Informs Annual Meeting 2017

TE69

INFORMS Houston – 2017

TE66

2 - Monitoring Communications: Aiming to Identify Periods of Unusually Increased Communications between Parties of Interest Ross Sparks, CSIRO.Computational Informatics, North Ryde, NSW, Australia, Ross.Sparks@data61.csiro.au Communication monitoring of large networks with aim of detecting unusually increased communications is a challenging task. This talk will cover some of the challenges when firstly the target sub-group of the network with potentially increased communications is known and changes in size over time; and secondly when the sub-group is unknown and needs to be found. The unknown sub-group case will be carried out under two situations. Firstly when the sub-group has a dominant leader and communication sparsity in the sub-group can be exploited. Secondly when sub-group has a collaborative leader and no communication sparsity generally exists in the sub-group. 3 - Two-sided Tolerance Intervals for Members of the (Log)-location-scale Family of Distributions Yili Hong, Virginia Polytechnic Institute, 213 Hutcheson Hall, Department of Statistics, Blacksburg, VA, 24061, United States, yilihong@vt.edu In this paper, we propose methods to calculate exact factors for two-sided control- the-centre and control-both-tails tolerance intervals for the (log)-location-scale family of distributions, based on complete or Type II censored data. With Type I censored data, exact factors do not exist. For this case, we developed an algorithm to compute approximate factors. Our approaches are based on Monte Carlo simulations. We also provide algorithms for computing TIs that control the probability in both tails of a distribution. 371D Sustainable and Responsible Supply Chain Management Sponsored: Energy, Natural Res & the Environment Environment & Sustainability Sponsored Session Chair: Jose Cruz, jose.m.cruz@uconn.edu 1 - Revenue and Demand Management of Smart Cities Shivani Shukla, PhD, University of Massachusets, Amherst, MA, United States, sshukla@som.umass.edu A smart city concept encompasses enhancement of transportation to support growth initiatives, environmental review to sustain modernization, and sustainable travel. Multiple public and private entities cooperate to share information and resources for collective efforts driven to achieve sustainability. This work focuses on measuring the effectiveness of such cooperative enterprises in the realm of demand management through the intended shift from the typical travel practices to more reliable alternatives. 2 - Supply Chain Sustainability through Risk Sharing Contracts In this paper, we develop a framework that captures sustainability trough d risk- sharing contracts in supply chain networks. Risk-sharing contracts specify who needs to incur the costs when supply chain disruptions occur. We present the behavior of the various decision-makers, derive the governing equilibrium conditions, establish the finite-dimensional variational inequality formulation, and present numerical examples. 3 - Supply Chain Network Competition and Sustainable Environmental Policies Min Yu, PhD, University of Portland, Portland, OR, United States, yu@up.edu, Jose Cruz, Dong Li We first develop a competitive supply chain network model consisting of a finite number of firms competing in an oligopolistic manner. We investigate the problem with progressive emission taxes, in which the emission tax function is a piecewise-defined function of the total emission. Second, we extend the single period model to a multiperiod model to study the firm’s long-term investment strategy on the sustainable energy-efficient technologies to take advantage of tax incentives while maximizing the net present value (NPV) of the investment. TE69 Jose Cruz, University of Connecticut, School of Business, 2100 Hillside Road, Storrs, CT, 06269-1041, United States, jose.m.cruz@uconn.edu

371A Analysis of Shared Mobility Systems Sponsored: Transportation Science & Logistics Sponsored Session

Chair: Mustafa Lokhandwala, Purdue University, 241 Sheetz Street, Apt 14, West Lafayette, IN, 47906, United States, mlokhand@purdue.edu 1 - A Study on the Phenomenon of E-hailing Taxi Order Discard Based on Customer Impatience and Arrival Time Fluctuation Analysis E-hailing applications are commonly believed to improve the traveling efficiency and service quality by increasing transport supply to meet customers’ demand. However, customers may discard the E-hailing taxi and take the cruising taxi instead to minimize the waiting time, depending on fluctuation of arrival time and individual patience. Based on data-driven approaches and human factor analysis, we propose an analytical framework to study the competition of E- hailing and cruising taxis by quantitatively evaluating the extent of reduced efficiency, thus optimizing the balance between system efficiency and service quality. 2 - Modeling Users’ Memory in Shared Bike Systems Dmitry Gimon, Assistant Professor, Fort Hays State University, 600 Park Street, Hays, KS, 67601, United States, dmitrygimon@gmail.com Shared resources, such as car or bicycle rental systems, have become a norm of life in cities. Since transportation patterns vary during the day, the key problem of these systems is network balance, which affects systems reliability and users satisfaction. We propose to model user satisfaction as a function of relevant concessions that this particular user or other users have made in deciding to use different drop-off/pickup sites. We compare existing and theoretical incentive models and users’ strategies to find the one that will result in the most balanced network or the one with the most satisfied users. This study contributes to the area of shared vehicle design and operations. 3 - Ride Sharing a Case Study of Shared Taxis in New York City Mustafa Lokhandwala, Graduate Student Purdue University, Purdue University, 235 Littleton Street, Apt 15, West Lafayette, IN, 47906, United States, mlokhand@purdue.edu, Hua Cai This study uses simulation to analyze the benefit of Taxi Sharing using agent- based modeling. The focus of our study is to build an agent-based simulation model using \color{red} real world \color{black}trip data obtained from the New York Taxi and Limousine Commission, model the decision making methodologies of the agents (the taxis and the riders) in a taxi sharing scenario and run this model using the most recent trip data within the GIS setting of New York City. The end result from this experiment is to optimize the number of taxis operating on the streets of New York City and also quantify the economic and environmental benefits of the same. 371B Quality Technology and Quantitative Management Invited Session Sponsored: Quality, Statistics and Reliability Sponsored Session Chair: Jing Li, Arizona State University, Tempe, AZ, 85287-8809, United States, jing.li.8@asu.edu 1 - Multilinear Principal Component Analysis for Statistical Modeling of Cylindrical Surfaces: A Case Study zhuying Jiang, Sparkzone Research Group, Beijing, China, 2579658790@qq.com, Yu Xin, Yinghao Jia, Yinghao Jia TE67 The work focuses the problem of modeling manufactured surfaces for statistical process control. The application of Multilinear principal component analysis (MPCA) is discussed. A real case study concerning cylindrical surfaces obtained by a lathe-turning process is taken as reference. The measurements related to a specific surface are stored in a matrix addressed by 2 index variables, while the observed data set related to several surfaces is stored in a 3rd- order tensor addressed by 3 indexes. Any 2.5- dimensional surface (i.e. where scalar measurements are sampled for each item by using a fixed grid of two spatial index variables) can be modeled with the proposed MPCA-based approach. Bianca Colosimo, Politecnico di Milano, Milan, Italy, biancamaria.colosimo@polimi.it, Massimo Pacella

435