2015 Informs Annual Meeting

MA77

INFORMS Philadelphia – 2015

4 - Optimize the Signal Quality of Health Index via Data Fusion for Degradation Modeling and Prognostics Abdallah Chehade, UW-Madison, 1513 University Avenue,

2 - Asymptotic Validity of the Bayes-inspired Indifference Zone Procedure Saul Toscano-palmerin, 113 Lake Street, Ithaca, NY, 14850, United States of America, st684@cornell.edu, Peter Frazier This talk considers the indifference-zone (IZ) formulation of the ranking and selection problem. Conservatism leads classical IZ procedures to take too many samples in problems with many alternatives. The Bayes-inspired Indifference Zone (BIZ) procedure, proposed in Frazier (2014), is less conservative than previous procedures, but its proof of validity requires strong assumptions. In this talk, we present a new proof of asymptotic validity that relaxes these assumptions. 3 - Reconstructing Input Models via Simulation Optimization Aleksandrina Goeva, Boston University, 111 Cummington Mall, Boston, MA, 02215, United States of America, agoeva@bu.edu, Henry Lam, Bo Zhang We consider the inverse problem of calibrating the distribution of a stochastic input model from only output data, in contexts where the input-output relation is accessible via stochastic simulation. We take a nonparametric approach, and formulate this problem as a stochastic program by maximizing the entropy of the input distribution subject to moment or tail-probability matching between simulation and empirical output. We propose an iterative scheme to approximately solve the program. 4 - Quantile Based Comparison for System Selection Demet Batur, Assistant Professor, University of Nebraska-Lincoln, CBA 209, Lincoln, NE, 68588, United States of America, dbatur@unl.edu, Fred Choobineh We present a fully-sequential selection procedure for comparing simulated systems based on a quantile of interest. The quantile of interest corresponds to a specific quantile of the simulated probability distribution of a comparison metric. The procedure is designed to asymptotically guarantee the selection of the best system or the best set of equivalent systems with a pre-specified probability of correct selection. Chair: Pritha Dutta, Doctoral Student, University of Massachusetts, Amherst, Isenberg School of Management, Amherst, MA, 01003, United States of America, pdutta@umass.edu 1 - The Value of Conversion for a Refinery Firm with Both Forward and Spot Procurement Mengmiao Chen, Fudan University, Lidasan Building, School of Management, 670 Guishun Rd, Yangpu District, Shanghai, China, 12110690007@fudan.edu.cn Our work analyzes the optimal procurement strategy, processing, and production decision of a refinery firm with both forward and spot procurement (hereafter, “dual sourcing”). Also the firm is capable of converting which improves the quality. A four-stage stochastic model is applied to investigate the value of dual sourcing and conversion. We find that both dual sourcing and conversion adds value to the refinery by improving the quality, unit profit, and enlarging the feasible producing region. 2 - Do Responsible Buyers Source from Responsible Suppliers? Hsiao-Hui Lee, Assistant Professor, University of Hong Kong, KKL 814 School of Business, Hong Kong - ROC, hhlee@hku.hk I examine the role of corporate social responsibility (CSR) in supply-chain formation. I first introduce the CSR similarity between buyers and suppliers as a selection criterion and examine why good (bad) buyers source from good (bad) suppliers. However, concerns over sourcing cost moderates the CSR similarity effect for good buyers, explaining why good buyers buy from bad suppliers. Supplier transparency (CSR signals) serves as a moderator to explain why bad buyers buy from good suppliers. 3 - The Effect of Commitment Completeness on Opportunism Alex Scott, Penn State University, 463A Business Building, University Park, PA, 16802, United States of America, alexscott@psu.edu Buyers often solicit non-contractual commitments from suppliers to provide services as the need arises. The level of detail of these commitments vary because, ceteris paribus, more detailed commitments are costlier and more time- consuming to develop than less detailed commitments. In this study, we examine how commitment completeness interacts with active and passive opportunism. We explore this question using a transactional dataset in the for-hire trucking sector. MA77 77-Room 300, CC Supply Chain Management V Contributed Session

Madison, WI, 53706, United States of America, chehade@wisc.edu, Changyue Song, Kaibo Liu

In this talk, a new signal-to-noise ratio (SNR) metric that is tailored to the needs of degradation signals is proposed. By maximizing this new metric, we develop a data fusion model to construct a health index (HI) via fusion of multiple degradation-based sensor data. The case study was based on the degradation dataset of aircraft gas turbine engines, which will demonstrate the effectiveness of developed HI for better characterization and prediction of the health condition of units.

MA75 75-Room 204B, CC New Research Topics on Innovation Cluster: New Product Development Invited Session

Chair: Manuel Sosa, Associate Professor of Technology and Operations Management, INSEAD, 1 Ayer Rajah Ave., Singapore, Singapore, manuel.sosa@insead.edu 1 - Technology Readiness Levels at 40: A Study of State-of-the-art Use, Challenges, and Opportunities Alison Olechowski, MIT, School of Engineering, Cambridge, United States of America, alisono@mit.edu, Steven Eppinger, Nitin Joglekar Since their introduction by NASA in the 1970s, the Technology Readiness Levels (TRLs) have become a widely used scale for assessing technology maturity during new product and system development. We empirically investigate current TRL usage in a cross-industry study, identifying challenges related to TRL implementation and use in technology-related decision-making. Some challenges are already addressed by uncommon best practices however others are opportunities for new methods and models. 2 - Idea Generation and the Role of Feedback Joel Wooten, University of South Carolina, Columbia, SC 29208, United States of America, joel.wooten@moore.sc.edu, Karl Ulrich In many innovation settings, ideas are generated over time and managers face a decision about if and how to provide in-process feedback about the quality of submissions. We use innovation tournament field experiments to examine the effect of feedback on idea generation and show individual-level differences between no feedback, random feedback, and directed feedback. 3 - Sole Inventor vs Team of Inventors: What’s Best? Tian Chan, INSEAD, 1 Ayer Rajah Avenue, Singapore, 138676, Singapore, TianHeong.Chan@insead.edu, Jurgen Mihm, Manuel Sosa History has often attributed sole individuals as the source of innovative breakthroughs. However, recent research has shown that teams of individuals are the ones that tend to produce breakthroughs. In this work we use patent data covering both function and form to systematically analyze the source of successful innovations. Our work moves towards reconciling the sole versus team conundrum by finding evidence of situations where the sole individual shine, and of situations where they do not. MA76 76-Room 204C, CC Simulation Optimization and Ranking and Selection Chair: Demet Batur, Assistant Professor, University of Nebraska- Lincoln, CBA 209, Lincoln, NE, 68588, United States of America, dbatur@unl.edu 1 - Probability of Correct Selection: More May Not Be Better! Yijie Peng, Fudan University, School of Management, Shanghai, China, 10110690016@fudan.edu.cn, Michael Fu, Jianqiang Hu, Chun-hung Chen We present a simple counterexample where the probability of correct selection decreases with additional sampling under certain allocation schemes. We then characterize the general setting where this phenomenon may occur, which highlights the importance of an appropriate allocation scheme. Simulation experiments illustrate our findings. Sponsor: Simulation Sponsored Session

171