Informs Annual Meeting 2017

TE38

INFORMS Houston – 2017

2 - The Impact of Buyer Pressure on Supplier Transparency Suvrat Dhanorkar, Assistant Professor, Pennsylvania State University, 466 Business Building, University Park, PA, 16802, United States, ssd14@psu.edu, Veronica H.Villena We explore conditions under which buyer pressure may succeed or fail to drive greater supplier transparency. To test the hypotheses, we combine data from the Carbon Disclosure Project’s (CDP) supply chain program and ASSET 4 on a sample of 700+ suppliers over the 2013-2015 period. The results show that the degree of influence exerted by buyer pressure is significantly lower when suppliers have in place internal environmental structures (i.e., an environmental strategy, a dedicated board representative and managerial incentives). 4 - The Optimal Incentive Supervision Contract for Ewaste Recycling with Retailers Self Interest Behavior Ziyu Zhai, Huazhong University of Science and Technology, Wuhan, China, zhaiziyu000@126.com We adopted the principal agent theory to explain this phenomenon and study the impact of incentive supervision contract on the self-interest behavior of retailers. Pure supervision and penalty on retailers does not always avoid the self-interest behavior of retailers, even when the revenue from selling recycled E-waste to individual is low and the producers have ability to control the recycling market; the hidden number of retailers recycling is weakly correlated with the supervision level and penalty degree, furthermore only the supervision level and penalty degree reach to be a certain threshold, self-interest behavior of retailers can be effectively avoided; 351F NSF Funding Opportunities for Service System Researchers converging with Social, Behavioral and Cognitive Sciences Sponsored: Service Science Sponsored Session Chair: Alexandra Medina-Borja, NSF/ UPRM, NSF/ UPRM, Falls Church, VA, 22046, United States, amedinab@nsf.gov 1 - NSF Funding Opportunities for Service System Researchers Converging with Social, Behavioral and Cognitive Sciences Alexandra Medina-Borja, PhD, US.National Science Foundation, 4201 Wilson Blvd., Arlington, VA, 23330, United States, alexandra.medinaborja@upr.edu Dr. Alexandra Medina-Borja, Program Director in the Directorate for Engineering at the National Science Foundation will provide an overview of the agency’s emerging funding opportunities for operations researchers and data scientists in programs that draw engineering and computer science researchers in collaboration with cognitive and behavioral scientists to study human-technology partnerships in service settings. Other funding opportunities for engineering and computer science departments as well as in graduate and undergraduate education will also be discussed followed by an open discussion and Q&A. TE36

2 - The Network of Cross-disciplinary Research in Health: Neuroscience Bridging Scientific Disciplines Ran Xu, Virginia Tech, Blacksburg, VA, 24060, United States, ranxu@vt.edu, Navid Ghaffarzadegan We map the progression of cross-disciplinary research studies in biological and health sciences by examining the topics of doctoral dissertations from US institutions. Examining dissertations from the past two decades, we analyze the changes in frequency of subjects that appear in each discipline, as well as how subjects co-appear in the same dissertations. We further conduct a topic-network analysis of the dissertations and examine emerging topics and links between the topics. Results show that behavioral sciences and engineering fields are playing an increasingly important role in biological and health sciences research. There are also important new bridges emerging around neuroscience. 3 - Modeling Long-term Success and Innovation in Career of Scientists in Academia Arash Baghaei Lakeh, Virginia Tech, 1145 Perry Street, 250 Durham Hall (0118), Blacksburg, VA, 24061, United States, arashb@vt.edu, Navid Ghaffarzadegan Success and innovation in a scientist’s career depend on multiple organizational factors such as availability of funding, resource allocation, capacity development, and learning through exploration. In this project, we investigate the dynamics of innovation and success in a scientist’s career through a system dynamics model. By capturing the important feedback mechanisms in the process of science production and modeling the stochastic nature of scientific inquiry, we study the conditions under which scientists can be successful and follow innovative projects in their careers. 352C Multi-Objective Methods for Infrastructure Analysis Sponsored: Multiple Criteria Decision Making Sponsored Session Chair: Kash Barker, University of Oklahoma, Norman, OK, 73019, United States, kashbarker@ou.edu 1 - Vehicle Routing Formulation for Infrastructure Network Recovery Nazanin Morshedlou, University of Oklahoma, 1135 E.Brooks Street, Apt 2, Norman, OK, 73071, United States, nazanin.morshedlou@ou.edu This research proposes an extension of the vehicle routing problem with a comprehensive formulation that accounts for scheduling work crews to recover an infrastructure network after a disruptive event. In this problem, work crews are routed from depots to disrupted network components to maximize the cumulative flow in the network in a desired time horizon. As disrupted network components possess different characteristics (e.g., level of disruption, the rate of recovery), we develop the formulation to address (i) different levels of requirements for disrupted components and (ii) the possibility of assigning more than one work crew to a single link to change its recovery rate. 2 - Integrated Performance Modeling of Water and Transportation Infrastructures with Physical-socioeconomic Interdependencies Shima Mohebbi, University of Oklahoma, Norman, OK, United States, mohebbi@ou.edu, Mingyang Li, Qiong Zhang, Qing Lu, E. Christian Wells This study develops an integrated performance modeling and decision framework to maintain and improve the operations of water and transportation infrastructures by considering their physical-socioeconomic interdependencies. We first develop system level performance measures to evaluate the vulnerability. Based on failure scenarios heuristically simulated, we then examine their consequences on system performances and characterize the degree of interdependency. A two-stage stochastic mathematical model is further developed to investigate the cost-effective maintenance and rehabilitation decisions. A case study will be demonstrated in the City of Tampa, FL. 3 - Optimizing Interdependent Infrastructure Resilience under the Dynamics of Individuals and Organizations Most research on infrastructure resilience assumes a centralized decision maker with a single well-defined objective function. In this work, we propose a model that incorporates a multi-objective societal layer to analyze and optimize the resilience of interdependent infrastructure networks. The proposed approach enables modeling how human and organizational aspects (e.g., biases, risk preferences, multiple interests, and decentralization, among others) influence infrastructure risk and resilience management decisions on the stakeholders involved in their use and operation, and how these impact the overall performance and resilience of infrastructure systems. TE38 Andres David Gonzalez, The University of Oklahoma, Norman, OK, United States, andres.gonzalez@ou.edu, Camilo Gómez-Castro, Hiba Baroud

TE37

352B Modeling to Inform Science Policy Sponsored: Service Science Sponsored Session

Chair: Maryam Alsadat Andalib, Virginia Institute of Technology, Virginia Institute of Technology, Blacksburg, VA, 24060, United States, maryam7@vt.edu 1 - System Dynamics and Operations Research: A New Synergy for Policy Modeling Richard C. Larson, Massachusetts Institute of Technology, E40-233, Cambridge, MA, 02139, United States, rclarson@mit.edu, Navid Ghaffarzadegan We offer a methodological hypothesis that many traditional OR models can be modified with SD feedback loops driven by forces outside of the traditional OR modeler’s universe. We argue that such models, even if simple and approximate, will be powerful and effective for policy modeling. We present examples of our recent models. These cases benefit from both schools of modeling. We think the fields of SD and OR can significantly benefit from collaboration. We further discuss the first steps towards integration of queueing models and feedback loop models: endogenous queueing models.

423