Search

We document the "good-on-paper effect," whereby people make more virtuous decisions on paper than on a digital device (e.g., tablet). In a series of 10 studies, we find that people are more likely to make virtuous decisions on paper than on a digital device because they perceive choices on paper as more real (i.e., tangible, actual, and belonging to the physical rather than the virtual world) and hence as more self-diagnostic (i.e., representative of who they are). We first show people express more interest in donating and volunteering (Studies 1a and 1b), are more likely to donate (Study 2), and put more effort into helping a charitable cause (Study 3) when these choices occur on paper (versus tablet)—a pattern of decision making we label the good-on-paper effect. Study 4 extends these findings to book choices (highbrow versus lowbrow) and to a device interaction that closely mimics writing on paper (i.e., tablet with digital pen). In the context of volunteering decisions, we then provide evidence for the sequential mediating roles of perceptions of realness and self-diagnosticity in the good-on-paper effect (Study 5 and Studies 6a and 6b). Finally, we show that chronic (Study 7) and situational (Study 8) perceptions of self-diagnosticity moderate this effect in the contexts of environmental protection and food choices (healthy versus indulgent), respectively. We discuss the theoretical and practical implications of these findings. [ABSTRACT FROM AUTHOR]

The journal is pleased to publish the abstracts of the six finalists of the 2023 Manufacturing and Service Operations Management Society's student paper competition. The 2023 prize committee was chaired by Ersin Korpeoglu (UCL), Simone Marinesi (Wharton), and Nur Sunar (UNC). The judges were Adam Elmachtoub, Adem Orsdemir, Agni Orfanoudaki, Alper Nakkas, Amrita Kundu, Antoine Desir, Antoine Feylessoufi, Anton Ovchinnikov, Anyan Qi, Arian Aflaki, Arzum Akkas, Ashish Kabra, Auyon Siddiq, Bilal Gokpinar, Bin Hu, Bob Batt, Bora Keskin, Brent Moritz, Can Zhang, Chloe Glaeser, Cuihong Li, Daniel Freund, Daniel Lin, David Drake, Divya Singhvi, Dongyuan Zhan, Ekaterina Astashkina, Elena Belavina, Elodie Adida, Emre Nadar, Enis Kayis, Fabian Sting, Fanyin Zheng, Fei Gao, Florin Ciocan, Francisco Castro, George Chen, Georgina Hall, Gloria Urrea, Gonzalo Romero, Guihua Wang, Guoming Lai, Heikki Peura, Hessam Bavafa, Hummy Song, Huseyin Gurkan, Ioannis Stamatopoulos, Iris Wang, Jiankun Sun, Jiayi Joey Yu, Jing Wu, Joel Wooten, John Silberholz, Jonas Oddur Jonasson, Jonathan Helm, Jose Guajardo, Junyu Cao, Kaitlin Daniels, Karen Zheng, Ken Moon, Kostas Bimpikis, Lennart Baardman, Lesley Meng, Lina Song, Luyi Yang, Mazhar Arikan, Mehmet Ayvaci, Meng Li, Mengzhenyu Zhang, Miao Bai, Michael Freeman, Mika Sumida, Ming Hu, Morvarid Rahmani, Mostafa Rezaei, Mumin Kurtulus, Nan Yang, Nazli Sonmez, Negin Golrezaei, Nektarios Oraiopoulos, Nikhil Garg, Nikos Trichakis, Nil Karacaoglu, Olga Perdikaki, Onesun Steve Yoo, Ovunc Yilmaz, Ozan Candogan, Panos Markou, Pengyi Shi, Philipp Cornelius, Qiuping Yu, Renyu Zhang, Robert Bray, Ruth Beer, Ruxian Wang, Saed Alizamir, Safak Yucel, Sanjith Gopalakrishnan, Santiago Gallino, Sarah Yini Gao, Scott Rodilitz, Sebastien Martin, Seyed Emadi, Sheng Liu, Shouqiang Wang, Siddharth Singh, Sidika Candogan, Sina Khorasani, So Yeon Chun, Somya Singhvi, Soo-Haeng Cho, Sriram Dasu, Stefanus Jasin, Stephen Leider, Suresh Muthulingam, Sytske Wijnsma, Taghi Khaniyev, Tian Chan, Tim Kraft, Tom Tan, Tugce Martagan, Vasiliki Kostamj, Velibor Misic, Vishal Agrawal, Xiaojia Guo, Xiaoshuai Fan, Xiaoyang Long, Yannis Bellos, Yao Cui, Yehua Wei, Yiangos Papanastasiou, Yi-Chun Chen, Yinghao Zhang, Ying-Ju Chen, Yinghao Zhang, Yuan-Mao Kao, Yuexing Li, Zhaohui (Zoey) Jiang, Zhaowei She, and Zumbul Atan. [ABSTRACT FROM AUTHOR]

We study the quadratic cycle cover problem (QCCP), which aims to find a node-disjoint cycle cover in a directed graph with minimum interaction cost between successive arcs. We derive several semidefinite programming (SDP) relaxations and use facial reduction to make these strictly feasible. We investigate a nontrivial relationship between the transformation matrix used in the reduction and the structure of the graph, which is exploited in an efficient algorithm that constructs this matrix for any instance of the problem. To solve our relaxations, we propose an algorithm that incorporates an augmented Lagrangian method into a cutting-plane framework by utilizing Dykstra's projection algorithm. Our algorithm is suitable for solving SDP relaxations with a large number of cutting-planes. Computational results show that our SDP bounds and efficient cutting-plane algorithm outperform other QCCP bounding approaches from the literature. Finally, we provide several SDP-based upper bounding techniques, among which is a sequential Q-learning method that exploits a solution of our SDP relaxation within a reinforcement learning environment. Summary of Contribution: The quadratic cycle cover problem (QCCP) is the problem of finding a set of node-disjoint cycles covering all the nodes in a graph such that the total interaction cost between successive arcs is minimized. The QCCP has applications in many fields, among which are robotics, transportation, energy distribution networks, and automatic inspection. Besides this, the problem has a high theoretical relevance because of its close connection to the quadratic traveling salesman problem (QTSP). The QTSP has several applications, for example, in bioinformatics, and is considered to be among the most difficult combinatorial optimization problems nowadays. After removing the subtour elimination constraints, the QTSP boils down to the QCCP. Hence, an in-depth study of the QCCP also contributes to the construction of strong bounds for the QTSP. In this paper, we study the application of semidefinite programming (SDP) to obtain strong bounds for the QCCP. Our strongest SDP relaxation is very hard to solve by any SDP solver because of the large number of involved cutting-planes. Because of that, we propose a new approach in which an augmented Lagrangian method is incorporated into a cutting-plane framework by utilizing Dykstra's projection algorithm. We emphasize an efficient implementation of the method and perform an extensive computational study. This study shows that our method is able to handle a large number of cuts and that the resulting bounds are currently the best QCCP bounds in the literature. We also introduce several upper bounding techniques, among which is a distributed reinforcement learning algorithm that exploits our SDP relaxations. [ABSTRACT FROM AUTHOR]

The article announces the winner of the 2023 Clemen-Kleinmuntz Decision Analysis Best Paper Award, which comes with a $2,000 prize. The winning paper, titled "Information Density in Decision Analysis," introduces the concept of "information density" as a new approach to sensitivity analysis in decision analysis. The paper provides a technical grounding of information density and offers examples for real-world implementation. Another paper, titled "An Empirical Comparison of Rank-Based Surrogate Weights in Additive Multiattribute Decision Analysis," was identified as a finalist for the award and provides valuable insights into converting ordinal rankings into attribute weights. The article encourages authors to continue submitting high-quality manuscripts for future consideration. [Extracted from the article]

