Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

TB24

n TB24 North Bldg 131B Value Brought by Digital Transformation Sponsored: EBusiness Sponsored Session Chair: Lu Yan, Indiana University, Bloomington, IN, 47401, United States 1 - Users’ Continuous Participation in the Online Weight-loss Communities

n TB26 North Bldg 132A E-commerce Sponsored: Service Science Sponsored Session Chair: Irfan Kanat, Ohio University, 123 Louise Lane, Athens, OH, 45701, United States 1 - A New Lp (0

Tongxin Zhou, University of Washington, 3927 Adams Lane NE, Seattle, WA, 98105, United States, Lu Yan, Yingfei Wang, Yong Tan Many weight-loss communities are designed to provide individuals with community support and, thus, help with their weight management. However, it has been noticed that such communities suffer from high drop-out rate. Therefore, we are interested in finding what factors influence users’ continuous participation. We adopt the Continuous-time Hidden Markov model to calibrate users’ participation trend. The hidden state we modeled is users’ commitment to weight-loss. We find that self-monitoring tools are effective for users with low commitment level. Social support tools such as friend and supporter functions exert different effects on users with different commitment levels. 2 - To Mine or Not to Mine: An Analysis of Crypto-mining Profitability Vipul Aggarwal, University of Washington, Seattle, WA, 98105, United States, Yong Tan In the cryptocurrency ecosystem, miners wield an inordinate amount of power as their participation can stagnate or extend a blockchain. Miners’ participation is incentivized through block fees and rewards subject to finding the right hash value at some given difficulty level. We investigate their mining preferences after the Bitcoin fork of August 2017 which resulted in the creation of Bitcoin cash (BCH). Due to fluctuation in difficulty levels of BCH, miners move frequently between the chains. However, profitability ratios do not justify such frequent movements in the face of competition from other miners. We analyze their mining behavior using a structural model to ascertain their motivations. 3 - Adaptive and Personalized Thermal Comfort Based HVAC Control Weilong Wang, Purdue University, 403 W. State Street, West Lafayette, IN, United States, Siliang Lu, Chaochao Lin Throughout a person’s whole life, he/she spends 90% of time indoors since buildings can provide satisfactory environments for human beings. As one of the key building systems, Heating, ventilation and air-conditioning (HVAC) system plays a key role in shaping the building performances. The effective and efficient HVAC operations not only achieve energy savings but also create a more comfortable environment for occupant indoors. We use Machine Learning technology to analyze differences of thermal sensation under the same conditions between different people with GMM clustering and then implement value iteration-based controller with the prediction model as the simulated environment. 4 - The Substitution Effect of Sharing-bike on Ride-sharing Platform Juan Qin, Harbin Institute of Technology, 4138 Brooklyn Ave NE, WA 98105, Seattle, WA, 98105, United States Sharing-bike was introduced to solve the “last-mile problem” whereas ride- sharing platform has been serving as a popular means for both short-distance and long-distance journeys. Does sharing-bike solve the “last-mile problem” and have a substitution effect on ride-sharing platform? Will geographic heterogeneity affect the way sharing-bike and ride-sharing platform compete? We take a natural experiment setting and conduct an RDD model to answer these research questions. We exploit granular order data from a leading ride-sharing platform in China and combine it with geographic information to look into the heterogeneous substitution effect. n TB25 North Bldg 131C Best Student Paper Competition – IV Sponsored: Service Science Sponsored Session Chair: Aly Megahed, IBM Research-Almaden, San Jose, CA, 95123, United States 1 - Service Science Best Student Paper Competition Aly Megahed, IBM Research - Almaden, San Jose, CA, 95123, United States This session consists of finalists presentations (judged by an expert panel) to determine the Best Student Paper Award for the Service Science Cluster.

284