Informs Annual Meeting 2017

MC19

INFORMS Houston – 2017

5 - Optimal Buffer Capacity of Steel Strips in Continuous Steel Pipe Manufacturing

depend on the state of the queueing system. The underlying Markov process can be a Markov chain with a finite/countable state space, or a (reflected) diffusion process. We study the ergodicity properties and characterize the joint stationary distributions of the queueing processes and underlying Markov processes under certain conditions.

Md Shahriar J.Hossain, Graduate Research Assistant, Louisiana State University, 4141 Burbank Dr. Apt. 3, Baton Rouge, LA, 70808, United States, msjhossain1@gmail.com, Bhaba R.Sarker The steel strip joining process in a seam welded pipe manufacturing line, is followed by a coil accumulator that facilitates buffer inventory. Size of the buffer determines the allowable time for strip joining. High buffer inventory increases cost for the accumulator but reduces the occurrence of line stoppage due to additional joining time. The problem is to find an appropriate buffer capacity that minimizes the total cost. Considering the issues of line stoppage, buffer storage and strip joining parameters, a total cost function is formulated and solved optimally in this research. 6 - Multi-item Inventory Optimization Approach under Multiple Target Levels Yossiri Adulyasak, Assistant Professor, HEC Montreal, 3000 Cote-Sainte-Catherine, GOL.department, Montreal, QC, H3T.2A7, Canada, yossiri.adulyasak@hec.ca, Eric Prescott Gagnon, Thierry Moisan This research focuses on a development of a multi-item inventory optimization approach under group target levels which is represented as a coupled Markov decision process (MDP). This problem is solved in a decomposition framework where a master problem will select a policy for each item that minimize overall costs while satisfying a set of targets such as a minimum cycle service level. A set of subproblems, one for each item, generates new inventory policies using an MDP approach. We discuss the practical considerations and extensions of the framework. 340A Queueing Systems Sponsored: Applied Probability Sponsored Session Chair: Yunan Liu, North Carolina State University, Raleigh, NC, 27695-7906, United States, yliu48@ncsu.edu 1 - Optimal Controls for Double-ended Queues with Time-varying Demand Continuous manufacturing has received much of recent attention both in academia and industry. We consider production systems with a time-varying demand, perishable inventory, and abandonment of backorders. The system incurs inventory-related costs of holding and perishment, and demand-related costs of waiting and abandonment. We study the finite-time production planning to minimize the sum of the demand and production related costs. We use fluid models to derive time-dependent production rates that are asymptotically optimal as the system scale increases; specifically, we prove a limit theorem to verify the MC17 Ling Zhang, North Carolina State University, Raleigh, NC, United States, lzhang42@ncsu.edu, Chihoon Lee, Xin Liu, Yunan Liu We study service differentiation in the presence of diverse customer needs and time-varying arrivals. We propose a new control family that exploits the head-of- line delay information. We refer to this family as the head-of-line-delay-ratio (HLDR) rule and show that it achieves desired differentiated service in a multi- class multi-server queue with non-stationary arrivals. 3 - Data-driven Inpatient Bed Assignment: Balancing Boarding and Overflowing using the P-model Balancing boarding and overflowing are a chanllenge to most public hospitals. To tackle with it, we provide a practical P-model approach, which is able to capture critical features of patient flow management, while keep the resulting optimization problem tractable. Based on a set of patient flow data from a hospital, our simulation study shows that the proposed approach can greatly reduce patients’ transfer delays and mitigate the time-of-day effect on boarding. Through the study with real data, we demonstrate that the P-model could be a useful tool for the control of queueing systems with time-sensitive service requirements. 4 - Queueing Models in Interactive Random Environments Guodong Pang, Penn State University, 310 Leonhard Bldg., Industrial and Manufacturing Engineering, University Park, PA, 16802, United States, gup3@psu.edu We study queueing models in interactive random environments. The arrival and/or service parameters are driven by a Markov process whose generators Shasha Han, NUS.Business School, Singapore, Singapore, shashahan@u.nus.edu, Shuangchi He, Hongchoon Oh asymptotic effectiveness of the optimal solution. 2 - Delay-based Service Differentiation in a Time-varying Environment Xu Sun, Columbia University, New York, NY, 10027, United States, xs2235@columbia.edu, Ward Whitt

MC18

340B Limit Order Book Sponsored: Applied Probability Sponsored Session

Chair: Raghu Pasupathy, Purdue University, pasupath@purdue.edu 1 - Optimal Placement of a Small Order under a Diffusive Limit Order Book Model

Jose E. Figueroa-Lopez, Washington University-St Louis, One Brookings Drive, St Louis, MO, 63130, United States, figueroa@math.wustl.edu, Hyoeun Lee, Raghu Pasupathy

We present a framework to study the optimal placement problem of a stock trader who wishes to clear his/her inventory by a predetermined time horizon by using a limit order or a market order. Building on diffusion approximations for LOB models, we consider the analogous problem in a diffusive market and characterize the optimal limit order placement under different market conditions. In particular, we propose a simple method to determine the optimal order placement and show its performance.

MC19

342A OM-Finance Interface Sponsored: Manufacturing & Service Oper Mgmt, iFORM Sponsored Session Chair: Jie Ning, Case Western Reserve University, Cleveland, OH, 44106, United States, jie.ning@case.edu Co-Chair: Qi Wu, Case Western Reserve University, Cleveland, OH, 44122, United States, qxw132@case.edu 1 - Financing and Supply Chain Network Structure John R.Birge, University of Chicago, Booth School of Business, 5807 South Woodlawn Avenue, Chicago, IL, 60637, United States, John.Birge@ChicagoBooth.edu Firms have differing incentives to form specific supply chain relationships depending on their relative positions downstream or upstream in the chain. These incentives including reliability of physical fulfillment and flexibility in financing. This talk will discuss a model that captures these dual roles of supply chain relationships and its implications for observed supply chain network configurations. 2 - Production and Capacity Management with Internal Financing Matthew J. Sobel, Case Western Reserve University - Retired, Department of Operations, 10900 Euclid Avenue, Cleveland, OH, 44106-7235, United States, matthew.sobel@case.edu, Jie Ning We analyze a price-taking firm that manages production and capacity, uses only internal financing, and faces stochastic market environments. The firm has two operationally independent production facilities, each of which makes two products, and a cash reserve which finances all operations and dividend issuance. We completely characterize the optimal policy and the endogenous values of the capacities and cash reserve, and show that they invite a real-option interpretation. We find that internal financing creates a spillover between the endogenous values of the two operationally independent facilities and we specify how this leads to interdependence of their optimal policies. 3 - Managing Inventory for a Multi-divisional Firm with Cash Pooling

Kevin Shang, Duke University, Fuqua School of Business, Duke University, Durham, NC, 27708, United States, khshang@duke.edu, Jianan Wang, Yi Yang

We consider a firm with N divisions, each replenishing inventory from an external supplier to satisfy a random, nonstationary demand in a finite horizon. The headquarter consolidates cash in a master account for external investments and purchasing inventory. The objective is to optimize the expected working capital at the end of the horizon. We derive near-optimal heuristics for the joint inventory and cash retention policy. We also investigate the value of cash pooling and the impact of cash on the order variability with correlated demands.

209