We consider a principal who periodically offers a fixed and costly nonmonetary reward to agents to incentivize them to invest effort over the long run. An agent's output, as a function of his effort, is a priori uncertain and is worth a fixed per-unit value to the principal. The principal's goal is to design an attractive reward policy that specifies how the rewards are to be given to an agent over time based on that agent's past performance. This problem, which we denote by P , is motivated by practical examples from both academia (e.g., a reduced teaching load) and industry (e.g., "Supplier of the Year" awards). The following "limited-term" (LT) reward policy structure has been quite popular in practice. The principal evaluates each agent periodically; if an agent's performance over a certain (limited) number of periods in the immediate past exceeds a predefined threshold, then the principal rewards him for a certain (limited) number of periods in the immediate future. When agents' outputs are deterministic in their efforts, we show that there always exists an optimal policy that is an LT policy and also, obtain such a policy. When agents' outputs are stochastic, we show that the class of LT policies may not contain any optimal policy of problem P but is guaranteed to contain policies that are arbitrarily near optimal. Given any ϵ > 0 , we show how to obtain an LT policy whose performance is within ϵ of that of an optimal policy. This guarantee depends crucially on the use of sufficiently long histories of the agents' outputs. We also analyze LT policies with short histories and derive structural insights on the role played by (i) the length of the available history and (ii) the variability in the random variable governing an agent's output. We show that the average performance of these policies is within 5% of the optimum, justifying their popularity in practice. We then introduce and analyze the class of "score-based" reward policies; we show that this class is guaranteed to contain an optimal policy and also, obtain such a policy. Finally, we analyze a generalization in which the principal has a limited number for rewards in any given period and show that the class of score-based policies, with modifications to accommodate the limited availability of the rewards, continues to contain an optimal solution for the principal. This paper was accepted by Jeannette Song, operations management. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2022.4482. [ABSTRACT FROM AUTHOR]

The article discusses the impact of digital technologies on the field of finance. It highlights how digital finance, including mobile banking, digital wallets, fintech innovations, and robo-advisors, has transformed consumer behavior and the global financial system. The article calls for research papers that explore the systemic impacts of digital finance, propose new methodologies for analysis, and examine the broader economic implications of its rapid growth. The special issue aims to address topics such as artificial intelligence in financial economics, digital currencies and payment systems, decentralized financing, the socioeconomic implications of digital technologies, digitization in various industries, network security, and regulation. The deadline for submission is June 30, 2025, and accepted papers will be published in regular issues of Management Science. [Extracted from the article]

The journal is pleased to publish the abstracts of the six finalists of the 2022 Manufacturing and Service Operations Management Society's student paper competition. The 2022 prize committee was chaired by Florin Ciocan (INSEAD), Ersin Korpeoglu (University College London), and Nikos Trichakis (Massachusetts Institute of Technology). The judges were Adam Elmachtoub, Adem Orsdemir, Agni Orfanoudaki, Alp Akcay, Alper Nakkas, Amrita Kundu, Amy Pan, Andrew Wu, Antoine DESIR, Anyan Qi, Arian Aflaki, Ashish Kabra, Auyon Siddiq, Bilal Gokpinar, Bob Batt, Bora Keskin, Can Zhang, Dan Iancu, Dan Iancu, Daniel Freund, Daniel Lin, Daniela Saban, David Drake, Dawson Kaaua, Ekaterina Astashkina, Elena Belavina, Elodie Adida, Emre Nadar, Fabian Sting, Fanyin Zheng, Fei Gao, Georgina Hall, Gizem Korpeoglu, Gonzalo Romero, Guoming Lai, Hessam Bavafa, Hummy Song, Ioannis (Yannis) Bellos, Ioannis Stamatopoulos, Iris Wang, Itir Karaesmen, Jiankun Sun, Jiankun Sun, Jiaru Bai, Jiayi Joey Yu, Jing Wu, Joel Wooten, John Silberholz, Jonathan Helm, Jose Guajardo, Karen Zheng, Ken Moon, Kenan Arifoglu, Kimon Drakopoulos, Kostas Bimpikis, Lennart Baardman, Lina Song, Luyi Gui, Luyi Yang, Miao Bai, Mika Sumida, Ming Hu, Mumin Kurtulus, Nazli Sonmez, Negin Golrezaei, Nektarios Oraiopoulos, Nil Karacaoglu, Nitin Bakshi, Nitish Jain, Nur Sunar, Olga Perdikaki, Ovunc Yilmaz, Ozan Candogan, Panos Markou, Pengyi Shi, Philip Zhang, Philipp Cornelius, Qi (George) Chen, Qiuping Yu, Ruslan Momot, Ruth Beer, S. Alex Yang, Saed Alizamir, Safak Yucel, Sanjith Gopalakrishnan, Santiago Gallino, Sarah Yini Gao, Scott Rodilitz, Sebastien Martin, Sheng Liu, Shouqiang Wang, Simone Marinesi, Sina Khorasani, So Yeon CHUN, Somya Singhvi, Soo-Haeng Cho, Soroush Saghafian, Sriram Dasu, Stefanus Jasin, Stephen Leider, Tian Chan, Tim Kraft, Tom Tan, Vasiliki Kostami, Velibor Misic, Vishal Agrawal, Xiaojia Guo, Xiaoshan Peng, Xiaoshuai Fan, Xiaoyang Long, Yangfang (Helen) Zhou, Yasemin Limon, Yehua Wei, Ying-Ju Chen, Yonatan Gur, Yuqian Xu, Zhaohui (Zoey) Jiang, Zhaowei She, and Zumbul Atan. [ABSTRACT FROM AUTHOR]

Analyzing Capacitated Two-Echelon Systems with Permutation-Dependent Separability Capacitated multiechelon systems are common in practice due to the escalating costs of labor and advanced manufacturing technology. However, identifying the optimal replenishment policies for such systems is a largely open area of research due to the intrinsic complexity, especially when there is an upstream bottleneck. In "A Permutation-Dependent Separability Approach for Capacitated Two-Echelon Inventory Systems", Shen, Yu, and Huh propose a new approach, that is, permutation-dependent separability, to tackle a capacitated two-echelon system in which the capacity of upstream stage can be the bottleneck. They show that the value function for the capacitated two-echelon system in each period is permutation-dependent separable, and that for each echelon, a permutation-dependent echelon base stock policy is optimal. The authors also develop efficient solution procedures on how to obtain the optimal policy. We consider optimal inventory replenishment policies for capacitated 2-echelon serial inventory systems, where the capacity of upstream echelon can be the bottleneck. We show that the optimal replenishment decisions in each period can be made one echelon at a time by introducing a procedure that can sequentially decompose a multidimensional optimization problem to a series of one-dimensional problems. We also introduce the notion of permutation-dependent separability. A permutation-dependent separable function is a function that can be decomposed as a sum of single-variable component functions under each nondecreasing order of variables. We find that the value function for the capacitated 2-echelon system in each period is permutation-dependent separable, and that, for each echelon, a permutation-dependent echelon base stock policy is optimal. [ABSTRACT FROM AUTHOR]

