Informs Annual Meeting 2017

TE39

INFORMS Houston – 2017

4 - Resilience-based Freight Transportation Infrastructure Network Design

2 - The Newsvendor with Uncertain Accounts Receivable Emre Berk, Bilkent University, Faculty of Business Administration, Ankara, 06800, Turkey, eberk@bilkent.edu.tr I revisit the classical newsvendor with different customer segments having different likelihoods of not being able to pay for the purchases they have already made. Under various collection scenarios, analytical results are obtained on the optimal production, marketing (promotion) and financial (collection) decisions. Basic trade-offs are discussed via illustrative examples. 3 - Effects of Overconfident Coos on Inventories. Jaeseok Na, KAIST, Seoul, Seoul, 082, Korea, Republic of, jsna@kaist.ac.kr, Bowon Kim, Jeong Eun Sim This study investigates the effects of overconfident Chief Operating Officers (COO) on the corporate inventory strategy of a large number of U.S. manufacturing companies. Overconfident managers tend to overestimate future outcomes; thus we hypothesize that overconfident COOs set their operational strategy more aggressively expecting to increase their sales. Using option-based proxies for the overconfidence, we find that such firms fail to maintain lean strategy. In addition, we suggest the relationship is moderated by market structure. We therefore conclude that understanding the behavioral aspects of managers provides valuable insights in managing inventories. 4 - Optimal Supply Chain Financing Strategies with Loss Aversions: Financing vs. Investment Ye Liu, Central University of Finance and Economics, Beijing, China, yeliuaug@gmail.com, Nina Yan This paper designs two supply chain financing schemes for the capital-constrained retailer, namely, supplier financing and supplier investment. Considering the retailer’s loss aversions, we comparatively analyze the optimal strategies for all participants under these two schemes and investigate the optimal financing choices for the retailer in different cases. 5 - Optimal Commodity Selling using Spot and Forward Markets Behzad Ghafouri, Graduate Research Assistant, University of Western Ontario, 1151 Richmond Street, Western Science Centre, Room 262, London, ON, N6A 5B7, Canada, b.ghafouri@gmail.com, Matt Davison, Somayeh Moazeni During contango oil markets, the offshore crude oil storage trade in which cheaper spot oil is purchased for higher forward sale is popular. The optimal sale of a commodity stored in a rental tanker over a finite horizon is studied. At each time step, the quantity of inventory sold at spot and the forward contract maturity for the remaining inventory are decided. Using the short/long-term Schwartz-Smith price model, the problem is formulated and analyzed as a Markov Decision problem. Approximate dynamic programming policies are investigated. The value of dynamic selling in the forward market and adaptively adjusting the forward maturities are compared to initial spot selling. 6 - A Bilevel Programming Approach to Optimize Credit Term Decisions in Supply Chains We present a new game-theoretic framework to model the problem of optimizing credit-term in a supplier-buyer supply chain. The interaction between supplier’s credit term decision and buyer’s order decision is captured in a Stackelberg game, where the supplier is the leader who determines credit terms for the transaction anticipating certain order quantities from the buyer; then the buyer reacts with its order quantity decision as a follower. A real world case study is presented to show benefit of our optimal credit term solution comparing with two fixed heuristic solutions. Haitao Li, Univ of Missouri-St Louis, College of Business Administration, 229 Express Script Hall, St. Louis, MO, 63121-4400, United States, lihait@umsl.edu

