Informs Annual Meeting 2017

SA16

INFORMS Houston – 2017

SA16

SA17

332F Inventory Management Contributed Session Chair: Sidika Tunc, University College London, London, United Kingdom, sidika.tunc.16@ucl.ac.uk 1 - The Need for Speed: Effects of Uncertainty Reduction in Patenting Mike Horia Teodorescu, Doctoral Candidate, Harvard Business School, Soldiers Field Road, Wyss House, Boston, MA, 02163, United States, miketeod@hbs.edu How does patent acceleration affect startup growth? Using a novel dataset obtained through a year-long government study, I look at the strategic implications of patent acceleration for startups and venture capital investors. The paper builds on the markets for technology literature. The empirical approach is a differences-in-differences regression to determine the effects of a previously unstudied executive policy change on the entrepreneurship ecosystem. A novel technique based on natural language processing is employed for firm matching. 2 - Multi-dimensional Proximity and Knowledge Networks of Product and Process Innovation Owais Anwar Golra, Doctoral Researcher, University of Edinburgh, 29 Buccleuch Place, Edinburgh, EH89JS, United Kingdom, S1256783@sms.ed.ac.uk Informal networks among manufacturing firms play important role in the transfer of knowledge in Industrial clusters. Proximity facilitates networking process, however our understanding is limited about the relationship between proximity dimensions and innovation networks. Particularly, it is unclear, how proximity dimensions shape the knowledge network of product and process innovations? We studied these networks in a textile cluster in Pakistan, and find significant influence of four proximity dimensions, excluding institutional proximity, on the network formation. Notably, the impact of proximity on process innovation network is stronger than the impact on product network. 3 - Strategic Information Technology Alignment, External Knowledge & Innovation Performance: An Empirical Study Mahmoud Abdullah, PhD Student, Yuan Ze, Zong Li, Taiwan, elakhdary@gmail.com This study draws on the strategic IT alignment perspective & the knowledge-based view (KBV) to explain the impacts of the alignment between IT strategy and external search activities on firm innovation performance. Since the focus of our study comprise both IT and business strategy. This study is using the key informant approach to collect data from senior IT executives and business strategic planners in the Taiwanese manufacturing firms. 4 - Technical Entrepreneurship in the Large Firm Eliezer Geisler, Illinois Institute of Technology, Chicago, IL, 60661, United States, geisler@stuart.iit.edu, Giuseppe Turchetti This paper explores the technical entrepreneurship (or “Intrapreneurship) in the large firm. We review the relevant literature with a focus on: what are the variables which impact the activities and the performance of entrepreneurs in a large firm? What are the barriers and facilitators to technical entrepreneurs as they pursue their endeavors in the large firm? These and similar questions are explored from a theoretical perspective by describing the state of the art in this topic. We explore the streams in the literature which explore these questions and Sidika Tunc, University College London, Level 38 One Canada Square, Canary Wharf, London, E14 5AA, United Kingdom, sidika.tunc.16@ucl.ac.uk, C.Gizem Korpeoglu, Ersin Korpeoglu In an innovation contest, an organizer decides on the contest duration and award to elicit solutions to an innovation-related problem from a number of independent agents. Each agent exerts effort to develop her solution by incurring a cost. This cost of effort decreases with the contest duration, and the quality of her solution depends on the agent’s uncertainty and productivity. We show that the agent’s effort increases with the contest duration, but the organizer limits the contest duration to guarantee agents’ participation in the contest. We also show that the optimal contest duration and award increase with the agent’s uncertainty and decrease with the agent’s productivity. we extract relevant propositions from this literature. 5 - Optimal Duration of Innovation Contests

340A Markov Decision Processes and Applications Sponsored: Applied Probability Sponsored Session Chair: Daniel F. Silva, Auburn University, Auburn, AL, 36849, United States, silva@auburn.edu 1 - Optimizing the Interaction between Residents and Attending Physicians Hayriye Ayhan, Georgia Institute of Technology, School of Industrial & Systems Eng, 755 Ferst Dr NW, Atlanta, GA, 30332- 0205, United States, hayhan@isye.gatech.edu, Sigrun Andradottir We analyze how attending physicians should allocate their time between the residents they supervise and their own responsibilities. Under the assumptions that the work of the attending physician can be preempted, and that a holding cost is incurred when residents and patients wait for a conference with the attending physician, we show that there are only two policies that could maximize the long-run average reward. Namely, it is optimal for the attending physician to start having consultations with the residents either when the residents can no longer examine new patients or as soon as there is a patient ready for conference. We then characterize when each of these policies is profitable. 2 - Control Policies for Queueing Systems with Time Windows Tugce Isik, Clemson University, 112 Summey St. Unit 2, Clemson, SC, 29631, United States, tisik@clemson.edu, Bahar Cavdar We focus on a queueing system where each job has a preferred time window for service. Every job in the system must be served by the end of the time window, and jobs may need to be outsourced if the capacity is not enough. Both early service and outsourcing are penalized. We show that the cost-optimal policy has a monotone structure and is of threshold-type when there is a single class of jobs. Based on these results, we develop a policy improvement heuristic and evaluate its performance. 3 - Discard or Install: Upgrading in the Presence of Spare Parts Stock Joni Driessen, j.p.c.driessen@tue.nl, Alan Scheller-Wolf, Joachim Jacob Arts, Geert-Jan Van Houtum We study how a company should manage an upgrade of a capital-intensive product when they have a number of old items (operating and spare parts) to be replaced by new items. The company must trade off the savings from exploiting the current old inventory versus the superior performance of the new items. We model this problem as a Markov Decision Process, and we decide in each period: (i) to replace any failed item by a new or old item, or not to replace it at all; (ii) and how many old item spare parts to salvage. We prove properties of the optimal policy, and generate insights into a replacement policy in general. 4 - Near-optimal Control of Complex Authentication Systems Oguz Toragay, Auburn University, 257 S.Gay Street, Dexter Arms A208, Auburn, AL, 36830, United States, ozt0008@auburn.edu, Daniel F.Silva We consider an authentication system, with several methods available, where requests arrive from users of several classes. The decision maker must dynamically assign each incoming request to a method, with the objective of maximizing security and minimizing latency and cost. We model the problem as a multi-class, parallel queueing system and solve it using MDP. The structure of optimal policies in systems with only one finite-capacity method is known. We use this structure to propose a new heuristic approach to construct near-optimal policies. Numerical experimentation shows that our approach offers near-optimal performance for a wide range of parameters.

8