The journal is pleased to publish the abstracts of the six finalists of the 2021 Manufacturing and Service Operations Management Society's student paper competition. The 2021 prize committee was chaired by Vishal Agrawal (Georgetown University), Florin Ciocan (INSEAD), and Yanchong Zheng (Massachusetts Institute of Technology). The judges were Adam Elmachtoub, Adem Orsdemir, Amrita Kundu, Antoine Desir, Anyan Qi, Arian Aflaki, Arzum Akkas, Ashish Kabra, Bin Hu, Bora Keskin, Brent Moritz, Can Zhang, Chloe Kim Glaeser, Dan Iancu, Daniel Freund, Daniel Lin, Daniela Saban, David F. Drake, Dawson Kaaua, Divya Singhvi, Ekaterina Astashkina, Elena Belavina, Elodie Adida, Enis Kayis, Ersin Korpeoglu, Evgeny Kagan, Fabian Sting, Fanyin Zheng, Fei Gao, Fernanda Bravo, Francisco Castro, Georgina Hall, Gonzalo Romero, Guangwen Kong, Guoming Lai, Hamsa Bastani, Hessam Bavafa, Hummy Song, Ioannis (Yannis) Stamatopoulos, Ioannis Bellos, Iris Wang, Jake Feldman, Jason Acimovic, Jiankun Sun, Jiaru Bai, John Silberholz, Joline Uichanco, Jonas Oddur Jonasson, Jose Guajardo, Kaitlin Daniels, Kenan Arifoglu, Lennart Baardman, Leon Valdes, Lesley Meng, Luyi Gui, Luyi Yang, Mary Parkinson, Mazhar Arikan, Mehmet Ayvaci, Miao Bai, Michael Freeman, Ming Hu, Morvarid Rahmani, Mumin Kurtulus, Nan Yang, Nektarios Oraiopoulos, Nikhil Garg, Nil Karacaoglu, Nitin Bakshi, Nur Sunar, Olga Perdikaki, Ovunc Yilmaz, Ozan Candogan, Ozge Sahin, Panos Markou, Pascale Crama, Pengyi Shi, Pnina Feldman, Qiuping Yu, Renyu Zhang, Ruslan Momot, Ruth Beer, Ruxian Wang, Saed Alizamir, Safak Yucel, Samantha Keppler, Sanjith Gopalakrishnan, Santiago Gallino, Sarah Yini Gao, Sebastien Martin, Serdar Simsek, Seyed Emadi, Shima Nassiri, Shouqiang Wang, Siddharth Singh, Simone Marinesi, So Yeon Chun, Somya Singhvi, Song-Hee Kim, Soo-Haeng Cho, Soroush Saghafian, Sriram Dasu, Stefanus Jasin, Stephen Leider, Suresh Muthulingam, Suvrat Dhanorkar, Tian Chan, Tim Kraft, Tom TAN, Tugce Martagan, Velibor Misic, Vishal Gupta, Weiming Zhu, Xiajun Amy Pan, Xiaoshan Peng, Xiaoyang Long, Yangfang (Helen) Zhou, Yehua Wei, Yiangos Papanastasiou, Ying-Ju Chen, Yinghao Zhang, Yoni Gur, Yuqian Xu, Zhaohui (Zoey) Jiang, Zumbul Atan. [ABSTRACT FROM AUTHOR]

The article presents the discussion on 21st century being termed as the century of disasters. Topics include blockchain for disaster logistics, wearable technologies for rescue teams and victims, and remote monitoring sensors for intelligence gathering; and Geographic Information System (GIS) for crisis mapping, social media for disaster information dissemination, and extraction of real-time intelligence.

The journal is pleased to publish the abstracts of the six finalists of the 2019 Manufacturing and Service Operations Management Society's student paper competition. The 2019 prize committee was chaired by Feryal Erhun (University of Cambridge), Antonio Moreno (Harvard University), and Yi Xu (University of Maryland). The other committee members were Elodie Adida, Vishal Agrawal, Arzum Akkaş, Mazhar Arıkan, Jiaru Bai, Gah-Yi Ban, Hamsa Bastani, Bob Batt, Elena Belavina, Ioannis Bellos, Kostas Bimpikis, Fernanda Bravo, Robert Bray, Eduard Calvo, Ozan Candogan, Tian Chan, Ying-Ju Chen, Soo-Haeng Cho, So Yeon Chun, Florin Ciocan, Pascale Crama, Ruomeng Cui, Kaitlin Daniels, Kris Ferreira, Santiago Gallino, Esma Gel, Chloe Kim Glaeser, Xiting Gong, Jose Guajardo, Ming Hu, Dan Iancu, Stefanus Jasin, Houyuan Jiang, Ashish Kabra, Itir Karaesmen Aydin, Enis Kayış, Diwas KC, Bora Keskin, Song-Hee Kim, Tim Kraft, Mirko Kremer, Mümin Kurtuluş, Guoming Lai, Daniel Lin, Fang Liu, Velibor Mišić, Suresh Muthulingam, Aris Oraiopoulos, Adem Orsdemir, Yiangos Papanastasiou, Chris Parker, Olga Perdikaki, Anyan Qi, Morvarid Rahmani, Guillaume Roels, Soroush Saghafian, Ozge Sahin, Burhan Sandıkçı, Juan Serpa, Pengyi Shi, Hummy Song, Brad Staats, Yannis Stamatopoulos, Sandra Sulz, Nur Sunar, Nicos Trichakis, John Turner, Jingqi Wang, Ruxian Wang, Shouqiang Wang, Yehua Wei, Wenqiang Xiao, Linwei Xin, Nan Yang, Renyue Zhang, Karen Zheng, and Weiming Zhu. [ABSTRACT FROM AUTHOR]

This paper introduces a new decision support tool to optimize advertising campaigns for new product launches based on learnings from an ex post analysis of prior campaigns. The launch of a new product is one of the most critical activities that product and brand managers are faced with. It requires a substantial communications budget to introduce the new product to the market. As the number of media channels proliferates, however, managers are increasingly held accountable to demonstrate the efficient use of resources. This article introduces a new decision support tool to optimize advertising campaigns for new product launches based on lessons learned from an ex post analysis of prior campaigns. The tool builds on a distinct data collection approach combined with econometric modeling to produce advertising elasticities, which is the key information in the media mix optimization. The approach was implemented at Mercedes-Benz and applied to four major new car launches in Germany in 2012 and 2013. It revealed estimated savings of 15%–30% or EUR 2 million per campaign from a more efficient use of resources. Data files and the online appendix are available at https://doi.org/10.1287/mksc.2018.1136. [ABSTRACT FROM AUTHOR]

Inventors cannot exploit new scientific discoveries if they do not pay attention to them. However, allocating attention to science is difficult because the scientific literature is complex, vast, fast-changing, and often unreliable. Inventors are therefore likely to rely on informational cues when screening new publications. I posit that inventors pay significantly less attention to discoveries "made in academia" than to those "made in industry" because they believe that the work of academic scientists will be less useful to them. I test this proposition by examining inventors' patent references to the scientific literature in the case of simultaneous discoveries made by at least one team based in academia and another based in industry. I find that inventors are 23% less likely to cite the academic paper than its "twin" from industry. My results highlight the importance of inventors' attention as a previously underexplored bottleneck shaping the translation of science into new technologies. [ABSTRACT FROM AUTHOR]

In this article, the author discusses various topics related to publication of the journal. Topics include changes in research areas of management science such as analytics; criteria for submission of research papers to the journal, such as research relevant to the management science community; and different departments related to selection of papers, like departments of Behavioral Economics, and Judgment and Decision Making.

The journal is pleased to publish the abstracts of the six finalists of the 2020 Manufacturing and Service Operations Management Society's student paper competition. The 2020 prize committee was chaired by Vishal Agrawal (Georgetown), Feryal Erhun (University of Cambridge), and Jun Li (University of Michigan). The judges were Adem Orsdemir, Antoine Desir, Anton Ovchinnikov, Anyan Qi, Arian Aflaki, Arzum Akkas, Ashish Kabra, Bin Hu, Bob Batt, Bora Keskin, Can Zhang, Carri Chan, Chloe Kim Glaeser, Daniel Lin, Eduard Calvo, Ekaterina Astashkina, Elena Belavina, Elodie Adida, Enis Kayış, Ersin Korpeoglu, Fabian Sting, Fang Liu, Fanyin Zheng, Fei Gao, Florin Ciocan, Gah-Yi Ban, Gizem Korpeoglu, Guihua Wang, Guillaume Roels, Guoming Lai, Hessam Bavafa, Hummy Song, Ioannis (Yannis) Stamatopoulos, Ioannis Bellos, Iris Wang, Itir Karaesmen, Jiankun Sun, Jiaru Bai, Jing Wu, Joann de Zegher, Joel Wooten, John Silberholz, Jose Guajardo, Kaitlin Daniels, Karen Zheng, Ken Moon, Kenan Arifoglu, Lennart Baardman, Leon Valdes, Lesley Meng, Linwei Xin, Luyi Gui, Luyi Yang, Mary Parkinson, Mazhar Arikan, Michael Freeman, Ming Hu, Morvarid Rahmani, Mumin Kurtulus, Nan Yang, Necati Tereyagoglu, Nektarios Oraiopoulos, Nikos Trichakis, Nil Karacaoglu, Nitin Bakshi, Niyazi Taneri, Nur Sunar, Olga Perdikaki, Ovunc Yilmaz, Ozan Candogan, Ozge Sahin, Panos Markou, Pascale Crama, Pengyi Shi, Pnina Feldman, Qiuping Yu, Renyu (Philip) Zhang, Robert Bray, Ruslan Momot, Ruxian Wang, Saed Alizamir, Safak Yucel, Samantha Keppler, Santiago Gallino, Serdar Simsek, Seyed Emadi, Shiliang (John) Cui, Shouqiang Wang, Simone Marinesi, So Yeon Chun, Song-Hee Kim, Soo-Haeng Cho, Soroush Saghafian, Stefanus Jasin, Suresh Muthulingam, Suvrat Dhanorkar, Tian Chan, Tim Kraft, Tom Tan, Tugce Martagan, Velibor Misic, Weiming Zhu, Xiaoshan Peng, Xiaoyang Long, Yasemin Limon, Yehua Wei, Yiangos Papanastasiou, Ying-Ju Chen, and Zumbul Atan. [ABSTRACT FROM AUTHOR]

