Informs Annual Meeting 2017

WA55

INFORMS Houston – 2017

4 - Models and Algorithms for the Traveling Salesman Problem with Time-dependent Service Times Paolo Toth, Professor, University of Bologna, DEI, Viale Risorgimento 2, Bologna, I-40136, Italy, paolo.toth@unibo.it, Valentina Cacchiani, Carlos Contreras Bolton We consider the “Traveling Salesman Problem with time-dependent Service Times” (TSPST), a variant of the classical Asymmetric TSP. In the TSPST, each customer requires a service time whose duration depends on the time at which the customer is visited. The TSPST calls for finding a Hamiltonian circuit minimizing the total duration of the circuit (i.e. the sum of the travel times and of the service times). A branch-and-cut algorithm and a metaheuristic are proposed. Computational experiments on benchmark TSPST instances are reported, showing the effectiveness of the proposed algorithms. 5 - Addressing Uncertainty in Meter Reading for Utility Companies using RFID Technology Debdatta Sinha Roy, PhD Candidate, Robert H. Smith School of Business, University of Maryland, 7699 Mowatt Lane, 3330 Van Munching Hall, College Park, MD, 20742, United States, debsroy@rhsmith.umd.edu, Bruce L. Golden, Edward Andrew Wasil Utility companies collect usage data from meters on a regular basis. It is done automatically using radio-frequency identification (RFID) technology. Each meter has a signal transmitter that is read by a receiver within a specified distance. It is a Close Enough Vehicle Routing Problem (CEVRP) on a street network. In practice, there is uncertainty while reading meters. The signal transmitted by an RFID tag is discontinuous and each meter differs with respect to the specified distance. These factors can lead to missed reads. We address this uncertainty using Bayesian statistics and data analytics and perform simulation experiments to show that we design improved routes for the vehicles. 362A Simulation Contributed Session Chair: Xianyong Wang, Clemson University, Clemson, SC, United States, xianyon@clemson.edu 1 - Models for Merging Fire and Emergency Medical Units Cheng Hua, Yale School of Management, Yale School of Management, P.O. Box 208200, New Haven, CT, 06520-8200, United States, cheng.hua@yale.edu, Arthur J.Swersey, Fernando Y. Chiyoshi, Ana Paula Iannoni, Arthur J. Swersey We consider an innovative emergency service spatial queuing system in which there are crossed trained personnel assigned to fire-medic units that respond to fire alarms or emergency medical calls. A fire-medic unit may be in one of three states: available, busy at a fire incident, or busy at an emergency medical call. To make the problem computationally tractable for typical real-world systems, we devise an approximation that solves two 2-states systems to determine steady state probabilities and performance measures and discuss its application in St. Paul, Minn. The model includes both fire-medic units and separate fire and emergency medical units. 2 - Antecedents of Service Provision by Manufacturing Firms Sina Golara, Doctoral Candidate, Arizona State University, 1249 East Spence Avenue, Apt 336, Tempe, AZ, 85281, United States, sgolara@asu.edu, Nasim Mousavi, Kevin Dooley This paper empirically investigates the drivers of servitization by manufacturing firms. We suggest that the extant diversification theory does not completely explain servitization. We study a sample of 2450 public manufacturing firms for the period of 1976 to 2006. We find that firms in industries characterized with high technology or firms in their earlier life cycle stage offer a higher level of service. We also demonstrate that a larger market share and larger stock of proprietary knowledge encourage manufacturers to introduce more services. 3 - Service Failure Recovery Strategies from the Consumer Perspective Ahasan Harun, PhD Student, University of North Texas, 316 Fry Str,eet Apt 203, Denton, TX, 76201, United States, MdAhasanUddin.Harun@unt.edu, Md Rokonuzzaman, Victor R. Prybutok This research develops and tests a model about consumer positive word of mouth intention after service failure. Antecedents such as service recovery strategies, recovery evaluation, and justice perception are evaluated in the context of the proposed theoretical framework. In addition to identifying the factors, we develop a survey to measure effective means to overcome service failure in a real situation. This approach overcomes the shortcomings of the currently predominant scenario-based survey research. WA54

4 - Sharing Economy: A Literature Review and Directions for Future Research Xianyong Wang, Student, Clemson University, 100 Sirrine Hall, Clemson, SC, 29634, United States, xianyon@clemson.edu This paper aims to examine both empirical and theoretical studies published in leading business journals regarding sharing economy and hopes to provide a better understanding of its scope, business models, operations strategies, and its impacts on well-established industries. This paper also presents a theoretical framework to explore the potential directions for future research. 362B Economics Contributed Session Chair: Jing Xu, University of Pennsylvania, Philadelphia, PA, United States, xjing@sas.upenn.edu 1 - Multidimensional Bayesian Persuasion Youzong Xu, Xi’an Jiaotong-Liverpool University, 111 Ren’ai Road, Business Building 423B, Suzhou, 215123, China, xu.youzong@wustl.edu This paper studies a Bayesian persuasion game on a multidimensional state and action space. We characterize Sender’s optimal persuasion strategies and show that, when the conflict between Sender and Decision Maker is not very large, Sender benefits more from using a signaling system that provides correlated information for different dimensions of the state. When the conflict between Sender and Decision Maker is large, Sender benefits more from using a signaling system that provides independent information for different dimensions of the state. Regardless what signaling system that Sender uses, Decision Maker’s welfare is not affected by Sender’s persuasion. 2 - Dynamic Duopolistic Competition with Sticky Prices under Uncertainty Luca Colombo, Deakin University, Melbourne Burwood Campus, We extend Fershtman and Kamien (Econometrica, 1987) to a stochastic environment. We study a differential game where two quantity-setting firms compete over an infinite time horizon under the assumption that the market price does not adjust instantaneously to the level specified by its demand function and is subject to stochastic fluctuations (governed by a Wiener process). We derive analytically the closed-loop (feedback) equilibrium strategies and the evolution path of the expected price. We show that demand uncertainty is pro (anti) competitive if market size is relatively small (large). 3 - Stackelberg Leadership in a Productive Asset Oligopoly Paola Labrecciosa, Monash University, Department of Economics, We study a differential oligopoly game where m Stackelberg leaders and n-m Stackelberg followers are engaged in the exploitation of a common-pool renewable resource over an infinite time horizon. At each point in time, firms sell their harvest in the marketplace at a price that depends on total harvest. We derive a multiple-leader feedback Stackelberg equilibrium and proceed with a comparative dynamics analysis w.r.t. various parameters of interest. We also compare and contrast the multiple-leader feedback Stackelberg equilibrium with the Cournot equilibrium and obtain a number of novel results on the relative efficiency of the two equilibria. 4 - Network Externalities in a Buyer-seller Search Platform Jing Xu, University of Pennsylvania, 209 S 33rd Street, Math Dept, Philadelphia, PA, 19104, United States, xjing@sas.upenn.edu I model a setting where buyers search for sellers directly or through a platform with lower search costs, and the platform charges both sides for the transactions it facilitates. Equilibrium strategies are determined by two coexisting positive and negative cross-group externalities: (i) while buyers appreciate more choices of available sellers on the platform, (ii) increasing the number of available sellers makes the search for low-priced and high-value sellers harder due to an unfavorable price dispersion. A platform optimally adopts a threshold strategy of targeting sellers with lower prices to balance the competing externalities. WA55 221 Burwood Hwy, Melbourne, VIC 3125, Australia, luca.colombo@deakin.edu.au, Paola Labrecciosa Clayton Campus, Wellington Road, Melbourne, VIC 3800, Australia, paola.labrecciosa@monash.edu, Luca Colombo

461