Mohamad Darayi, Pennsylvania State University-Great Valley, 201-F Wadsack Dr., Malvern, PA, United States, mud415@psu.edu, Nazanin Morshedlou, Kash Barker Improving freight transportation infrastructure network design (i) by the addition of capacitated links/nodes or (ii) by increasing the capacity of current network components are explored as options to strengthen the functionality of the network and its effect on the multiple industries relying upon it. Considering different disruption scenarios, this research addresses a network completion decision making problem with respect to the network resilience, as well as multi- industry impacts. 352D The Analytics of Modern Political Campaigning Sponsored: The INFORMS Section on Practice Sponsored Session Chair: Douglas Alan Samuelson, InfoLogix, Inc., Annandale, VA, 22003-3807, United States, samuelsondoug@yahoo.com 1 - Polling, Psychology and Propaganda in Forecasting Electoral Outcomes Dimitris Vayenas, Metapoll, London, United Kingdom, dimitris@metapoll.co.uk Having predicted the outcomes of recent elections the session addresses the underlying theories that are calling for a re-evaluation of political campaigning. By utilising formal models developed to quantify transparency in many-to-many communications, accurate forecasting is attainable based on the motives of the population as addressed by the campaigns; in the case of Brexit with a 0.1% deviation one month before the elections. The shortcomings of polling methodologies and how analytics can help mitigate the impact of unreliable responses are also addressed. Models of human socialty and motivators are employed in the mapping of quantifiable sentiments and reactions to propaganda. 2 - Voting Access and Delay Theodore T. Allen, Ohio State University, 210 Baker Systems, 1971 Neil Ave, Columbus, OH, 43210-1271, United States, allen.515@osu.edu Election officials are often poor users of waiting line analyses including simulations. As a result, they often fail to account for longer ballots in certain locations requiring more resources. Also, they can fail to provision enough ballots so that lines are uniform and minimal. Evidence from elections in New Mexico and North Carolina is discussed. In both cases, there is evidence that one party in power used lines (with somewhat unclear intentionality) to improve its own election results likely, for some elections, changing the outcomes. 3 - Targeting and Virtual Tribalism Douglas Alan Samuelson, InfoLogix, Inc., 8711 Chippendale Court, Annandale, VA, 22003-3807, United States, samuelsondoug@yahoo.com Recent US elections and marketing data indicate that the country is evolving toward a form of virtual tribalism. Virtual tribes are defined by shared values and perceptions rather than kinship, but they work the same way: mutual reinforcement of information and opinions, excluding outsiders. Several political campaigns have exploited this tribal structure to target selected groups of voters with focused messages, especially via non-traditional media. We review these methods and phenomena and discuss some of their implications. 352E Operations/Finance Interface Contributed Session Chair: Haitao Li, Univ of Missouri-St Louis, St Louis, MO, United States, lihait@umsl.edu 1 - Bank’s Effect on Entry of Capital-constrained Firm Guanmei Liu, PhD, Shanghai Jiao Tong University, 1954 Huashan Roda, Shanghai, 200030, China, lgmei2006@sjtu.edu.cn This paper studies the effect of finance on new entry in industry market, when an entrant and an incumbent both need to borrow money from a common bank. We derive two firms’ optimal production quantities and bank’s optimal interest rate in two scenarios: monopoly without entry and competition with entry. We find that bank may not deter entry in one of three cases: 1)two firms are symmetric, 2)incumbent only has advantage in working capital, 3)incumbent has advantage in market size or production cost or both but with more working capital. Interestingly, only when incumbent has no or a little working capital, his larger market size or lower production cost advantage could make bank deter entrant. TE40 TE39

TE41

352F Health Care, Other Contributed Session

Chair: Anna Paula Galvão Scheidegger, Texas A&M University, College Station, TX, United States, apscheidegger@tamu.edu 1 - Preliminary Integrated Approach to Modeling the Prescription and Illicit Opioid Epidemic Jacqueline Garrahan, Healthcare Systems Engineering Institute, Northeastern University, Boston, MA, 02115, United States, j.garrahan@northeastern.edu, Iulian Ilies, James C. Benneyan We formulated a differential equations model of the prescription and illicit opioid co-epidemic and parametrized this model for Massachusetts aggregate and county-level populations as case studies. We optimized parameter values to reproduce historic overdoses by using a grid search approach, identified a prior systemic change leading to increased prevalence, and simulated the consequences of potential interventions. Results identified chronic prescribing and the prescription-to-illicit use pipeline as major drivers of epidemic growth, providing avenues for future research and model expansion.

424