Tenure decisions, key decisions in academic institutions, are primarily based on subjective assessments of candidates. Using a large-scale bibliometric database containing 198,310 papers published 1975-2012 in the field of operations research (OR), we propose prediction models of whether a scholar would perform well on a number of future success metrics using statistical models trained with data from the scholar's first five years of publication, a subset of the information available to tenure committees. These models, which use network centrality of the citation network, coauthorship network, and a dual network combining the two, significantly outperform simple predictive models based on citation counts alone. Using a data set of the 54 scholars who obtained a Ph.D. after 1995 and held an assistant professorship at a top-10 OR program in 2003 or earlier, these statistical models, using data up to five years after the scholar became an assistant professor and constrained to tenure the same number of candidates as tenure committees did, made a different decision than the tenure committees for 16 (30%) of the candidates. This resulted in a set of scholars with significantly better future A-journal paper counts, citation counts, and h-indexes than the scholars actually selected by tenure committees. These results show that analytics can complement the tenure decision-making process in academia and improve the prediction of academic impact. [ABSTRACT FROM AUTHOR]

Marketing Science has introduced a new section "Marketing Science: Frontiers," focused on publishing timely research with high potential for impact. The section is positioned as "different, but equal" relative to regular Marketing Science with the same high quality standards, but differentiated contribution criteria and a shorter Science-like format to highlight the core contribution and maximize readability and impact. The section will encourage competition among authors for publishing timely and contemporaneously relevant research--undervalued attributes in traditional contribution evaluation--on topics with high impact potential. In exchange, it will accept papers that make major contributions on one "primary" dimension (methodological, modeling or substantive), with more relaxed thresholds on the non-primary dimensions compared to traditional top journals, and offer faster reviews and time to print. Authors benefit from the promise of first-mover impact rewards, while the field benefits from faster entry and a larger volume of novel, timely and relevant ideas. The section will have a distinct editorial structure and a one round conditional accept/out review process. The editorial elaborates on the purpose of the section, its editorial structure and publication process. [ABSTRACT FROM AUTHOR]

The journal is pleased to publish the abstracts of the six finalists of the 2016 Manufacturing and Service Operations Management Society's student paper competition. The 2016 prize committee was chaired by Sameer Hasija (INSEAD), Nicos Savva (London Business School), and Tolga Tezcan (London Business School). The other committee members were: Philipp Afeche, Vishal Agrawal, Aydin Alptekinoglu, Dimitrios Antritsos, Nilay Argon, Mazhar Arikan, Alessandro Arlotto, Arash Asadpour, Atalay Atasu, Nitin Bakshi, Gah-Yi Ban, Opher Baron, Robert Batt, Elena Belavina, Omar Besbes, Kostas Bimpikis, Robert Bray, Carri Chan, Xin Chen, Ying-Ju Chen, Soo-Haeng Cho, So Yeon Chun, Florin Ciocan, Sarang Deo, Lingxiu Dong, Santiago Gallino, Srinagesh Gavirneni, Manu Goyal, Shuangchi He, Jonathan Helm, Ming Hu, Dan Iancu, Foad Iravani, Srikanth Jagabathula, Nitish Jain, Yash Kanoria, Fikri Karaesmen, Diwas KC, Saravanan Kesavan, Bora Keskin, Sang Kim, Song-Hee Kim, Pascale Krama, Mirko Kremer, Harish Krishnan, Mumin Kurtulus, Guoming Lai, Cuihong Li, Jun Li, Lauren Lu, James Luedtke, Victor Martinez de Albinez, Mili Mehrotra, Alex Mills, Toni Moreno, Nektarios Oraiopoulos, Anton Ovchinnikov, Yiangos Papanastasiou, Chris Parker, Rodney Parker, Ali Parlakturk, Alfonso Pedraza Martinez, Ramandeep Randhawa, Paat Rusmevichientong, Chris Ryan, Soroush Saghafian, Ozge Sahin, Burhaneddin Sandikci, Nicola Secomandi, Stephen Shechter, Pengyi Shi, Amitabh Sinha, Milind Sohoni, Brad Staats, Alireza Tahbaz-Salehi, Nicos Trichakis, Owen Wu, Wenqiang Xiao, Nan Yang, Fuqiang Zhang, Jiawei Zhang, Yao Zhao, Karen Zheng, Yong-Pin Zhou, and Leon Zhu. [ABSTRACT FROM AUTHOR]

The article offers information on the journal's call for papers for the special issue on human-algorithm relationships, with topics like algorithm aversion, customer satisfaction in machine-driven communication, and algorithm transparency.

The Micali–Vazirani (MV) algorithm for finding a maximum cardinality matching in general graphs, which was published in 1980, remains to this day the most efficient known algorithm for the problem. The current paper gives the first complete and correct proof of this algorithm. The MV algorithm resorts to finding minimum-length augmenting paths. However, such paths fail to satisfy an elementary property, called breadth first search honesty in this paper. In the absence of this property, an exponential time algorithm appears to be called for—just for finding one such path. On the other hand, the MV algorithm accomplishes this and additional tasks in linear time. The saving grace is the various "footholds" offered by the underlying structure, which the algorithm uses in order to perform its key tasks efficiently. The theory expounded in this paper elucidates this rich structure and yields a proof of correctness of the algorithm. It may also be of independent interest as a set of well-knit graph-theoretic facts. Funding: This work was supported in part by the National Science Foundation [Grant CCF-2230414]. [ABSTRACT FROM AUTHOR]

This paper develops algorithms to solve strong-substitutes product-mix auctions: it finds competitive equilibrium prices and quantities for agents who use this auction's bidding language to truthfully express their strong-substitutes preferences over an arbitrary number of goods, each of which is available in multiple discrete units. Our use of the bidding language and the information it provides contrasts with existing algorithms that rely on access to a valuation or demand oracle. We compute market-clearing prices using algorithms that apply existing submodular minimization methods. Allocating the supply among the bidders at these prices then requires solving a novel constrained matching problem. Our algorithm iteratively simplifies the allocation problem, perturbing bids and prices in a way that resolves tie-breaking choices created by bids that can be accepted on more than one good. We provide practical running time bounds on both price finding and allocation and illustrate experimentally that our allocation mechanism is practical. Funding: E. Baldwin and P. Klemperer were supported by the Economic and Social Research Council [Grant ES/L003058/1]. P. W. Goldberg and E. Lock were supported by a JP Morgan faculty fellowship during the work on the final version of the paper. Supplemental Material: The online companion is available at https://doi.org/10.1287/moor.2019.0248. [ABSTRACT FROM AUTHOR]

