Recent Posts by Eric Kelso

Eric Kelso is Vice President of Product Management for Enterprise Optimizer at River Logic. He has over two decades experience building and implementing prescriptive analytics optimization software, and has successfully completed more than 50 major projects in the discrete, process manufacturing and consumer packaged goods industries.
Find me on:

Why Simulation is a Must for Optimization-based Scenario Planning

In the age of Big Data, corporate leaders have more power at their disposal than ever before. Meanwhile, globalization has increased the number and quality of competitors fighting for the same customers. There is enormous pressure on planners to achieve economical, efficient strategies that increase return on investment with as little risk as possible. In the past, intuition was a planners’ tool of choice, but most today are tech savvy enough to have moved onto better techniques. With the sheer number of variables to consider, there are not enough hours in the day to cope with the resultant information overload.

Read More

A Prescriptive-First Approach to Driving Business Impact with Analytics – Part I

Author’s comment: this is part 1 of 2 of a hypothetical case study. The story described below can be any company willing to consider using a Prescriptive Analytics approach to improve their decision-making abilities.

Read More

Prescriptive Analytics: It’s More Than Rules and Big Data

I recently read Prescriptive models take analytics one step beyond, written by Scott Robinson, Louisville Metro government, and published by SearchBusinessAnalytics.

Read More

­­­­Enterprise Optimizer Modeling Series: A Chemical Company Example

This is the first blog post in a series aimed at highlighting important Enterprise Optimizer® (EO) modeling features and uses. All information below is based on an EO model created for a real customer. High-level background description is provided, but no confidential information is disclosed.


River Logic partnered with a Top 5 consulting company to design, build and implement a Sales & Operations Planning (S&OP) modeling platform for a large chemical company (“ChemCo”) with operations throughout the Americas and Europe. The impetus for this project was a definable gap in their S&OP planning process.

Read More

Microsoft Office Excel Scenario Manager and Prescriptive Analytics

River Logic’s Enterprise Optimizer® and Microsoft Office Excel® have a long and close working relationship. For two decades, EO’s prescriptive analytics-based models have read data from and written data to Excel workbooks. Consistently one of EO’s most popular data sources, Excel makes building prototype EO models or conducting quick, one-off consulting projects considerably easier. It has also been an excellent option to analyze EO model solution results, either with Excel’s built-in features or using add-on technologies like Tableau.

Read More

8 Data Requirements for Advanced Logistics Planning

As I began to write about advanced logistics planning, the first thing that popped into my head, literally, was “damn motorcycles.” I’m currently at 35,000 feet, flying home after a couple of weeks with my wife visiting relatives and friends in Thailand. My mother-in-law’s home in central Bangkok used to be on a quiet, dead-end street (“soi” in Thai). At least, that was until about 10 years ago, when a major Asian parcel delivery company determined (somehow) that the best location for their Thai head office was further in our soi!

Read More

Monte Carlo Optimization for Beginners

The Monte Carlo method is a well-known simulation technique that uses statistical random sampling to solve mathematical problems. In use for about 85 years, many variants exist across a wide range of disciplines. If you're not familiar with these, I suggest reading this Wiki page.

Read More

Results of the 2015 Operations Research Survey

Read More

Why Optimization Modeling Should Move Beyond Using Scripts

In the first decade of my career, most optimization models that I authored also required one or more scripts. They were necessary to automate tasks such as importing data, setting parameters, solving the model, executing another program (e.g., to crunch numbers), generating reports, and so on. Most were written in DOS shell, with an occasional UNIX shell depending on the platform used.

Read More

River Logic's Learnings from INFORMS 2015

“I used to use [AIMMS, GAMS, MPL…] but I don’t anymore.”

I heard this comment repeatedly this week at the Institute for Operations Research and Management Sciences (INFORMS) conference in Philadelphia, PA. My colleague, Rohan Guthal, and I staffed the River Logic booth from November 1-3. We met with a variety of professors, students and practitioners. In nearly every case, when asked – “What optimization software are you using?” — that was the response we heard: they used to use a certain program, but no longer do. More on this below...

Read More
Supply Chain Brief