Informs Annual Meeting 2017

TC56

INFORMS Houston – 2017

4 - Prediction Based Electricity Forward Pricing Model in Japan Electric Power Exchange Yuji Yamada, Professor, University of Tsukuba, 3-29-1 Otsuka Bunkyo-ku, Tokyo, 112-0012, Japan, yuji@gssm.otsuka.tsukuba.ac.jp In this paper, we demonstrate the prediction based electricity forward pricing model in JEPX, where we decompose the log-price process into the trend function and its residual based on the generalized additive model (GAM). We then apply the autoregressive moving average model (ARMA) for the residuals to estimate the conditional expectation of JEPX spot price, and derive the forward price with arbitrarily delivery period based on the Esscher transformation. 5 - Accelerating Scenario Decomposition Schemes using Optimal Scenario Grouping Deepak Rajan, Lawrence Livermore National Laboratory, Box 808, L-495, Livermore, CA, 94551, United States, rajan3@llnl.gov, Kevin C. Ryan, Shabbir Ahmed, Santanu Subhas Dey We present a scenario grouping scheme for accelerating scenario decomposition algorithms to solve stochastic integer programs. The scenario grouping scheme is quite general and applicable to any scenario decomposition scheme, which can be implemented in a distributed framework. We illustrate the effectiveness of the proposed scheme on hard-to-solve stochastic unit commitment instances. 362A Environmental Aspects of Transportation Systems Sponsored: Transportation Science & Logistics Sponsored Session Chair: Gisele Mendy Bilek, University of Pau and the Adour countries, 8 Allees des Platanes, Bayonne, 64100, France, gisele.mendy-bilek@univ-pau.fr 1 - A Soft System Based Framework for Sustainable Freight Transportation Ravi Shankar, Professor, Indian Institute of Technology Delhi, New Delhi, India, ravi1@dms.iitd.ernet.in, Vijayta Fulzele The major challenge for any freight transporter is to effectively develop proper fit between organisation goals and sustainability initiatives. Soft System Methodology has been adopted to design this framework. This fundamental and comprehensive framework helps in better adoption of sustainability in freight operations. 2 - Energy-efficient Peer-to-peer Ridesharing Optimization Yanshuo Sun, Research Scientist, Univ of Maryland, 1173 Glenn Martin Hall, College Park, MD, 20742, United States, gexingba@gmail.com, Zhi-Long Chen, Lei Zhang The Peer-to-peer (P2P) dynamic ridesharing has been promoted as a promising mobility option to reduce the system-wide cost by consolidating similar trips into one. As the problem shares some features with pickup and delivery problems, even static problem instances are computationally very expensive and exact solutions can be found only for small or medium-sized problems. A highly efficient set packing type formulation is thus proposed to solve relatively large problems (e.g., 500 vehicles and 1000 riders) optimally in almost real-time. Real- world tests in the Washington DC-Baltimore region are performed to explore the potential of dynamic P2P ridesharing for system-wide energy use reduction. 3 - Modeling Enablers of Dedicated Freight Corridor for Sustainable Freight Transportation Ravi Shankar, Professor, Indian Institute of Technology-Delhi, India is witnessing an unprecedented decline in modal share of rail freight transportation. Dedicated freight corridors (DFCs) are being implemented to reduce carbon emissions and decongest freight traffic. This study attempts to identify enablers of DFC to achieve sustainability in freight transportation. Exploratory factor analysis is employed to identify hidden constructs and total interpretive structural modeling has been used to explicate the inter-relationship among these enablers. 4 - The Impact of a Mandatory Program for Measuring and Reporting Co2 Emissions on Environmental Practices: the Case Study of Road Transportation in France Gisele Mendy Bilek, Associate Professor, IAE Business School, University of Pau and the Adour countries, 8 Allees des Platanes, Bayonne, 64100, France, gisele.mendy-bilek@univ-pau.fr, Nathalie Fabbe-Costes, Arnaud Bilek, Elodie Kacoui Maurin Since 2013, a mandatory program to inform CO2 emissions of transport services on French territory has been implemented. This research analyzes the motivations and constraints associated with the application of the device and the New Delhi, 110016, India, r.s.research@gmail.com, Devendra Kumar Pathak, Devendra Choudhary TC54

impact on environmental practices. On the basis of a national survey, the researchers propose a typology of companies applying the decree, identifies factors of influence and possible avenues for improvement. A model based on systems dynamics theory makes it possible to simulate several scenarios with regard to the evolution of the device.