Diagnostic errors are common and can result in serious patient harm. Making the right diagnosis often requires significant diagnostic effort. Yet most physician payment schemes are procedure based and do not account for diagnostic effort or accuracy because of observability issues. In this paper, we develop a parsimonious model to examine the impact of a physician payment scheme on a physician's decisions to (1) exert diagnostic effort and (2) perform a confirmatory test. High effort provides an informative (though imperfect) signal of the patient's true state; the test is confirmatory in that it is a prerequisite for diagnosing a severe condition. Our model uses a two-step diagnostic process to capture the interaction between the physician's diagnostic effort and testing decisions. We show that under a fee-for-service payment scheme, the physician may view the diagnostic effort and the confirmatory test as either complementary or substitutive, depending on the additional revenue from testing. We also reveal nonmonotonic properties such that a more patient-centered physician may not exert more effort or provide a more accurate diagnosis. In addition, either a flat or differentiated payment scheme may be optimal. We also show that an alternative payment scheme, under which the revenue from the confirmatory test is contingent on its result, can induce the social optimum under certain conditions. With the advent of artificial intelligence as part of the standard of care and its increasing use as confirmatory tests, our research has implications for the design of physician payment systems in light of concerns about the potential erosion of individual attention. This paper was accepted by Jayashankar Swaminathan, operations management. Funding: T. Dai was supported by the Johns Hopkins Discovery Award (2022–2024) and the Hopkins Business of Health Initiative Seed Grant (2022–2023). Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.4937. [ABSTRACT FROM AUTHOR]

This paper shows how convex incentives in vertical contracts between manufacturers and retailers can induce sales behavior with costs to consumers. We examine this problem in the automotive sector, where manufacturers commonly motivate new vehicle sales through dealer incentive programs with large discrete bonuses determined by monthly sales targets. Using subprime car loans from over 3,500 dealerships, we document high default rates on new car loans originated at the end of the month—the period when dealerships attempt to secure target-based bonuses by intensifying efforts to sell new cars. We provide evidence consistent with the observed higher default rates resulting from customers purchasing new vehicles at month end. New car purchases stretch borrower budgets and expose borrowers to rapid depreciation, which consigns the borrower with negative equity through much of the loan term. Our results imply that the quartile of customers with the highest payment-to-income ratio see default rates increase from 13.6% to 19.7% on the last day of the month. Although consumers bear high costs from increased defaults, we find no evidence that lenders that purchase the loans are hurt by the default increase. Our results demonstrate how the behaviors induced by convex incentive schemes for sales are borne by customers. This paper was accepted by David Simchi-Levi, business strategy. Supplemental Material: The data files and online appendix are available at https://doi.org/10.1287/mnsc.2023.4935. [ABSTRACT FROM AUTHOR]

In this paper, we study the design and analysis of experiments conducted on a set of units over multiple time periods in which the starting time of the treatment may vary by unit. The design problem involves selecting an initial treatment time for each unit in order to most precisely estimate both the instantaneous and cumulative effects of the treatment. We first consider nonadaptive experiments, in which all treatment assignment decisions are made prior to the start of the experiment. For this case, we show that the optimization problem is generally NP-hard, and we propose a near-optimal solution. Under this solution, the fraction entering treatment each period is initially low, then high, and finally low again. Next, we study an adaptive experimental design problem, in which both the decision to continue the experiment and treatment assignment decisions are updated after each period's data are collected. For the adaptive case, we propose a new algorithm, the precision-guided adaptive experiment algorithm, which addresses the challenges at both the design stage and the stage of estimating treatment effects, ensuring valid post-experiment inference, accounting for the adaptive nature of the design. Using realistic settings, we demonstrate that our proposed solutions can reduce the opportunity cost of the experiments by more than 50%, compared with static design benchmarks. This paper was accepted by George Shanthikumar, data science. Funding: S. Athey and G. Imbens were supported by the Office of Naval Research [Grant N00014-19-1-2468]. M. Bayati was supported by the National Science Foundation [Grant CMMI 1554140]. Supplemental Material: The e-companion and data are available at https://doi.org/10.1287/mnsc.2023.4928. [ABSTRACT FROM AUTHOR]

In a wide variety of applications, including healthcare, bidding in first price auctions, digital recommendations, and online education, it can be beneficial to learn a policy that assigns treatments to individuals based on their characteristics. The growing policy-learning literature focuses on settings in which policies are learned from historical data in which the treatment assignment rule is fixed throughout the data-collection period. However, adaptive data collection is becoming more common in practice from two primary sources: (1) data collected from adaptive experiments that are designed to improve inferential efficiency and (2) data collected from production systems that progressively evolve an operational policy to improve performance over time (e.g., contextual bandits). Yet adaptivity complicates the problem of learning an optimal policy ex post for two reasons: first, samples are dependent and, second, an adaptive assignment rule may not assign each treatment to each type of individual sufficiently often. In this paper, we address these challenges. We propose an algorithm based on generalized augmented inverse propensity weighted (AIPW) estimators, which nonuniformly reweight the elements of a standard AIPW estimator to control worst case estimation variance. We establish a finite-sample regret upper bound for our algorithm and complement it with a regret lower bound that quantifies the fundamental difficulty of policy learning with adaptive data. When equipped with the best weighting scheme, our algorithm achieves minimax rate-optimal regret guarantees even with diminishing exploration. Finally, we demonstrate our algorithm's effectiveness using both synthetic data and public benchmark data sets. This paper was accepted by Hamid Nazerzadeh, data science. Funding: This work is supported by the National Science Foundation [Grant CCF-2106508]. R. Zhan was supported by Golub Capital and the Michael Yao and Sara Keying Dai AI and Digital Technology Fund. Z. Ren was supported by the Office of Naval Research [Grant N00014-20-1-2337]. S. Athey was supported by the Office of Naval Research [Grant N00014-19-1-2468]. Z. Zhou is generously supported by the New York University's 2022–2023 Center for Global Economy and Business faculty research grant and the Digital Twin research grant from Bain & Company. Supplemental Material: The data files are available at https://doi.org/10.1287/mnsc.2023.4921. [ABSTRACT FROM AUTHOR]

Using two proxies for investors' political affiliation, we document sharp differences in stock returns between firms likely dominated by Democratic investors (blue stocks) and those dominated by Republican investors (red stocks) during the COVID pandemic. Red stocks have 20 basis points higher risk-adjusted returns than blue stocks on COVID news days (Partisan Return Gap). Lockdown policies, COVID cases, industry and firm fundamentals only explain at most 40% of the return gap. Polarized political beliefs about COVID, revealed through people's social distancing behavior, contribute to about 40% of the return gap beyond the fundamental channel. Our paper provides partisanship as a novel aspect in understanding abnormal stock returns during the pandemic. This paper was accepted by Lukas Schmid, finance. Supplemental Material: The data and e-companion are available at https://doi.org/10.1287/mnsc.2023.4913. [ABSTRACT FROM AUTHOR]

Random coefficient multinomial logit models are widely used to estimate customer preferences from sales data. However, these estimation models can only allow for products with positive sales; this selection leads to highly biased estimates in long tail markets, that is, markets where many products have zero or low sales. Such markets are increasingly common in areas such as online retail and other online marketplaces. In this paper, we propose a two-stage estimator that uses machine learning to correct for this bias. Our method first uses deep learning to predict the market shares of all products, where the neural network's structure mirrors the random coefficient multinomial logit model's data generating process. In the second stage, we use the predictions of the first stage to reweight the observed shares in a way that corrects for the induced bias and maintains the causal interpretation of the structural model. We show that the estimated parameters are consistent in the number of markets. Our method performs well on simulated and real long tail data, producing accurate estimates of customer behavior. These improved estimates can subsequently be used to provide prescriptive policy recommendations on important managerial decisions such as pricing, assortment, and so on. This paper was accepted by Gabriel Weintraub, revenue management and market analytics. Supplemental Material: The data files are available at https://doi.org/10.1287/mnsc.2023.4893. [ABSTRACT FROM AUTHOR]

Limited access to large-scale data is a key obstacle to building machine learning (ML) applications in practice, partly due to a reluctance of information exchange among data owners out of privacy and data security concerns. To address this "information silo" problem, federated learning (FL) techniques have been proposed to enable decentralized model training via an orchestrating central server and have received increasing attention in several industries (including healthcare and finance). Despite its superior privacy protection property, adoption of FL is limited by a lack of systematic understanding of its underlying economics. In this paper, we take an analytical approach to answer two questions: (1) when do data owners prefer to form a FL partnership over building ML models by themselves and (2) how can different contractual mechanisms be used to promote repeated contributions to FL (the cooperative outcome that benefits all participants). We formulate an iterated prisoner's dilemma (IPD) model that accounts for unique FL characteristics, including the specification of the payoff matrix and the involvement of a central server to sanction noncooperation. We find that partnership formation requires participants to be not too forward-looking in temporal preferences, which is contrary to the conventional wisdom in IPD. Furthermore, to promote repeated contributions, it is insufficient to only rely on penalties imposed by the central server or by participants for noncooperation, but a combination of both is enough. Our work advances theoretical understanding of the economics of FL and provides prescriptive insights that can inform FL participant selection and contract design. This paper was accepted by D. J. Wu, information systems. Funding: This work was partially supported by Cisco Research. Supplemental Material: The online appendices are available at https://doi.org/10.1287/mnsc.2023.00611. [ABSTRACT FROM AUTHOR]

