Feb. 5, 2020
Moore School associate professor Pelin Pekgün was recently elected to the 2020 INFORMS Board of Directors as the vice president of membership and professional recognition.
INFORMS, the Institute for Operations Research and the Management Sciences, is the largest international association for professionals in operations research, management science and analytics with more than 12,500 members worldwide.
Pekgün, who has been on the Moore School management science faculty since 2012, has been actively involved with INFORMS for more than a decade. She has served as the chair of membership and member services and the professional recognition committees and as a member of the Franz Edelman Award and Daniel H. Wagner Prize committees. She has assisted with organizing tracks and sessions at annual conferences and has given numerous presentations. She is also currently an editorial board member and associate editor for the INFORMS Journal on Applied Analytics.
Regarding her research and teaching, Pekgün said she has been passionate about closing the gap between industry and academia and increasing the applicability of academic research in practice. Pekgün focuses much of her research on the intersection of two areas within marketing and operations: pricing and revenue management and supply chain management, with a particular interest in understanding customer behavior and developing operational and pricing strategies that can incorporate this behavior.
Her most recently published paper, “Does Forecast-Accuracy Based Allocation Induce Customers to Share Truthful Order Forecasts?” explores how in some industries, buyers often submit forecasts, or estimates, for future orders, also known as “soft orders,” Pekgün said. She worked on the research with Minseok Park, a Moore School Ph.D. alumnus and an assistant professor at Salisbury University, Pınar Keskinocak, the William W. George Chair and professor at Georgia Tech, and Mani Janakiram, an adjunct professor at Arizona State University and senior director and senior principal engineer for Intel. Published in the October 2019 issue of Productions and Operations Management journal, the paper was Park’s doctoral dissertation.
“Since soft orders represent the intent of purchasing and are often not legally binding, buyers tend to submit inflated order forecasts to secure capacity from their supplier,” Pekgün explained. “The supplier, knowing this forecast inflation behavior of the buyers, may discount or even ignore the forecasts received, leading to lower service levels and longer delivery times, which may further drive the buyers to inflate their forecasts.”
Pekgün and her colleagues decided in their study to test the impact of an allocation policy in a lab setting. They developed an interactive game that simulates supply chain where one supplier sells a key component to two buyers, who in turn sell to consumers. Buyers share forecasts of future orders with their suppliers. The participants in the game play the role of a buyer, while the supplier is automated.
“Our experimental findings suggest that rewarding forecast accuracy in allocating inventory can significantly improve the order forecast accuracy of the buyers and reduce their forecast inflation behavior,” Pekgün said. “Interestingly, even without communication of the policy, buyers learn over time that more accurate forecasts lead to better service from their supplier and improve their order forecast accuracy.”
Beyond forecasting, Pekgün studies pricing and revenue management for the hotel industry. Unlike the airline industry, which she thinks has been extensively studied when it comes to revenue management, she said hotel revenue management has not received enough attention, and it is an interesting area with unique challenges.
In a paper published a couple of years ago in the Manufacturing & Service Operations Management journal, Pekgün worked with then Ph.D. student Ӧvünç Yılmaz, now an assistant professor at the University of Notre Dame, and fellow Moore School management science professor Mark Ferguson to determine how hotels can best price their room upgrades. As Yılmaz’s doctoral dissertation, the research focused on the discounts guests are given for a room upgrade after they reserve a standard room but are only charged for the upgrade if the premium room is available when they arrive at the hotel.
Pekgün said hotel guests may act strategic in booking standard rooms with the hope they will receive an offer for an upgrade to a premium room at a much-discounted price. Through an analytical model, the research team found that the success of a standby upgrade program is directly tied to the type of guests who frequent the hotel and the availability of premium rooms. When a hotel faces repeat guests, who recognize the higher likelihood of being awarded a standby upgrade and thus act strategically, the hotel should decrease the discount offered for a room upgrade.
In another hotel study, Pekgün focused on understanding the effect of user reviews on hotel demand. Analyzing the text of reviews with Moore School Ph.D. student Sanghoon Cho and marketing professor Ramkumar Janakiraman, the team found that negative reviews, relative to competition, are more influential in guests' booking decisions than positive reviews, particularly when they are paying higher prices.
Pekgün is currently working on developing a new model and estimation method for personalized pricing of hotel rooms, room upgrades and a hotel’s ancillary services in collaboration with Oracle Labs.