2016 INFORMS Annual Meeting Program

SC67

INFORMS Nashville – 2016

SC67 Mockingbird 3- Omni Decision Analysis Approaches and Predictive Modeling to Managing Uncertainty in Manufacturing and Service Systems Design & Operations Sponsored: Quality, Statistics and Reliability Sponsored Session Chair: Zhenyu Kong, Virginia Tech, 1145 Perry Street, Blacksburg, VA, 24060, United States, zkong@vt.edu 1 - Self-organizing Network For Variable Clustering And Predictive Modeling Hui Yang, Penn State University, huy25@engr.psu.edu Rapid advancement of sensing and information technology brings the big data, which presents a gold mine of the 21st century to advance knowledge discovery. However, big data also brings significant challenges for data-driven decision making. In particular, it is common that a large number of variables (or predictors, features) underlie the big data. Complex interdependence structures among variables challenge the traditional framework of predictive modeling. This paper presents a new methodology of self-organizing network for variable clustering and predictive modeling. 2 - Forecasting Of Weather-driven Damage In A Distribution System Of Electric Power Zhiguo Li, IBM, hardthinking@gmail.com Electric utilities spend a large amount of resources and budget on managing unplanned outages, the majority of which are driven by weather. A major ongoing effort is to improve their emergency preparedness process, in order to reduce outage time, reduce repair costs, and improve customer satisfaction. This paper proposes a method for forecasting the number of damages of different types that will result from a weather event in a power distribution system. The proposed method overcomes practical issues with sparsity of historical damage and weather records, and its performance is evaluated on real utility data. This work is the core of an approach called Outage Prediction and Response Optimization. 3 - Prognostics Of Surgical Site Infections Using Dynamic Health Data Yan Jin, University of Washington - Seattle, yanjin@uw.edu, Shuai Huang Surgical Site Infection (SSI) is a national priority in healthcare research. To achieve better SSI risk prediction models, there have been emerging mobile health (mHealth) apps that can closely monitor the patients and generate continuous measurements of many wound-related variables and other evolving clinical variables. Since existing predictive models of SSI have quite limited capacity to utilize the evolving clinical data, we develop the corresponding solution to improve these mHealth tools with decision-making capabilities for SSI prediction. We derive efficient algorithms and demonstrate the advantage of our new predictive model on a real-world dataset. 4 - Spatiotemporal Model With Dirichlet Process Mixing For Nonnormal And Nonstationary Data Jia Liu, Virginia Tech, jliu@vt.edu In real-life, sensor data often violate assumptions of normality and stationarity required by many prevalent statistical methods. In order to acquire accurate prediction and interpolation by sensor data, a nonparametric spatiotemporal model is proposed, which takes non-normality and non-stationarity of data into account. In this model, spatial correlation is captured by Dirichlet process mixture model using particle filter. Moreover, temporal correlation is incorporated into this model by using recurrent Dirichlet process. This model can be used in various fields with data exhibiting non-normality and non-stationarity to achieve accurate interpolation and prediction. SC68 Mockingbird 4- Omni Panel Discussion: Funding Opportunities Sponsored: Quality, Statistics and Reliability Sponsored Session Moderator: Abhishek K Shrivastava, Florida State University, Tallahassee, FL, United States, ashrivastava@fsu.edu Co-Chair: Hui Wang, FSU, TBD, TBD, FL, 00000, United States, hwang10@fsu.edu 1 - Panel Discussion On Funding Opportunities Abhishek Shrivastava, Florida State University, FAMU-FSU College of Engineering, Tallahassee, FL, 32310, United States, ashrivastava@fsu.edu

