2016 INFORMS Annual Meeting Program

WB33

INFORMS Nashville – 2016

2 - The Analytics Of A Business Traveler’s Alert System Fletcher Lu, Associate Professor, University of Ontario Institute of Technology, Faculty of Business and IT, 2000 Simcoe Street North, Oshawa, ON, L1H 7K4, Canada, fletcher.lu@uoit.ca An insurance company is minimizing its business clients’ exposure to dangerous and risky environments by providing a dynamic mobile travel alert system. I present the analytics of this alert system, which gathers world-wide current information using big data analytics to rank alerts regarding dangerous situations and matches the alert information to clients’ relevant travel information such as location, travel time, health issues, etc. to send to the clients. Included in the system’s alerts are safety recommendations and instructions. 3 - Risk Sharing Between The Insured And The Insurer Christopher Gaffney, Drexel University, 3141 Chestnut St, LeBow College of Business, Philadelphia, PA, 19104, United States, ctg39@drexel.edu Risk reduction is a key benefit of insurance. We derive many of the properties that govern the way in which risk is shared, with a particular focus on the properties of insurer and insured variance and covariance. Additionally, we consider the conception of optimal risk sharing, and discuss how an insurance contract can constitute such an arrangement. 4 - Filtering For Risk Assessment Of Interbank Network Majeed Simaan, PhD Student, RPI, 110 8th Street, Pittsburgh Building, TROY, NY, 12180, United States, simaam@rpi.edu, Aparna Gupta, Koushik Kar We develop a framework for risk assessment in an interbank network, where banks interact with each other via short-term debt contracts. Focusing on the demand side of liquidity and omitting credit rationing, the framework identifies the interbank network structure and its degree of interconnectedness. Identification is facilitated by a statistical learning procedure that reverse engineers signals (transactions) observed in the interbank market and conducts inference about banks’ individual and network-level characteristics. The results from the simulation study undermine the value of integration, even when the network is identified in its simplest forms. 5 - Decision Making Under Uncertainty Using Monte Carlo Simulation: Case Of The Offshore Wind Industry Supply Chain David Menachof, Peter Thompson Chair in Port Logistics, The University of Hull, Logistics Institute, Hull University Business School, Kingston upon Hull, HU6 7RX, United Kingdom, d.menachof@hull.ac.uk, Negar Akbari Offshore wind energy has emerged as an emission free source of energy globally, especially across Northern European countries. The projects are now moving further from shore and into deeper waters, which in turn increases the levelized cost of energy (LCOE) and related risks. Furthermore, government support schemes are also subject to uncertainties that influence investment in such projects. A Monte Carlo Simulation model is proposed that aims to consider the risks within the project and suggests ways that risks can be incorporated in the project evaluation phase including port selection. An offshore wind project case is considered and the results are reported. Chair: Jorge Pereira, Universidad Adolfo Ibáñez, Avda. Padre Hurtado 750, office 216-C, Viña del Mar, 2530852, Chile, jorge.pereira@uai.cl 1 - Sequencing In Mixed Model Assembly Lines Mary Beth Kurz, Clemson University, 271 Freeman Hall, Clemson, SC, 29634-0920, United States, mkurz@clemson.edu, Anas Alghazi Mixed model assembly lines are used by manufacturers to satisfy customer’s demand for products’ customization while keeping the cost down. Since customization is available via different options, some orders that include labor- intensive options require more time to be assembled. In order to minimize work overload throughout the assembly line, a sequencing decision must be taken to sequence customer’s orders such that work overload is minimized. We investigate this problem which is tackled in two different ways in the literature; the car sequencing problem and the mixed model sequencing problem. 2 - A Systematic Literature Review Of Rolling Methods For Production Planning Reha Uzsoy, North Carolina State University, Dept. of Industrial & Systems Engg, 300 Daniels Hall Camps Box 7906, Raleigh, NC, 27695-7906, United States, ruzsoy@ncsu.edu, Rafael Gumaraes Wollmann, Raimundo de Sampaio The effective implementation of production plans over time requires them to be updated a new information becomes available. This motivates the extensive use of rolling methods in both industry and academia. Rolling methods for production planning can be implemented in different ways and terminology is not always WB33 203B-MCC Production and Scheduling II Contributed Session

