Informs Annual Meeting 2017

MC64

INFORMS Houston – 2017

MC64

MC65

370E Big Data II Sponsored: Data Mining Sponsored Session Chair: Subhashish Samaddar, Georgia State University, s-samaddar@gsu.edu

370F Maritime Transportation Sponsored: TSL, Freight Transportation & Logistics Sponsored Session Chair: Harilaos Psaraftis, hnpsar@dtu.dk Co-Chair: Thalis Zis, Technical University of Denmark, Lyngby, 2800, Denmark, tzis@dtu.dk 1 - A Hub-and-spoke Approach to the Liner Shipping Network Design Rommert Dekker, Professor of Quantitative Logistics, Erasmus University-Rotterdam, Burg. Oudlaan 50, P.O. Box 1738, Rotterdam, 3000 DR, Netherlands, rdekker@ese.eur.nl, Judith Mulder We present an hub-and-spoke approach to the Liner Shipping Network Design problem with transhipment costs. We apply it to the cases from Brouer et al (2016) and show that for large instances costs can be reduced considerably. Finally we show an application to the Asia-Europe lane. 2 - Optimal Investment for a Resilient Global Port Network Ali Asadabadi, University of Maryland, 7700 Adelphi Road, Apt 34, Hyattsville, MD, 20783, United States, ali.asadabadi@gmail.com, Elise Miller-Hooks Ports are critical components of the global supply chain, supplying key connections between land- and maritime-based transport modes. They operate in cooperative, but competitive environments wherein individual port throughput is linked through an underlying transshipment network. This presentation models and analyzes protective investment strategies aimed at enhancing resilience to disruption and protecting a port’s market share. 3 - Berth Scheduling to Reduce In-port Emissions with Cold Ironing Provision Thalis Zis, Technical University of Denmark, Lyngby, 2800, Denmark, tzis@dtu.dk, Harilaos ZisPsaraftis3, Michail ZisGkolias Cold Ironing is a technology that allows vessels at berth to receive shorepower for their energy requirements at the port. This paper studies the (discrete space & dynamic arrival) berth scheduling problem with berth cold ironing capabilities. The problem is formulated as a multi-objective problem where the terminal operator seeks to simultaneously maximize the utilization of its cold ironing berths (i.e., minimize emissions) and customer satisfaction. Problem instances are solved for various penetration rates of cold ironing capabilities by vessels calling at the port, and a sensitivity analysis on the number of cold ironing-ready berths is conducted. 4 - Vessel Scheduling Problem in a Liner Shipping Route with Perishable Products Maxim A. Dulebenets, Florida A&M.University-Florida State University, 2300 Bluff Oak Way, Apt. 8408, Tallahassee, FL, 32311, United States, mdlbnets@gmail.com, Eren Erman Ozguven, Ren Moses, Thobias Sando Some of the products, transported in a containerized form by vessels, are perishable in nature. Perishable products deteriorate due to certain operational and environmental factors. This study proposes a novel non-linear mixed integer mathematical model for the vessel scheduling problem with perishable products, which explicitly captures decay of perishable products on board the vessels. Numerical experiments demonstrate that the developed mathematical model can serve as an effective planning tool for liner shipping companies and tackle important operational aspects.

Co-Chair: Milton Soto-Ferrari, miltonrene.sotoferrari@wmich.edu 1 - Methodology for Systematic Monitoring of Radiotherapy Administration to Breast Cancer Patients Milton Soto-Ferrari, Indiana State University, 30 N.7th St, Terre Haute, IN, 47809, United States, miltonrene.sotoferrari@wmich.edu The proposed methodology presents a Bayesian based quality control framework for the monitoring of patients to predict radiation receipt based on the associations among clinical and non-clinical variables. The framework was developed using information from a cancer center with a sample of 1922 patients from years 2009-2014. As result of the implementation, a total of 48 patients were flagged. The quality control framework flagged patients with a metastatic disease as condition for an aggressive treatment in the center. Physicians from the review panel denoted that radiation treatment is offered as a palliative treatment for patients in a metastatic phase to mitigate side effects and pain. 2 - Assessing the Accuracy of Influenza Incidence Characterization During an Emergency Diana Prieto, Western Michigan University, Industrial and Manufacturing Engineering, 4601 Campus Dr. (Office E-213), Kalamazoo, MI, 49008-5336, United States, diana.prieto@wmich.edu, Yuwen Gu In the United States, emergent influenza outbreaks pose challenges in public health operation. One of such challenges is in the process of collection and testing of specimens to characterize the incidence trend. It is currently unknown whether this process is capable of retrieving accurate and timely incidence trends for response planning. In this talk, we will discuss the results from a simulation based evaluation of the operational features that allow the specimen collection and testing in the state of Michigan. 3 - Forecasting for Hotel Revenue Management by using AI and Other Methods Subhashish Samaddar, Georgia State University, Managerial Science, P.O. Box 4014, Atlanta, GA, 30302-4014, United States, s-samaddar@gsu.edu, Somnath Mukhopadhyay, Satish Nargundkar Forecast accuracy is the most important factor for achieving additional revenue from hotel revenue management (HRM) models. HRM analyst can forecast product demands for all revenue buckets of the room inventory products to maximize the overall revenue. However, the literature on HRM is sparse on details of RM forecasts. This article uses archived hotel data from a major hotel chain to compare room bookings forecast accuracies of traditional forecasting methods, Bayesian Networks (BN) methods, random forest methods and AI methods such as neural networks (NN). 4 - Analysis of HealthcareService Quality from Hospital Data at the National Level Satish V. Nargundkar, Georgia State University, 2802 Fairlane Dr., Atlanta, GA, 30340, United States, snargundkar@gsu.edu, Subhashish Samaddar The secondary data from patient surveys from all hospitals in the US is obtained from Medicare and different attributes of healthcare service is analyzed at the city, state and national levels. 5 - Analyzing Disease Data for Detection: A Comparison of Methods Subhashish Samaddar, Georgia State University, Atlanta, GA, United States, subhashish.samaddar@gmail.com, Somnath Mukhopadhyay Clinical data is used for disease detection and the error rates obtained by using Neural Network, Linear Programming, and Decision Trees are compared and reported.

226