What is the relationship between price, profit, and waste? Let’s find out using a Statsmodels ML regressor, a Gurobipy optimizer, Plotly charts, scenario testing, and a slick decision workflow built and run in Nextmv.
.png)
Follow this step-by-step tutorial to go from a forecasted demand to optimized routes for delivery and similar use cases. Create and customize decision apps using OR-Tools, HiGHS, Pyomo, and more.
It’s time to understand the behavioral impacts of a new decision model under real-world conditions compared to a production model. Switchback testing enables this, helping build confidence in a new model’s rollout in a safe and measured way.
I retired our on-call scheduling spreadsheet — and you can too. Here’s how I built a custom decision model that generates and sends optimized schedules to the PagerDuty API.
Deploy and run a Statsmodels ML regressor, Gurobipy price avocado optimizer, and end-to-end decision workflow on Nextmv via a Jupyter Notebook on Google Colab. Plus, visualize results using Plotly and perform a scenario test to determine relationship between price and profit.
In under ten minutes, we demo how to deploy and run your local Python decision model from a Jupyter Notebook as a fully featured decision app on the Nextmv platform for simpler collaboration and a smarter workflow.
Go from a local decision model to a remotely hosted decision application (with API endpoints) that you can easily share, test, and confidently promote to production. See the model-to-app workflow using a Jupyter Notebook and Nextmv in under 30 minutes.
Join us for a tour of the newest features and get a sneak peek of what’s to come. See how our decision science platform helps teams launch and scale optimization projects faster and with more confidence.
If you develop decision models in Python, this presentation will save you time (and the added effort of building and maintaining DecisionOps tools). Accelerate development of your optimization models with features for testing, deploying, managing, and collaborating.
Learn how to use vehicle activation penalties to encourage vehicle efficiency. This is sometimes known as prioritizing backhaul when going back to a depot.
Learn how to configure your vehicle routing problem (VRP) to have multiple pickups precede a dropoff. In this example, learn how to set two pickups at two different locations precede a dropoff stop for a pickup and delivery problem.
Learn how to model continuous moves for a vehicle going back and forth from a depot.
Create robust and interactive decision apps with AMPL as the optimization layer, Nextmv as the DecisionOps layer, and Streamlit as the UI and visualization layer. Learn how with a facility location example.
When you’re ready to have a candidate model make true operational decisions, it’s time for switchback testing. Kick off an experiment and analyze how your new decision model measures up to your current model in production.
With Nextmv, you can customize an optimization model for your use case without wading into linear inequalities. From creating your own value function to adding custom constraints, learn best practices for representing business logic as code.