A challenge in same-day delivery operations is that delivery requests are typically not known beforehand, but are instead revealed dynamically during the day. This uncertainty introduces a trade-off between dispatching vehicles to serve requests as soon as they are revealed to ensure timely delivery and delaying the dispatching decision to consolidate routing decisions with future, currently unknown requests. In this paper, we study the dynamic dispatch waves problem, a same-day delivery problem in which vehicles are dispatched at fixed decision moments. At each decision moment, the system operator must decide which of the known requests to dispatch and how to route these dispatched requests. The operator's goal is to minimize the total routing cost while ensuring that all requests are served on time. We propose iterative conditional dispatch (ICD), an iterative solution construction procedure based on a sample scenario approach. ICD iteratively solves sample scenarios to classify requests to be dispatched, postponed, or undecided. The set of undecided requests shrinks in each iteration until a final dispatching decision is made in the last iteration. We develop two variants of ICD: one variant based on thresholds, and another variant based on similarity. A significant strength of ICD is that it is conceptually simple and easy to implement. This simplicity does not harm performance: through rigorous numerical experiments, we show that both variants efficiently navigate the large state and action spaces of the dynamic dispatch waves problem and quickly converge to a high-quality solution. Finally, we demonstrate that the threshold-based ICD variant achieves excellent results on instances from the EURO Meets NeurIPS 2022 Vehicle Routing Competition, nearly matching the performance of the winning machine learning–based strategy. History: This paper has been accepted for the Transportation Science Special Issue on DIMACS Implementation Challenge: Vehicle Routing Problems. Funding: This work was supported by TKI Dinalog, Topsector Logistics, and the Dutch Ministry of Economic Affairs and Climate Policy. Supplemental Material: The online appendix is available at https://doi.org/10.1287/trsc.2023.0111. [ABSTRACT FROM AUTHOR]

This paper describes an exact branch-and-cut (B&C) algorithm for the split delivery vehicle routing problem. The underlying model is based on a previously proposed two-index vehicle flow formulation that models a relaxation of the problem. We dynamically separate two well-known classes of valid inequalities, namely capacity and connectivity cuts, and use an in-out algorithm to improve the convergence of the cutting phase. We generate no-good cuts from feasible integer solutions to the relaxation using a recently proposed single-commodity flow formulation in the literature. The exact methodology is complemented by a very effective adaptive large neighborhood search (ALNS) heuristic that provides high-quality upper bounds to initiate the B&C algorithm. Key ingredients in the design of the heuristic include the use of a tailored construction algorithm, which can exploit the situation in which the ratio of the number of customers to the minimum number of vehicles needed is low, and the use of a route-based formulation to improve the solutions found before, during, and after the ALNS procedure. An earlier version of this work was submitted to the DIMACS (Center for Discrete Mathematics and Theoretical Computer Science) implementation challenge, where it placed third. On sets of well-known benchmark instances for limited and unlimited fleet variants of the problem, we demonstrate that the heuristic provides very competitive solutions, with respective average gaps of 0.19% and 0.18% from best-known values. Furthermore, the exact B&C framework is also highly competitive with state-of-the-art methods, providing solutions with an average optimality gap of 1.82%. History: This paper has been accepted for the Transportation Science Special Section on DIMACS Implementation Challenge: Vehicle Routing Problems. Supplemental Material: The online appendices are available at https://doi.org/10.1287/trsc.2022.0353. [ABSTRACT FROM AUTHOR]

Production and inventory routing problems consider a single-product supply chain operating under a vendor-managed inventory system. A plant creates a production plan and vehicle routes over a planning horizon to replenish its customers at minimum cost. In this paper, we present two- and three-index formulations, implement a branch-and-cut algorithm based on each formulation, and introduce a local search matheuristic-based algorithm to solve the problem. In order to combine all benefits of each algorithm, we design a parallel framework to integrate all three fronts, called the three-front parallel branch-and-cut algorithm (3FP-B&C). We assess the performance of our method on well-known benchmark instances of the inventory routing problem (IRP) and the production routing problem (PRP). The results show that our 3FP-B&C outperforms by far other approaches from the literature. For the 956 feasible small-size IRP instances, our method proves optimality for 746, being the first exact algorithm to solve all instances with up to two vehicles. 3FP-B&C finds 949 best known solutions (BKS) with 153 new BKS (NBKS). For the large-size set, our method provides two new optimal solutions (OPT), and finds 82% of BKS, being 70% of NBKS for instances with up to five vehicles. This result is more than twice the number of BKS considering all heuristic methods from the literature combined. Finally, our 3FP-B&C finds the best lower bounds (BLB) for 1,169/1,316 instances, outperforming all previous exact algorithms. On the PRP, our method obtained 278 OPT out of the 336 instances of benchmark set of small- and medium-size instances being 19 new ones in addition to 335 BKS (74 NBKS) and 313 BLB (52 new ones). On another set of PRP with medium- and large-size instances, our algorithm finds 1,105 BKS out of 1,440 instances with 584 NBKS. Besides that, our 3FP-B&C is the first exact algorithm to solve the instances with an unlimited fleet, providing the first lower bounds for this subset with an average optimality gap of 0.61%. We also address a very large-size instance set, the second exact algorithm to address this set, outperforming the previous approach by far. Finally, a comparative analysis of each front shows the gains of the integrated approach. History: This paper has been accepted for the Transportation Science Special Issue: DIMACS Implementation Challenge: Vehicle Routing. Funding: C. M. Schenekemberg was supported by the São Paulo Research Foundation (FAPESP) [Grant 2020/07145-8]. A. A. Chaves was supported by FAPESP [Grants 2018/15417-8 and 2016/01860-1] and Conselho Nacional de Desenvolvimento Científico e Tecnológico [Grants 312747/2021-7 and 405702/2021-3]. L. C. Coelho was supported by the Canadian Natural Sciences and Engineering Research Council [Grant 2019-00094]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/trsc.2022.0261. [ABSTRACT FROM AUTHOR]

Quantify the uncertainty to decide and explore better In statistical inference, large-sample behavior and confidence interval construction are fundamental in assessing the error and reliability of estimated quantities with respect to the data noises. In the paper "Uncertainty Quantification and Exploration for Reinforcement Learning", Dong, Lam, and Zhu study the large sample behavior in the classic setting of reinforcement learning. They derive appropriate large-sample asymptotic distributions for the state-action value function (Q-value) and optimal value function estimations when data are collected from the underlying Markov chain. This allows one to evaluate the assertiveness of performances among different decisions. The tight uncertainty quantification also facilitates the development of a pure exploration policy by maximizing the worst-case relative discrepancy among the estimated Q-values (ratio of the mean squared difference to the variance). This exploration policy aims to collect informative training data to maximize the probability of learning the optimal reward collecting policy, and it achieves good empirical performance. We investigate statistical uncertainty quantification for reinforcement learning (RL) and its implications in exploration policy. Despite ever-growing literature on RL applications, fundamental questions about inference and error quantification, such as large-sample behaviors, appear to remain quite open. In this paper, we fill in the literature gap by studying the central limit theorem behaviors of estimated Q-values and value functions under various RL settings. In particular, we explicitly identify closed-form expressions of the asymptotic variances, which allow us to efficiently construct asymptotically valid confidence regions for key RL quantities. Furthermore, we utilize these asymptotic expressions to design an effective exploration strategy, which we call Q-value-based Optimal Computing Budget Allocation (Q-OCBA). The policy relies on maximizing the relative discrepancies among the Q-value estimates. Numerical experiments show superior performances of our exploration strategy than other benchmark policies. Funding: This work was supported by the National Science Foundation (1720433). [ABSTRACT FROM AUTHOR]

