2015 Informs Annual Meeting

SD76

INFORMS Philadelphia – 2015

SD74 74-Room 204A, CC Panel Discussion: Funding Opportunities Sponsor: Quality, Statistics and Reliability Sponsored Session

3 - Time Series Forecasting for Nonlinear and Non-stationary Processes: A Review and Comparative Study Changqing Cheng, USF, 4202 E. Fowler Avenue, Tampa, FL, United States of America, cheng.changqing@gmail.com, Hui Yang, Zhenyu Kong, Satish Bukkapatnam Forecasting of time series data is critical for the monitoring of complex systems. This article presents a review of nonlinear and non-stationary time series forecasting models and a comparison of their performances in certain real-world applications. Conventional approaches do not adequately capture the system evolution, from the standpoint of forecasting accuracy, computational effort, and sensitivity to quantity and quality of a priori information in these applications. SD73 73-Room 203B, CC Multicriteria and Multiobjective Models in Risk, Reliability and Maintenance Sponsor: Quality, Statistics and Reliability Sponsored Session Chair: Adiel De Almeida Filho, Assistant Professor, Universidade Federal de Pernambuco, Caixa Postal 7471, Recife, PE, 50630-971, Brazil, adieltaf@googlemail.com Co-Chair: Adiel T De Almeida, Professor, Universidade Federal de Pernambuco, Caixa Postal 7462, Recife, PE, 50630-971, Brazil, almeidaatd@gmail.com 1 - Multicriteria and Multiobjective Models for Risk, Reliability and Maintenance Decision Analysis Adiel T De Almeida, Professor, Universidade Federal de Pernambuco, Caixa Postal 7462, Recife, PE, 50630-971, Brazil, almeidaatd@gmail.com, Adiel De Almeida Filho, Marcelo H Alencar, Rodrigo J P Ferreira Many decisions on Risk, Reliability and Maintenance (RRM) problems involve multiple objectives and multicriteria methods (MCDM). RRM decision problems may affect strategic results of organizations, human life (e.g. safety) and the environment, in order to comply with modern society demands. A framework for structuring MCDM in the RRM context is presented, based on the reference Multicriteria and Multiobjective Models for Risk, Reliability and Maintenance Decision Analysis. 2 - Multicriteria and Multiobjective Models on Maintenance Outsourcing Adiel De Almeida Filho, Assistant Professor, Universidade Federal de Pernambuco, Caixa Postal 7471, Recife, PE, 50630-971, Brazil, adieltaf@googlemail.com, Thalles V Garcez, Adiel T De Almeida This work presents key aspects of MCDM approaches for decisions on maintenance outsourcing, which including contract selection and supplier selection. Given the MCDM nature of maintenance outsourcing, models to address this problem include maintainability, dependability, quality of repair besides cost, which are detailed in the reference Multicriteria and Multiobjective Models for Risk, Reliability and Maintenance Decision Analysis. 3 - Multidimensional Risk Evaluation Based on Multicriteria and Multiobjective Models Marcelo H Alencar, Universidade Federal de Pernambuco, Av. Professor Morais Rego, 1235., Recife, PE, 50670-901, Brazil, marcelohazin@gmail.com, Adiel T De Almeida, Rodrigo J P Ferreira, Adiel De Almeida Filho, Thalles V Garcez When considering a multidimensional risk evaluation model a broader view is enabled by considering financial, environmental and safety aspects simultaneously. This work presents MCDM approaches in two different contexts: natural gas pipelines and underground electricity distribution system. These models are part of the reference Multicriteria and Multiobjective Models for Risk, Reliability and Maintenance Decision Analysis. 4 - Multicriteria and Multiobjective Models on Preventive and Condition-based Maintenance Rodrigo J P Ferreira, Assistant Professor, Universidade Federal de Pernambuco, Av. Professor Morais Rego, 1235., Recife, PE, 50670-901, Brazil, rodjpf@gmail.com, Cristiano A,V, Cavalcante, Adiel T De Almeida, Adiel De Almeida Filho Classical optimization models consider only the cost optimization for defining the best time interval for preventive maintenance. This work presents a framework to support the selection of time interval for equipment preventive maintenance involving reliability and availability aspects besides cost. A complete framework for such decisions is presented in the reference Multicriteria and Multiobjective Models for Risk, Reliability and Maintenance Decision Analysis.

Chair: Abhishek Shrivastava, Assistant Professor, FAMU-FSU College of Engineering, Dept of Industrial & Manufacturing Eng, Tallahassee, FL, 32310, United States of America, ashrivastava@fsu.edu 1 - Panel Discussion: Funding Opportunities Moderator:Abhishek Shrivastava, Assistant Professor, FAMU-FSU College of Engineering, Dept of Industrial & Manufacturing Eng, Tallahassee, FL, 32310, United States of America, ashrivastava@fsu.edu, Panelists: Fariba Fahroo, Sylvia Spengler, Diwakar Gupta In this panel, program officers from NSF and DARPA will discuss funding opportunities in their programs. The panelists are Dr. Fariba Fahroo, Dr. Sylvia Spengler and Dr. Diwakar Gupta. Software Demonstration Cluster: Software Demonstrations Invited Session 1 - IBM -Optimization Group - IBM Decision Optimization for Python IBM Optimization Group In this tutorial, you will learn how to use the new CPLEX Modeling API for Python with standard development tools. You will learn how to install this open source library, configure your environment to use it and write optimization models in few minutes. At the end of the session, you will be able to write your own optimization model and solve them with the free IBM Decision Optimization offers (either with the DOcloud trial or with the Community Edition of CPLEX Optimization Studio). The hands-on part of the tutorial requires a 64-bit machine (Windows, Linux, or Max) and access to wifi, but all participants will be able to follow along. Follow us on twitter @IBMoptimization. and Optimization Sponsor: Simulation Sponsored Session Chair: Chun-hung Chen, George Mason University, 4400 University Drive, MS 4A6, SEOR Dept, GMU, Fairfax, VA, 22030, United States of America, cchen9@gmu.edu 1 - Slice Sampling Approaches to Stochastic Optimization John Birge, Professor, University of Chicago Booth School of Business, 5807 S Woodlawn Ave, Chicago, IL, 60637, United States of America, john.birge@chicagobooth.edu Crude Monte Carlo approaches are often used to approximate stochastic optimization problems, but such approaches suffer from both high variance and bias. This presentation will describe the slice sampling approach, which can significantly reduce variance, and mode identification, which can eliminated bias. 2 - Simulation Analytics Barry Nelson, Walter P. Murphy Professor, Northwestern University, Dept. of IEMS, 2145 Sheridan Road, C210, Evanston, IL, 60208, United States of America, nelsonb@northwestern.edu, Yujing Lin The influence of queueing theory on stochastic simulation is substantial, including simulation’s emphasis on long-run averages of predetermined performance measures. As a result, the analysis of simulation data and real-world “big data” are entirely different. In this talk we explore how a data analytics perspective could significantly change the use of simulation. SD75 75-Room 204B, CC SD76 76-Room 204C, CC Emerging Development in Simulation

143