well-defined in the literature. The objective of this paper is to characterize rolling methods, focusing on the quality of the resulting production planning decisions, using the Systematic Literature Review methodology and suggest unifying themes and directions for future research. 3 - A Simulated Annealing Approach For Scheduling Jobs On Identical Parallel Machines Pravin Tambe, RCOEM, Katol Road, Nagpur, 440013, India, tambepp@gmail.com, Makarand Kulkarni Identical parallel machine scheduling problem for minimizing the total tardiness is a very important scheduling problem, but there have been many difficulties in solving large size identical parallel machine scheduling problem with too many jobs and machines. Metaheuristic approach like Simulated Annealing(SA) has shown efficient results in solving the combinatorial optimization problem. In this paper, a hybrid approach of a SA algorithm combined with backward-forward heuristic has been proposed for solving identical machine scheduling problem for minimizing the total tardiness. A numerical example of scheduling jobs on identical high pressure die casting machines is presented. 4 - Cross-training Policies For Team Cost And Robustness Jordi Olivella, Assistant professor, Universitat Politecnica de Catalunya, Avda. Diagonal 647, Barcelona, 08028, Spain, jorge.olivella@upc.edu, David A Nembhard We assess alternative cross-training policies for work-teams considering cost, and levels of cross training. The policies are assessed with respect to their robustness to demand-mix variation and absenteeism coverage. We employ simulation to examine instances where cross training can be used to help meet a fixed demand scenario, and with instances where cross-training can help to meet demand mix variability. Current results indicate that when minimizing cross-training costs, policies related to equalizing the cross-training level among the workforce, may provide improvement in terms of robustness without additional cost. 5 - The Robust Simple Assembly Line Balancing Problem Jorge Pereira, Universidad Adolfo Ibáñez, Avda. Padre Hurtado 750, office 216-C, Viña del Mar, 2530852, Chile, jorge.pereira@uai.cl In this work we consider the simple assembly line balancing problem (SALBP) with uncertainty on the operation time of the tasks. In order to deal with the uncertainty, we put forward a robust version of the problem in which the solution can handle a limited number of disruptions. Several new lower bounds as well as adaptations of the current state-of-the-art procedures for the SALBP are proposed. These methods are tested on a computational experiment, and the results show that the method is able to solve large-sized instances within reduced running times, outperforming available procedures in the literature. 204-MCC Operations Analysis in Healthcare Sponsored: Manufacturing & Service Oper Mgmt, Healthcare Operations Sponsored Session Chair: Tinglong Dai, Johns Hopkins University, Baltimore, MD, United States, dai@jhu.edu Co-Chair: Daniel Ding, University of British Columbia, 2053 Main Mall, Vancouver, V6T 1Z2, Canada, daniel.ding@sauder.ubc.ca 1 - Allocation Of ICU Beds During Periods Of High Demand Huiyin Ouyang, University of North Carolina, ouyanghuiyin@gmail.com We formulate an MDP model for admission decisions in an ICU where patients’ health conditions changeover time according to Markovian probabilities. We find that the optimal decision can depend on the mix of patients in the ICU and provide an analytical characterization of the optimal policy. We also identify conditions under which the optimal policy is state-independent. 2 - Late-onset Neonatal Sepsis Prediction Using Supervised Learning Techniques Nadia Aly, College of William and Mary, naaly@email.wm.edu In this study, we derive features from the R-R intervals (distance between R peaks) and apply novel machine learning algorithms to predict if an infant will be diagnosed with a sepsis infection within the next twelve hours. The dataset used in this study consisted of the R-R intervals recorded by monitoring electrodes in the NICU for approximately 3,000 infants. A Support Vector Machine model, outperformed all other models with a 0.07% false alarm rate and a 91.70% classification accuracy. These encouraging results imply potential clinical applications for the NICU to implement this algorithm on real-time heart rate data to influence decisions on when to proceed with diagnostic procedures. WB34

406