Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

TA57

3 - The Effect of Visibility on Consumer Trust of Social Responsibility Disclosures

1 - Task Selection and Workload Bradley R. Staats, University of North Carolina at Chapel Hill, Campus Box 3490, McColl 4720, Chapel Hill, NC, 27599-3490, United States, Diwas S. KC How individuals manage their tasks is central to operations. Recent research focuses on how increasing workload individuals can increase service time. As the number of tasks increases workers may also manage their workload by a different process - task selection. We theorize and test that under conditions of increased workload individuals may choose to complete easier tasks to manage their load. We label this behavior Task Completion Bias (TCB). Using 2 years of data from an emergency department we find support for TCB and show it improves short-term productivity. However, we find that an overreliance on this task selection strategy hurts performance in the long run. 2 - Heuristic Thinking in Patient Care Diwas S. Kc, Emory University, 1300 Clifton Road, Goizueta Business School, Atlanta, GA, 30322, United States This paper studies heuristic thinking and cognitive bias using a natural experiment from the field. The setting for the study is a set of acute care hospitals, where we examine the care process and discharge decisions for individual patients. Determining a patient’s suitability for discharge is cognitively taxing, calling for the decision maker to draw on up-to-date clinical expertise and detailed information. We postulate that bounded rationality in decision making leads the care provider to substitute clinical readiness for discharge - a more cognitively complex attribute, with a more easily accessible heuristic. 3 - Recovering from Distress: The Impact of Critical Incidents on Operational Performance Jonas Oddur Jonasson, Assistant Professor, MIT Sloan School of Management, 30 Memorial Drive, E62-588, Cambridge, MA, 02142, United States, Hessam Bavafa In service operations settings, where the difficulty of jobs can be unpredictable, workers sometimes encounter critical incidents (CIs) — tasks or situations which are sufficiently disturbing to challenge workers’ usual coping mechanisms. In the context of ambulance services we find that encountering a CI negatively affects subsequent operational performance. The effect is strong for patient-pickup (a complex, non-standardized task) and weaker for patient handover to the hospital (a more standardized task). 4 - Timeliness and Compliance with Standard Operating Procedures Reidar Hagtvedt, University of Alberta School of Business, 2-43 Business Bldg, Edmonton, AB, T6G 2R6, Canada, Kenneth L. Schultz, Trish Reay, Sarah Forgie Compliance with standard operating procedures when looking ahead, or in the very short term, is notoriously difficult to measure. In this paper, we examine two previously ignored aspects of compliance with hand-hygiene regulations in a hospital setting. First, we allow teleological cues to prompt hand-hygiene by re- coding the time of the cue forward. Second, we examine the time immediately after a disruption, to see if a reduction in hand-hygiene is measurable. We use data from a tertiary Canadian teaching hospital. Joint Session HAS/Practice Curated: Dynamic Models of Patient Health Sponsored: Health Applications Sponsored Session Chair: Maria Esther Mayorga, North Carolina State University, Raleigh, NC, 27695, United States 1 - Dynamic Patient Routing with Nurse Workload Considerations Siddhartha Nambiar, North Carolina State University, Raleigh, NC, 27606, United States, Maria Esther Mayorga, Muge Capan We develop a multi-class, multi-server queueing model to route incoming patients to different server pools in an Emergency Department setting, to minimize their sojourn time and to manage the nurses’ workload. There is a lack of consensus in analytics and decision science regarding how to quantify workload in healthcare and how to use it in an analytic framework. We first review literature to determine how to define workload, and to identify components, available in electronic health records data, that allow us to measure workload. Following this, we develop both analytic and simulation models to get optimal routing policies and compare them to other workable, realistic policies. n TA57 West Bldg 101B

Leon Valdes, University of Pittsburgh, Joseph M. Katz Graduate School of Business, 119B Mervis Hall, Pittsburgh, PA, 15260, United States, Tim Kraft, Yanchong Zheng We conduct an incentivized lab experiment to investigate the role of supply chain visibility on consumer trust of social responsibility (SR) communications. Specifically, we design a three-player game with the roles of Consumer, Seller, and Worker. The Worker’s payment is private information, but the Seller observes a signal about this payment. The Seller can invest to improve this signal’s quality, which captures the level of supply chain visibility. Finally, the Seller communicates with the Consumer about the observed signal. With this game, we study whether the level of visibility affects the Consumer’s trust in the Seller’s communication.

n TA55 North Bldg 232C Joint Session Sports/Practice Curated: Sports Analytics IV

Sponsored: SpORts Sponsored Session Chair: Stephen Hill, University of North Carolina-Wilmington, 601 South College Road, Wilmington, NC, 28403-5611, United States 1 - Optimizing Pitcher Rotations Cody O’Brien, Slippery Rock University, Slippery Rock, PA, 16057, United States, Bradely Schweitzer, Justin Long Replacing a starting pitcher with a relief pitcher is rarely backed by analytics. By taking an analytical approach, a manager may be able to detect when a pitcher’s performance is starting to decline before it has a negative impact. Historical data is used to calculate multiple indicators of pitcher performance to create a baseline of how each individual pitcher should perform. During a game, current indicators will be calculated and compared to the baseline statistics to understand how the pitcher is performing. When his performance declines passed a set threshold, a relief pitcher will be called in. A fully-functioning GUI is implemented to make the model more accessible. 2 - Simulating Major League Baseball Games Brad Schweitzer, Slippery Rock University, Slippery Rock, PA, United States, Justin Long The game of baseball can be explained as a Markov Process, as each state is independent of all states except the one immediately prior. Probabilities are calculated from historical data on every state transition from 2011 to 2016 for Major League Baseball games and are then grouped to account for home-field advantage, offensive player ability, and pitcher performance. Using the probabilities, transition matrices are developed and then used to simulate a game play-by-play. For a specific game, the results give the probability of a win as well as expected runs for each team. 3 - Examination of Gambling Lines in College Football Mikhail Gordon, Doctoral Candidate, University of Pittsburgh,Pittsburgh, PA, 15224, United States An investigation of college football team characteristics in the context of gambling, using historical line information. An emphasis is placed on games with large spreads, defined as a spread greater than 15 points. 4 - Social Network Analysis in Field Sports: Irish (Gaelic) Football Vincent Hargaden, Associate Professor, University College Dublin, School of Mechanical & Materials Engineering, Engineering & Materials Science Centre, Belfield, Dublin 4, Ireland Irish (Gaelic) football is a 15-a-side field sport played through a combination of hand and foot passing. Each year from May through September, the elite teams complete in conference and play-off games to determine the title winner. Similar to many team sports (basketball, soccer), it is of interest to analyse the players’ cooperation from a network perspective. Using passing data between players from games involving the three-in-a-row title winning team (2015-2017), we take a social network analysis approach to identify interactions and team style of play.

n TA56 West Bldg 101A Joint Session HAS/Practice Curated: Behavioral Drivers for Decision Making and Productivity: Empirical Evidence Sponsored: Health Applications Sponsored Session Chair: Jónas Oddur Jónasson

263