2016 INFORMS Annual Meeting Program

SB27

INFORMS Nashville – 2016

SB27 201A-MCC Diagnosis Under Uncertainty Sponsored: Manufacturing & Service Oper Mgmt Sponsored Session Chair: Sarang Deo, Indian School of Business, Hyderabad, India, sarang_deo@isb.edu Co-Chair: Tinglong Dai, Johns Hopkins University, Baltimore, MD, United States, dai@jhu.edu 1 - False Diagnosis And Overtreatment In Services Senthil Veeraraghavan, University of Pennsylvania, senthilv@wharton.upenn.edu In many services, consumers must rely on experts to identify the type of service they need. In such service, diagnosis is a crucial step in which the expert identifies the problem and provides the corresponding treatment. The information asymmetry leads to inefficiencies in the form of overtreatment. Overtreatments are expensive but also require more service capacity and time, and thus result in longer delays and higher waiting costs for services. However, we find that such delays act as a natural “fraud cost” and mitigates expert cheating and induce honesty. Experts high capacity utilization are less prone to overtreat. 2 - Conspicuous By Its Absence: Diagnostic Expert Testing Under Uncertainty Tinglong Dai, Assistant Professor, Johns Hopkins University, 100 International Drive, Baltimore, MD, 21202, United States, dai@jhu.edu, Shubhranshu Singh Diagnostic experts, such as medical doctors, are crucial for evaluating the state of the world. All diagnostic experts are not equally competent, and even the best experts are imperfect. We model the decision-making process of a diagnostic expert, who is altruistic but concerned about reputation. Our paper presents interesting insights about the expert’s test-ordering behavior primarily driven by reputation concerns. 3 - Incentizing Less-than-Fully-Qualified Providers For Early Diagnosis Of Tuberculosis In India Sarang Deo, Indian School of Business, sarang_deo@isb.edu Milind Sohoni, Neha Jha A major driver of TB epidemic in India is delay in diagnosis by less-than-fully- qualified providers (LTFQs), who are typically the first point of contact for patients. This work is motivated by pilots funded by international donors to provide monetary incentives to LTFQs to induce earlier referral and diagnosis. Using a game-theoretic model, we show that the optimal structure of the incentive referral contract (whether to pay for all referrals or only for confirmed referrals) depends on the quality of diagnosis of the provider. We calibrate our model results using realistic parameter estimates obtained from primary and secondary data sources. 4 - Medical Guideline Making When Litigation Is A Concern: The Role Of Ubiquitous Health Information Mehmet U Ayvaci, University of Texas-Dallas, 800 W. Campbell Rd. SM33, Richardson, TX, 75080, United States, Mehmet.Ayvaci@utdallas.edu, Yeong In Kim, Srinivasan Raghunathan, Turgay Ayer We examine the optimal formulation of guidelines in a generic health screening with consideration for the physician’s increased liability risk under ubiquitous health information and information technologies. We find that under the litigation concern, the social planner strategically provides imprecise guidelines with vague recommendations regarding which patients should undergo the test while providing precise guidelines regarding the physician’s decisions based on test results. Strategic vagueness in guidelines balances the trade-off between the reduction of defensive medicine and supply of the health service.

SB28 201B-MCC MSOM Student Paper Competition Finalists – II Sponsored: Manufacturing & Service Oper Mgmt Sponsored Session Chair: Sameer Hasija, Insead, 1 Ayer Rajah Avenue, Singapore, 138676, Singapore, sameer.hasija@insead.edu Co-Chair: Tolga Tezcan, London Business School, Regent’s Park, London, NW1 4SA, United Kingdom, ttezcan@london.edu Co-Chair: Nicos Savva, London Business School, Regent’s Park, London, NW1 4SA, United Kingdom, nsavva@london.edu - Economies of Scale and Scope in Hospitals Michael Freeman, University of Cambridge, Cambridge, United Kingdom. mef35@cam.ac.uk Abstract to come 3 - Online Decision-Making with High-Dimensional Covariates Hamsa Bastani, Stanford University, Stanford, CA, bayati@stanford.edu Big data has enabled decision-makers to tailor decisions at the individual-level in a variety of domains such as personalized medicine and online advertising. This involves learning a model of decision rewards conditional on individual-specific covariates. In many practical settings, these covariates are high-dimensional; typically only a small subset of the observed features are predictive of a decision’s success. We formulate this problem as a multi-armed bandit with high- dimensional covariates, and present a new efficient bandit algorithm based on the LASSO estimator. Our regret analysis establishes that our algorithm achieves near-optimal performance in comparison to an oracle that knows all the problem parameters. The key step in our analysis is proving a new oracle inequality that guarantees the convergence of the LASSO estimator despite the non-i.i.d. data induced by the bandit policy. Furthermore, we illustrate the practical relevance of our algorithm by evaluating it on a real-world clinical problem of warfarin dosing. 4 - Real-time Optimization of Personalized Assortments Negin Golrezaei, USC Marshall School of Business, Los Angeles, CA, golrezae@usc.edu Abstract to come SB29 202A-MCC Innovations in the Operations-Marketing Interface Sponsored: Manufacturing & Service Oper Mgmt Sponsored Session Chair: Jose A Guajardo, University of California-Berkeley, Berkeley, CA, United States, jguajardo@berkeley.edu 1 - Does Online Learning Work In Retail? Serguei Netessine, INSEAD, serguei.netessine@insead.edu Marshall L Fisher, Santiago Gallino We partnered with Experticity, a firm that provides online training modules for retail Store Associates, and Dillard’s, a leading department store chain whose more than 50,000 Store Associates had access to the Experticity product training modules. We found that as Store Associates engaged in training over time, their sales rate increased by 1.8 percent for every module taken. We also found that willingness to engage in voluntary training was an indicator of raw talent; those Store Associates who engaged in training were 20 percent more productive prior to any training, and 46 percent more productive after training, than those who took no training. 2 - Business Models In The Sharing Economy: Manufacturing Durable Goods In The Presence Of Peer-to-peer Rental Markets Zhe Zhang, Carnegie Mellon University, 4800 Forbes Avenue, Pittsburgh, PA, 15213, United States, zhezhang@cmu.edu Vibhanshu Abhishek, Jose A Guajardo Business models focusing on providing access to assets rather than on transferring ownership of goods have become an important recent industry trend. Motivated by this trend, this research analyzes the interaction between a manufacturer of durable goods and a peer-to-peer marketplace, characterizing market outcomes under alternative market structures. 2

52