2015 Informs Annual Meeting

WB30

INFORMS Philadelphia – 2015

WB31 31-Room 408, Marriott Data Analytics for Manufacturing and Healthcare Enterprise System Sponsor: Data Mining Sponsored Session Chair: Kaibo Liu, Assitant Professor, UW-Madison, 1513 University Avenue, Madison, WI, 53706, United States of America, kliu8@wisc.edu 1 - Quantitative Imaging in Medicine Teresa Wu, Arizona State University, Tempe, AZ, United States of America, teresa.wu@asu.edu The ASU-Mayo Clinic Imaging Informatics Laboratory is a collaborative effort between the Industrial Engineering program at Arizona State University and the Department of Radiology at Mayo Clinic Arizona. Our goal is to improve patient care by analyzing and managing information in radiology images and databases. In this talk, I will briefly discuss some on-going projects on the use of quantitative imaging in the clinical context. 2 - Process Execution Monitoring and Controlled Violations Russell Barton, Senior Associate Dean, Penn State, Smeal College of Business, 210 Business Building, University Park, PA, 16802, United States of America, rrb2@psu.edu, Akhil Kumar Service processes do not lend themselves to SPC methods common in manufacturing settings. Monitoring activity timing and activity sequencing presents special opportunities for statistical characterization, and opportunities for taking corrective action. Violations in activity timing and/or sequencing are unavoidable. We show how to monitor a running process, and through constraint satisfaction find a schedule for its completion to minimize total penalty from the violations. 3 - Data Driven Approach for Modeling the Coupled Dynamics of Machine Degradation and Repair Processes Hoang Tran, Texas A&M University, College, TX, United States of America, tran@tamu.edu, Satish Bukkapatnam We proposed a data driven approach to model the coupled dynamics of recurring degradation and restoration processes that take place in manufacturing systems. Unlike previous methods, interactions between the two processes that influence downtimes and throughput rate can be explicitly considered. Theoretical and numerical analyses prove that our model can capture multimodal distribution and dynamic couplings between the time between failures and the time to repair. WB32 32-Room 409, Marriott Data Mining in Health Care Contributed Session Chair: Hamed Majidi Zolbanin Oklahoma State University, 309 S. West St, Unit 6, Stillwater, Ok, 74075, United States of America, hamed.majidi@gmail.com 1 - Predicting Inpatient Ward Demand Based on The Emergency Department Patient Characteristics Nooshin Valibeig, Northeastern University, 334 Snell engineering, Boston, 02115, United States of America, nooshin.valibeig@gmail.com, Jacqueline Griffin Bed assignment is the process of assigning patients to the targeted ward in a reasonable time or to an overflow ward when the assignment time increases. Predicting demand for each ward helps bed managers to decrease assignment time and overflow assignments which results in lower costs and better quality of care. In our study we apply data mining methods on historical data of an emergency department(ED) to predict the probability of inpatient admission and potential targeted ward for ED patients. 2 - Mining Process Patterns from Noisy Audit Logs with Application to Emr Systems He Zhang, Assistant Professor, University of South Florida, 4202 E. Fowler Avenue, CIS1040, Tampa, 33620, United States of America, hezhang@usf.edu, Sanjay Mehrotra, David Liebovitz, Carl Gunter, Bradley Malin We present a four-step framework to analyze process models with noise. The first step is to establish correlations among events and separate each trace of access logs into blocks. These blocks are then clustered into several groups and the original traces of access logs are transformed to traces consisting of high level blocks. The traces are then clustered into subgroups, each of which can be used to analyze the process. We implement our approach using data from a large academic medical center.

3 - Why Revenue Management is a Good Thing? Emmanuel Carrier, Delta, emmanuel.carrier@delta.com From its roots in the airline industry, RM has expanded to many industries such as hospitality, retail and B2B. While they are affecting a growing number of B2C and B2B transactions, RM practices have become increasingly controversial with consumers. In this paper, we look at long series of empirical data to show that RM is a win-win strategy for producers and consumers and leads to higher utilization rates. We discuss how to keep this legacy alive given the emergence of “big data” techniques. 4 - A Heuristic Approach to Predicting Customer Lifetime Values for Apartment Tenants Jian Wang, Vice President, Research & Development, The Rainmaker Group, 4550 North Point Parkway, Alpharetta, GA, 30022, United States of America, jwang@letitrain.com Estimating tenant lifetime values is important for apartment revenue management. We propose a heuristic approach to predicting renewal likelihoods and estimating tenant lifetime values. We then present empirical results based on real apartment data. WB30 30-Room 407, Marriott Information Systems I Contributed Session Chair: Xu Han, Uconn School of Business, 2100 Hillside Rd, Storrs, CT, 06268, United States of America, xu.han@business.uconn.edu 1 - The Impact of it Maturity and IS Planning Process on IS Planning Success Tomoaki Shimada, Associate Professor Of Operations Management, Kobe University, 2-1 Rokkodai-cho, Nada-ku, Kobe, 657-8501, Japan, shimada@b.kobe-u.ac.jp, Robert De Souza, James Ang, Yoshiki Matsui, Darren Ee In this study, we examine the impact of information technology (IT) maturity and information system (IS) planning process on IS planning success. Using data collected from the self-administrated questionnaire survey, we show complex relationships between IT maturity and IS planning success as well as between IS planning process and IS planning success in a non-linear regression approach. 2 - Why Do High-tech Firms Offer Perks at Work? Xuan Ye, PhD Student, New York University, 44 West 4th ST, KMEC 8-186, New York, NY, 10012, United States of America, xye@stern.nyu.edu, Prasanna Tambe We study whether and why high-tech firms rely more heavily on non-wage benefits, such as free meals, transportation subsidies, and athletic facilities (“work-related perks”). We find that employers engaged in IT innovation are more likely to offer work-related perks. Additionally, we find that high-tech firms offer work-related perks to attract and motivate IT workers who can quickly adapt to technological change. 3 - Merger and Acquisitions in it Industry Kangkang Qi, Michigan State University, 632 Bogue Street, BCC N204, East Lansing, United States of America, qikang@broad.msu.edu I study M&A in IT industry. There are three major research questions: (1) Does making M&A really help IT firms with long term profitability and/or innovation gains? (2) For each specific M&A, what are the antecedents that are related to market reaction? (3) Can we attribute the misalignment between firms’ characteristics and M&A decision to CEO overconfidence/narcissism? 4 - Nursing Home Rating System Fraud Detection Xu Han, Uconn School of Business, 2100 Hillside Rd, Storrs, CT, 06268, United States of America, xu.han@business.uconn.edu, Niam Yaraghi, Ram Gopal Potential fraud may exist in the rating procedure of CMS’s Nursing Home Compare System, leading to misuses of ratings. This study empirically examines the factors affecting the ratings. We find a significant association between ratings and profits, pointing to a financial incentive to cheat. We show that this association does not always lead to legitimate efforts, but can induce cheating. A prediction model is then developed, and 6% of the suspect nursing homes are identified as likely cheaters.

410