Dial M for Simulation For years, systems of stochastic differential equations (SDEs) were simulated by discretization, inevitably introducing a bias, which can be difficult to quantify accurately. To circumvent this, some attempts have been made to simulate exactly various models from the SDE solution. These approaches prove capable of producing accurate results. A serious drawback of such an approach, nevertheless, is the implicit need to use extensive numerical methods, which make the entire simulation computationally heavy and quite impracticable. In the paper "Unified moment-based modeling of integrated stochastic processes," Kyriakou, Brignone, and Fusai present a methodological framework based on M(oments) for the simulation of such models that overcomes earlier limitations. Theoretical results and extensive numerical experiments show that the proposed approach allows accurate simulation of complex stochastic models with low computational effort. In this paper, we present a new method for simulating integrals of stochastic processes. We focus on the nontrivial case of time integrals, conditional on the state variable levels at the endpoints of a time interval through a moment-based probability distribution construction. We present different classes of models with important uses in finance, medicine, epidemiology, climatology, bioeconomics, and physics. The method is generally applicable in well-posed moment problem settings. We study its convergence, point out its advantages through a series of numerical experiments, and compare its performance against existing schemes. Supplemental Material: The e-companion is available at https://doi.org/10.1287/opre.2022.2422. [ABSTRACT FROM AUTHOR]

Efficient Learning Algorithms for the Best Capped Base-Stock Policy in Lost Sales Inventory Systems Periodic review, lost sales inventory systems with lead times are notoriously challenging to optimize. Recently, the capped base-stock policy, which places orders to bring the inventory position up to the order-up-to level subject to the order cap, has demonstrated exceptional performance. In the paper "UCB-Type Learning Algorithms with Kaplan–Meier Estimator for Lost Sales Inventory Models with Lead Times," Lyu, Zhang, and Xin propose an upper confidence bound–type learning framework. This framework, which incorporates simulations with the Kaplan–Meier estimator, works with censored demand observations. It can be applied to determine the optimal capped base-stock policy with a tight regret with respect to the planning horizon and the optimal base-stock policy with a regret that matches the best existing result. Both theoretical analysis and extensive numerical experiments demonstrate the effectiveness of the proposed learning framework. In this paper, we consider a classic periodic-review lost-sales inventory system with lead times, which is notoriously challenging to optimize with a wide range of real-world applications. We consider a joint learning and optimization problem in which the decision maker does not know the demand distribution a priori and can only use past sales information (i.e., censored demand). Departing from existing learning algorithms on this learning problem that require the convexity property of the underlying system, we develop an upper confidence bound (UCB)-type learning framework that incorporates simulations with the Kaplan–Meier estimator and demonstrate its applicability to learning not only the optimal capped base-stock policy in which convexity no longer holds, but also the optimal base-stock policy with a regret that matches the best existing result. Compared with a classic multi-armed bandit problem, our problem has unique challenges because of the nature of the inventory system, because (1) each action has long-term impacts on future costs, and (2) the system state space is exponentially large in the lead time. As such, our learning algorithms are not naive adoptions of the classic UCB algorithm; in fact, the design of the simulation steps with the Kaplan–Meier estimator and averaging steps is novel in our algorithms, and the confidence width in the UCB index is also different from the classic one. We prove the regrets of our learning algorithms are tight up to a logarithmic term in the planning horizon T. Our extensive numerical experiments suggest the proposed algorithms (almost) dominate existing learning algorithms. We also demonstrate how to select which learning algorithm to use with limited demand data. Supplemental Material: The online appendix is available at https://doi.org/10.1287/opre.2022.0273. [ABSTRACT FROM AUTHOR]

Anytime Valid Comparison of Sequential Forecasters How do we compare forecasters that each make a probabilistic prediction on a sequence of events (e.g., weather and sports)? In "Comparing Sequential Forecasters," Choe and Ramdas propose flexible approaches to sequential inference on the mean score difference between any two forecasters. To estimate this time-varying quantity, the authors propose a sequential analog of confidence intervals, called confidence sequences (CSs). These CSs correctly cover the score difference under continuous monitoring, and the evaluator can freely peek at the scores to stop the experiment ("anytime valid"). The authors further develop a complementary anytime valid approach called e-processes, which quantify the evidence against the claim that one forecaster is never better than the other in mean scores. The validity of these methods does not depend on stationarity or other assumptions on how the score differences evolve sequentially. In their paper, the authors showcase CSs and e-processes for comparing real-world baseball and weather forecasters. Consider two forecasters, each making a single prediction for a sequence of events over time. We ask a relatively basic question: how might we compare these forecasters, either online or post hoc, avoiding unverifiable assumptions on how the forecasts and outcomes were generated? In this paper, we present a rigorous answer to this question by designing novel sequential inference procedures for estimating the time-varying difference in forecast scores. To do this, we employ confidence sequences (CS), which are sequences of confidence intervals that can be continuously monitored and are valid at arbitrary data-dependent stopping times ("anytime-valid"). The widths of our CSs are adaptive to the underlying variance of the score differences. Underlying their construction is a game-theoretic statistical framework in which we further identify e-processes and p-processes for sequentially testing a weak null hypothesis—whether one forecaster outperforms another on average (rather than always). Our methods do not make distributional assumptions on the forecasts or outcomes; our main theorems apply to any bounded scores, and we later provide alternative methods for unbounded scores. We empirically validate our approaches by comparing real-world baseball and weather forecasters. Funding: A. Ramdas acknowledges funding from the National Science Foundation Division of Mathematical Sciences [Grant 1916320]. Research reported in this paper was sponsored in part by the DEVCOM Army Research Laboratory under Cooperative Agreement W911NF-17-2-0196 (ARL IoBT CRA). Supplemental Material: The e-companion is available at https://doi.org/10.1287/opre.2021.0792. [ABSTRACT FROM AUTHOR]

The stochastic approximation (SA) method stands as the foundational mathematical tool for modern large-scale optimization and machine learning. Therefore, gaining a theoretical understanding of SA algorithms is of fundamental interest. In their paper titled "A Lyapunov Theory for Finite-Sample Guarantees of Markovian Stochastic Approximation," Chen et al. present a unified Lyapunov framework for the finite-sample analysis of a Markovian SA algorithm under a contractive operator with respect to an arbitrary norm. The key novelty lies in the construction of a smooth Lyapunov function called the generalized Moreau envelope. The authors demonstrate the effectiveness of their SA results in the context of reinforcement learning (RL), specifically through popular algorithms such as variants of temporal difference (TD) learning and Q-learning. As byproducts, the results provide theoretical insights into the efficiency of bootstrapping in TD learning with eligibility traces and the bias-variance tradeoff in off-policy learning. This paper develops a unified Lyapunov framework for finite-sample analysis of a Markovian stochastic approximation (SA) algorithm under a contraction operator with respect to an arbitrary norm. The main novelty lies in the construction of a valid Lyapunov function called the generalized Moreau envelope. The smoothness and an approximation property of the generalized Moreau envelope enable us to derive a one-step Lyapunov drift inequality, which is the key to establishing the finite-sample bounds. Our SA result has wide applications, especially in the context of reinforcement learning (RL). Specifically, we show that a large class of value-based RL algorithms can be modeled in the exact form of our Markovian SA algorithm. Therefore, our SA results immediately imply finite-sample guarantees for popular RL algorithms such as n-step temporal difference (TD) learning, TD (λ) , off-policy V-trace, and Q-learning. As byproducts, by analyzing the convergence bounds of n-step TD and TD (λ) , we provide theoretical insight into the problem about the efficiency of bootstrapping. Moreover, our finite-sample bounds of off-policy V-trace explicitly capture the tradeoff between the variance of the stochastic iterates and the bias in the limit. Funding: This work was supported by RTX, the National Science Foundation [Grants 2019844, 2107037, 211247, 2112533, 2144316, and 2240982], and the Machine Learning Laboratory at University of Texas at Austin. Supplemental Material: The online appendix is available at https://doi.org/10.1287/opre.2022.0249. [ABSTRACT FROM AUTHOR]

