Following are lightly edited excerpts from Rob’s recent webinar (watch the recording here). An optimization model can perform beautifully in development but struggle or fail when it reaches production. The reason is straightforward: production changes the rules. A prototype often operates against a controlled dataset and can be given considerable time to find an answer. […]
Following are lightly edited questions and answers from Patricia’s presentation on Orchestrating Optimization with Precision Agents at the INFORMS Analytics+ conference and in a subsequent webinar (watch the recording here and see Patricia’s previous blog post for reference.) Question: What is the difference between an agent and a tool? Patricia: The agent is the decision […]
This post is based on Patricia’s presentation at the INFORMS Analytics+ Conference and in a subsequent webinar (watch the recording here). For decades, enterprise optimization systems have been built around a familiar assumption: if you can centralize enough data, encode enough constraints, and build a sufficiently sophisticated model, you can produce better operational decisions. That […]
Built on precision, safety, and long investment cycles, the rail industry has historically adopted innovation deliberately rather than aggressively. Significantly, AI is reshaping that pattern. Insights from the first two years (2024 and 2025) of the AI In Rail survey, conducted by Princeton Consultants with the Northwestern University Transportation Center (NUTC), point to a clear […]
At the INFORMS Analytics+ Conference in Indianapolis, Irv presented a workshop along with Siddhesh Pore of Databricks, https://www.databricks.com/. They discussed data-first principles and best practices, data prep and engineering, and optimization integrations on the Databricks platform. Following is an edited excerpt from the Q&A at the end of the program. Audience Member: I enjoyed the […]
In a Gurobi webinar on Data-First Optimization Development, Irv discussed the transformative benefits of pandas and best practices using the gurobipy-pandas library, and walked through an example. Following is a lightly edited excerpt from Irv’s presentation and the Q&A with practitioners from around the world.  In emphasizing Data-First optimization development, I have drawn on […]
Princeton Consultants employs a 7-step methodology that improves the likelihood of success and deliverability in optimization projects. Based on our solution development and deployment for many clients in varied industries, this approach helps create useful documentation, shortens the overall development time, improves code maintainability, and provides a natural feedback loop to business requirements. The main […]
At Princeton Consultants, we have started to do the majority of our optimization model development using python with the pandas library (https://pandas.pydata.org/). We have successfully developed models using the Gurobi Python interface, and the DO CPLEX Python interface for IBM ILOG CPLEX Optimization Studio. Based on these successes, we have codified our approach for rapid […]
For successful implementation, it is critical to understand how data feeds the optimization system. There are two sources of data. The first is automatic feeds. I have been involved in building systems where one database is feeding another database, which is feeding another database, and eventually it feeds the model that you have. What happens […]
Over a year ago, I discovered a wonderful article by Hadley Wickham introducing the concept of “Tidy Data.” Here is a complete reference: Hadley Wickham, “Tidy Data”, Journal of Statistical Software, Volume 59(10), August 2014. Wickham, a major contributor to the open source R package for statistical analysis, was addressing the needs of statisticians in […]