In this panel, program officers from NSF will discuss funding opportunities in their programs. The panelists are Dr. Joanne Culbertson, Dr. David Mendonca, Dr. Jon Leland and Dr. Alexandra Medina-Borja 2 - Panelist Alexandra Medina-Borja, US National Science Foundation, 2507 Fowler St, Falls Church, VA, 22046, United States, alexandra.medinaborja@upr.edu 3 - Panelist David Mendonca, NSF, Arlington, VA, 22230, United States, mendonca@nsf.gov 4 - Panelist Joanne Culbertson, National Science Foundation, 4201 Wilson Boulevard, Arlington, VA, 22230, United States, jculbert@nsf.gov SC70 Acoustic- Omni Transportation, Freight III Contributed Session Chair: Carlos Alberto Gonzalez-Calderon, Rensselaer Polytechnic Institute, 4 25th St, Apt 5, Troy, NY, 12180, United States, gonzac8@rpi.edu 1 - A Multi-commodity Intermodal Traffic Assignment Between Rail And Truck Lokesh Kumar Kalahasthi, Rensselaer Polytechnic Institute, 22 College Ave, Troy, NY, 12180, United States, kalahl@rpi.edu Trilce Marie Encarnacion, Jose Holguin-Veras, John E Mitchell The goal of the paper is to obtain an optimization model that gives a freight traffic assignment on a combined network of road and rail; that could be used to assess the freight modal split including vehicle types and intermodal transfers. Authors of this paper have conducted In-Depth-Interviews (IDI) with shippers, carriers and receivers regarding the factors influencing their mode choice. The challenge is to incorporate the findings from these IDIs into a mathematical model. Major findings include commodity type, backhaul, shipment limit, transfer time, reliability in transit time restrictions. The model also incorporates the variation in the rail pricing based on origin and destination. 2 - Reliable Routing Of Multicommodity Road-rail Intermodal Freight Under Uncertainty M. Majbah Uddin, University of South Carolina, 300 Main Street, Civil and Environmental Engineering, Columbia, SC, 29208, United States, muddin@cec.sc.edu, Nathan Huynh A reliable routing model for multicommodity shipments on a road-rail intermodal freight transport network, where network elements are subject to uncertainty, is proposed. A stochastic mixed integer program is formulated which minimizes not only operational costs but also penalty cost associated with unsatisfied demand. This study provides a novel distribution-free approach to ensure probabilistic guarantees on the resulting routing plan. Case study on a small network reveals the key characteristics of the proposed model. 3 - Shipment Consolidation And Dispatching With Cross-docks Sinem Tokcaer, Izmir University of Economics, Fevzi Cakmak Mh, Sakarya Cd No:156, Izmir, 35330, Turkey, sinem.tokcaer@ieu.edu.tr, Ahmet Camci, Ozgur Ozpeynirci Freight forwarders dealing with long haul transportation establish their own consolidation systems in order to reduce costs by economies of scale and efficient use of owned or rented vehicles. Such consolidation systems usually include cross-docking terminals to provide additional services and reduce the travelling time of the vehicles. We propose a shipment consolidation and dispatching problem with cross-docks, and develop a mathematical programming model. The model suggests the consolidation and transportation plans. We propose lower and upper bounds, develop a Variable Neighborhood Search algorithm, and test the performances of develop methods on randomly generated instances. 4 - Freight Trip Generation (FTG), Freight Generation FG) And Service Trip Attraction (STA) In New York City (NYC) And Capital Region

Carlos Alberto Gonzalez-Calderon, Rensselaer Polytechnic Institute, 4 25th St, Apt. 5, Troy, NY, 12180, United States, gonzac8@rpi.edu, Jose Holguin-Veras, Shama Campbell, Lokesh Kumar Kalahasthi

This paper presents a thorough analyses and econometric models explaining the Freight Trip Generation (FTG), Freight Generation (FG) and the Service Trips Attraction (STA) in the New York City and Capital Region. The team conducted a detailed survey including the number of deliveries (received), shipments (sent), type of cargo, weight of shipment, industry sector, truck type, who transports the cargo (vendor or receiver). This study serves as a tool for transportation planners in understanding the freight patterns in urban areas.

92