TC56

362C Health Care, Other Contributed Session Chair: Bhavin Shah, Indian Institute of Management, Indore, Indore - Madhya Pradesh, India, bhavinj@iimidr.ac.in 1 - Assessing In-stent-restenosis Risks of Patients After Coronary Stent Placement Halenur Sahin, Middle East Techical University, Ankara, Turkey, halenur.sahin@metu.edu.tr, Serhan Duran, Ertan Yakici Aging and some lifestyle habits cause plaque accumulation in cardiac veins and narrows the arteries. Stents are used in balloon angioplasty to keep the vessel open but the stented vessel has a risk of re-narrowing due to recovery response of the stented segment. It is called in-stent-restenosis. Our objective is to develop a model to classify patients due to their risks of restenosis by using OR and statistical tools. With this model, it is intended to bring patient specific follow-up proposals to protect them from an emergency case, more medication requirement, and risk of losing effectiveness of treatment. In this context, mathematical models and effective solution methodologies will be presented. 2 - A Discussion of the Capital and Human Resources of the Aging Care System in Taiwan Ching Ying Huang, Assistant Professor, National Cheng Kung University, Tainan, Taiwan, isastat@mail.ncku.edu.tw Aging in place let the elderly stay in the familiar environment not feel be abandoned. How the local community provides the care assistance to the aging population and increases their independence is an issue. In Taiwan, the care organizations are classified into three levels, A, B, and C that helps the daily life of the local elderly. Under limited resources, how to run the care system in an efficient way is difficult. We mimic the system of an A-class care organization with ARENA. Four types of the aging services, energy station, day care center, dementia day care, and housing are discussed. The research attempts to answer how much the capitals and the resources should input in an A-class care organization. 3 - Competition Between For-profit and Not-for-Profit Hospitals Steven Mark Shugan, University of Florida, 2030 NW. 24th For-profit and not-for-profit hospitals differ on many factors including their objectives. Using a two-stage game theoretic optimization, we model competition between these hospitals. With an extensive database of US hospitals, we show, consistent with model predictions, nonprofits are more profitable than for-profits, nonprofits have higher equilibrium prices than for-profits, competition increases the differences between these hospitals, and nonprofit make greater investments in premium medical services. We provide additional empirical findings and implications for hospital and public policy. 4 - Measuring Complexity in Healthcare Systems Farzad Zaerpour, University of Calgary, 407 10 Brentwood CMNW, Calgary, AB, T2L2L6, Canada, farzad.zaerpour@haskayne.ucalgary.ca, Diane P.Bischak, Mozart Batista de Castro Menezes In this work we study a complexity measure and its relationship with performance measures within the healthcare system. We borrow the measure from manufacturing systems to assess how well a healthcare system can manage the complexity induced by the patient group mix. We illustrate the use of the measure in an emergency department (ED) setting. An emergency department is a complex system with random arrivals of patients who may have different classifications based on personal history and health condition. We present findings on the impact of complexity levels on ED quality performance. 5 - Homophily Effect of Doctor Behavior Over Time in Online Healthcare Communities: Evidence from China Panpan Wang, Shanghai Jiao Tong University, Shanghai, China, wpp_cq@qq.com, Jifeng Luo This research empirically investigate the patterns of physicians’ adoption and usage behaviors in online healthcare community over time across locations. Based on existing literature, we propose that doctors’ adoption behavior is more influenced by their peers similar both in geography and demography. We collected data on historical online activities of 26499 physicians in 31cities of China from a large Chinese healthcare website. We use logit model to study doctors’ adoption behavior. Homophily effect(i.e. demographic similarity and geographical proximity) of doctor’s behavior over time will be reported. Avenue, Gainesville, FL, 32605, United States, steven.shugan@warrington.ufl.edu, Jihwan Moon

355