Finding the Best Path to Your Target

Algorithm Helps Organizations Find Their Sweet Spot

Jul 25, 2019
As Featured In 
Operations Research

Political candidates, manufacturers and even online game designers can hit their performance targets with increased regularity using a new algorithm developed by professor Ilya O. Ryzhov at the University of Maryland’s Robert H. Smith School of Business.

The prediction model works best in situations where decision makers have a complex set of variables to consider and a predetermined target — rather than a general desire to maximize results as much as possible.

An example occurs during presidential campaigns. Candidates use polling data to identify locations where political rallies and media buys might have the biggest impact. The goal is not to maximize results as much as possible, but to push candidates past the threshold to win elections in key “swing states.”

Boosting someone’s popularity by 5 or 10 percentage points is wasted effort if the starting point is 30% or 60%. The real objective is to select regions where the candidate’s rating is as close to 50% as possible.

A similar situation occurs in manufacturing and quality control. The goal is usually not zero product defects, which might be too expensive to achieve. Instead, process designers determine an acceptable tolerance level and run simulations to find the most efficient path to the target.

With online gaming platforms like Microsoft’s Xbox Live, the goal is to match opponents whose skill levels are as close together as possible.

Simulation optimization often relies on a “ranking and selection” framework, which assumes a desire to maximize mean performance. Ryzhov uses a “targeting and selection” framework instead, which assumes a predetermined target.

His proposed algorithm then simulates the option with the best chance of matching the target — or the alternative with the highest expected "Brownian local time," a mathematical term used to predict the proximity of a moving variable to a predetermined location. "The local time argument is a novel and potentially useful idea for sequential selection problems that involve targeting," he says.

Read more: The Local Time Method for Targeting and Selection, by Ilya O. Ryzhov is featured in Operations Research, Vol. 66, No. 5.

About the Author(s)


Ilya Ryzhov is an Associate Professor of Operations Management in the Department of Decision, Operations and Information Technologies. He received his Ph.D. in Operations Research and Financial Engineering from Princeton University in 2011. His research deals with the role of information in decision analysis, exploring the way in which new information influences and improves decision-making strategies. Information has an economic value that can be measured and balanced against other economic concerns as part of the decision-making process. Dr. Ryzhov develops efficient ways to achieve this balance, with applications in pricing, revenue management, and optimization of energy costs. His work has appeared in Operations Research.

More in


How To Guard Against Cyber Attacks
Can centralized IT decision-making protect an organization from cybersecurity breaches?
Jun 01, 2020
A New Approach to Data-driven Modeling
Research offers new approach for data-driven predictive analytics models as an alternative or compliment to AI approaches.
Apr 30, 2020
A Mathematically Rigorous Way To Analyze Statistics from Simulations
New research from Maryland Smith’s Michael C. Fu offers a rigorous way to analyze statistics generated from simulation models.
Apr 30, 2020