models for forecasting, scheduling, and routing, with a Live Demo.
- The talk is about using Python, OR-Tools, and DecisionOps to solve optimization models for on-demand delivery.
- The three core models are forecasting, scheduling, and routing.
- Forecasting is not typically done with an optimization solver, but it can be used to generate synthetic data for testing.
- Scheduling involves minimizing the sum of residuals where it’s the absolute value, with a mix of integer programming and constraint programming.
- Routing involves finding the optimal sequence of stops.
- The talk uses a synthetic dataset for demonstration purposes.
- The speaker has 17 years of experience with Python and has seen the adoption of data science in Python explode over the last few years.
- The speaker’s company, NextMove, uses Python for optimization modeling.
- The talk includes a demo of using OR-Tools to solve a scheduling model and a routing model.
- The speaker mentions that OR-Tools is a hybrid solver and a multi-paradigm library that supports linear programming, mixed integer programming, and other types of optimization.
- The talk also mentions the use of LAD (least absolute deviations) regression, which is a type of regression that is more robust to outliers than traditional least squares regression.
- The speaker encourages attendees to explore other tools and libraries for optimization modeling.