This paper examines the potential of large language models to mimic human survey respondents and to derive their preferences. We explore the viability of large language models (LLMs), specifically OpenAI's GPT-3.5 and GPT-4, in emulating human survey respondents and eliciting preferences, with a focus on intertemporal choices. Leveraging the extensive literature on intertemporal discounting for benchmarking, we examine responses from LLMs across various languages and compare them with human responses, exploring preferences between smaller, sooner and larger, later rewards. Our findings reveal that both generative pretrained transformer (GPT) models demonstrate less patience than humans, with GPT-3.5 exhibiting a lexicographic preference for earlier rewards unlike human decision makers. Although GPT-4 does not display lexicographic preferences, its measured discount rates are still considerably larger than those found in humans. Interestingly, GPT models show greater patience in languages with weak future tense references, such as German and Mandarin, aligning with the existing literature that suggests a correlation between language structure and intertemporal preferences. We demonstrate how prompting GPT to explain its decisions, a procedure we term "chain-of-thought conjoint," can mitigate, but does not eliminate, discrepancies between LLM and human responses. Although directly eliciting preferences using LLMs may yield misleading results, combining chain-of-thought conjoint with topic modeling aids in hypothesis generation, enabling researchers to explore the underpinnings of preferences. Chain-of-thought conjoint provides a structured framework for marketers to use LLMs to identify potential attributes or factors that can explain preference heterogeneity across different customers and contexts. History: Olivier Toubia served as the senior editor. This paper was accepted through the Marketing Science: Frontiers review process. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mksc.2023.0306. [ABSTRACT FROM AUTHOR]

The paper studies whether and under what conditions policy interventions, such as strengthening censorship, can help a social planner reduce social harm caused by the consumption of sensitive content. Censorship is widely observed in both developed and developing countries across the globe. Many policy makers and platforms filter out certain sensitive content to reduce its consumption and the corresponding social damage. However, censorship cannot perfectly eliminate all sensitive content, and content creators have various information transmission channels to circumvent censorship systems. In this paper, we construct a game-theoretic model to study the creation, consumption, and control of sensitive content. Our paper seeks to answer whether policy interventions, such as strengthening censorship, can help a social planner reduce social harm caused by the consumption of sensitive content. First, when censorship strength is at a low or high level, tightening censorship limits the spread of sensitive content; however, when censorship strength is at an intermediate level, strengthening censorship can lead to higher social harm. This is because stricter censorship can motivate creators to choose a more harmful, secret channel rather than a less harmful, open channel to transmit messages. The larger consumption volume of sensitive content in the secret channel can generate greater social damage. Our first result implies that regulators who seek to minimize social damage may take a loose censorship strategy. Second, our paper reveals that raising creators' content creation cost can result in higher damage for the society. Third, when creators need to incur a higher cost upon being censored out, social harm can increase. History: Anthony Dukes served as the senior editor. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mksc.2022.0453. [ABSTRACT FROM AUTHOR]

This paper investigates the interplay between advertising and decentralized content creation on digital content platforms. This paper investigates the interplay between advertising and decentralized content creation on digital content platforms, such as YouTube, TikTok, and Instagram. Unlike traditional media, content on these platforms is produced by numerous creators competing for viewer attention. This study addresses how a platform's advertising policy (specifically, the level of advertising embedded in organic content) and various market factors, including viewers' satiation rates and their distaste for ads, influence decentralized content creation, exploring the ensuing implications for the platform's advertising and revenue-sharing policy design. Using a game-theoretical model, the research reveals that the level of advertising, viewers' satiation rates, and their distaste for ads can have a nonmonotonic impact on the equilibrium content quality. Contrary to conventional wisdom, a low level of advertising could be advantageous for platforms when viewers are inclined to consume high volumes of content, factoring in the strategic influence on content creation. Additionally, revenue sharing with creators depends on factors such as the viewers' satiation rates and the creators' creative costs. This study offers content platforms strategic guidance on refining their advertising and revenue-sharing policies in the creator economy. History: Anthony Dukes served as the senior editor. Funding: This work was supported by the National Natural Science Foundation of China [Grant 72332004]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mksc.2022.0387. [ABSTRACT FROM AUTHOR]

This paper shows that consumers prefer organic listings in the top-ranked positions to sponsored listings of the same product/position in an online marketplace. Sponsored product listings on online marketplaces are third-party sellers' ads blended in organic product listings. This paper investigates a seller's managerial questions: whether a sponsored listing outperforms an organic listing and how the performance varies by positions. Our large-scale field study on a mobile app with experimental and natural variation finds that consumers prefer organic listings in the top-ranked positions to sponsored listings of the same product/position. Consumers become indifferent between sponsored and organic listings in the lower-ranked positions. A mechanism check suggests that top-ranked organic listings are perceived as more credible than seller-subsidized sponsored listings. Despite consumers' preference for organic listings, a simulation analysis shows that an advertising seller may benefit financially if a lower-ranked organic listing can be replaced with a top-ranked sponsored listing. History: Olivier Toubia served as the senior editor. Funding: J. Shi acknowledges financial support from INFORMS Society for Marketing Science Doctoral Dissertation Early Stage Research Grants. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mksc.2021.0293. [ABSTRACT FROM AUTHOR]

In this paper, we introduce the daily berth planning problem for Xiamen Hai-Tian Container Terminal (XHCT) at the Port of Xiamen, China, and propose the development and implementation of a berth planning system. The aim of the berth planning problem is to optimize daily berth plans by considering various decisions, including berth allocation, quay crane assignment, fairway traffic control, and berthing safety requirements. Among these decisions, the berthing safety requirement is a novel but practical problem in berth allocation that concerns the resource allocation related to berthing safety and interrelation with other decisions. A mathematical programming-driven methodological framework is designed with a 0-1 integer linear programming model for problem formulation and a highly efficient decomposition heuristic algorithm for solving the problem. This framework establishes the core for the berth planning system. The adoption of the berth planning system contributes to the increase of container throughput and berth capacity by transforming the planning process of XHCT. Moreover, the mathematical programming-driven daily berth planning informs further intelligent operations development in the Port of Xiamen and other container ports. History: This paper was refereed. Funding: This research was supported by the National Natural Science Foundation of China [Grants 72394360, 72394362, 72025103, 71831008, 72361137001, 72071173, and 72371221]. [ABSTRACT FROM AUTHOR]

Problem definition: Physicians spend more than five hours a day working on Electronic Health Record (EHR) systems and more than an hour doing EHR tasks after the end of the workday. Numerous studies have identified the detrimental effects of excessive EHR use and after-hours work, including physician burnout, physician attrition, and appointment delays. However, EHR time is not purely an exogenous factor because it depends on physician usage behavior that could have important operational consequences. Interestingly, prior literature has not considered this topic rigorously. In this paper, we investigate how physicians' workflow decisions on when to perform EHR tasks affect: (1) total time on EHR and (2) time spent after work. Methodology/results: Our data comprise around 150,000 appointments from 74 physicians from a large Academic Medical Center Family Medicine unit. Our data set contains detailed, process-level time stamps of appointment progression and EHR use. We find that the effect of working on EHR systems depends on whether the work is done before or after an appointment. Pre-appointment EHR work reduces total EHR workload and after-work hours spent on EHR. Post-appointment EHR work reduces after-work hours on EHR but increases total EHR time. We find that increasing idle time between appointments can encourage both pre- and post-appointment EHR work. Managerial implications: Our results not only help us understand the timing and structure of work on secondary tasks more generally but also will help healthcare administrators create EHR workflows and appointment schedules to reduce physician burnout associated with excessive EHR use. History: This paper has been accepted in the Manufacturing & Service Operations Management Frontiers in Operations Initiative. Funding: The research conducted for this paper received partial funding from the Center of Business for Health at the Kenan-Flagler Business School, University of North Carolina at Chapel Hill. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2023.0028. [ABSTRACT FROM AUTHOR]