John Turner

As Associate Professor of Operations & Decision Technologies at the Paul Merage School of Business at UC Irvine, my research specializes in applying rigorous optimization-based methods to real-world problems. My interests include revenue management, large-scale optimization and decomposition methods, online advertising allocation, sports scheduling, environmental policy, and health care management.

I am interested in new models, new insights, and new applications, as well as new theory and efficient algorithms for solving such problems.

I teach management science, operations management, and revenue management, and integrate real-world examples from advertising and other industries into the classroom.

Research Highlights – Online Advertising & Operations / Marketing Interface

In one research stream, I apply concepts from the fields of revenue management and optimization to interesting problems in the online advertising industry and other marketing/operations problems. My contributions in this stream include:

  • Identifying how to effectively schedule ads into console-based 3D video games (INFORMS George B. Dantzig Dissertation Award winner)
  • Defining a modeling paradigm for a broad class of display advertising, called Guaranteed Targeted Display Advertising, along with an efficient duality-based algorithm that simultaneously clusters audience segments while producing an optimal allocation plan (INFORMS George B. Dantzig Dissertation Award winner)
  • Developing a hierarchical planning algorithm that uses patterns to satisfy ads with guaranteed reach and frequency requirements (Yahoo! Faculty Research & Engagement Award winner)
  • Proposing a new way to measure dispersion of ad exposures across different user types and time, based on the Gini Coefficient and Lorenz Curves, and developing an associated efficient optimization decomposition method to solve Gini-based ad allocation.
  • Studying user browsing and purchase behavior on a large travel website, to understand how the number of hotels evaluated by the user is connected to purchase likelihood. Our results reveal that evaluation set size and purchase are negatively correlated and that factors typically presumed to be associated with purchase – i.e., when users sort search results by price or quality, request many rooms, disclose that there are many guests in their party, or arrive from other search engines and/or partner sites – actually relate to larger evaluation sets but lower purchase probability. In contrast, when users filter the search results, we observe smaller evaluation sets and higher purchase probability.

Research Highlights – Healthcare Logistics / Location Optimization / Queueing & Routing

Research Highlights – Sports Analytics

  • Have you ever wondered how a sports league can resume after a disruption (e.g., player strike, COVID), and optimally choose a subset of games to play to conclude the season in a compressed timeframe? We use predictive and prescriptive analytics to make this decision, and consider a novel rankings-based objective that maximizes the concordance (i.e., similarity) between the end-of-season rankings of our planned shortened season and the counterfactual full season (i.e., the rankings that would have occurred had no disruption occurred and the season was played out in full).

Research Highlights – General Methodological Contributions to the Field of Optimization

  • Benders Decomposition is a widely-used computational technique for solving large optimization problems that are important in many industries, such as supply chain planning problems. The Benders algorithm is iterative, and often takes many steps to reach a near-optimal solution. Our work contributes to the body of literature that attempts to find ways to speed up Benders. Specifically, our methodological approach generates “deep” cuts at each iteration, which “cut off” or exclude from consideration as many points in each iteration as possible. Viewed through the lens of Deep Benders Cuts, we bring several well-known cut selection strategies in Benders decomposition under the general umbrella of depth-maximizing cut selection strategies.

Research Highlights – Supply Chain / Logistics

  • Through a collaboration with a large retailer, we developed a tool to assist the retailer in selecting full-truckload transportation carriers while simultaneously negotiating fuel surcharges during its request-for-proposal contracting process, so that retailer and carrier share the diesel price risk

 

For a full chronological list of my publications, please visit my Research page. You can also find my official faculty profile website at: http://merage.uci.edu/go/jturner.