Informs Annual Meeting 2017

MB54

INFORMS Houston – 2017

4 - Traffic Equilibrium in a System with Interactions Between Drivers and Traffic Control Wei-Hua Lin, Associate Professor, University of Arizona, Systems and Industrial Engineering, Tucson, AZ, 85721, United States, whlin@email.arizona.edu We will assess how traffic pattern would change in response to a traffic control system in which the interaction between traffic control and individual drivers is made possible. Both short-term and long-term effects will be evaluated. Properties associated with the traffic equilibrium resulting from the interaction will be examined. 362A Application of Geospatial and IoT Data in Agriculture Invited: Agricultural Analytics Invited Session Chair: Siyuan Lu, IBM, IBM, Yorktown Heights, NY, 10598, United States, lus@us.ibm.com 1 - Thoreau: A Fully Buried Wireless Sensor Network for Soil and Plant Science Supratik Guha, PhD, University of Chicago, Chicago, IL, 60637, United States, guha@uchicago.edu, William Kent, Xufeng Zhang, Arseniy Andreyev, Cristina Negri, Monisha Ghosh, Jacob Gold I will describe our research that has led to an end-to-end fully buried wireless sensor network for monitoring soil properties as a function of time on the University of Chicago campus in Hyde Park, Chicago. The network, named Thoreau, uses commercial sensors for monitoring temperature, volumetric water content, water potential and electrical conductivity as representative examples that are integrated into low power sensor packages (including the microprocessor unit, radio and antenna) that are buried 6-14 inches below the ground. A SigFox low power IoT wireless network is used for connectivity and the wireless signal path includes subterranean transmission followed by refraction into the air above ground. All data is curated on the Cloud and available via thoreau.uchicago.edu in open access mode. We will describe our experiences with Thoreau over the past 9 months, and we argue that scalable, affordable, ubiquitous subterranean sensing networks are within reach and will have significant impact on agriculture, the environment, and our understanding of soil and plant science. 2 - Advances in Agricultural Analytics Joshua Woodard, Cornell University, Ithaca, NY, United States, jdw277@cornell.edu By 2050, the population will approach 10 billion people, placing enormous strain on the Earth’s resources, and food security is at increased risk from climate change. Meanwhile, the increased availability of high resolution geospatial, environmental, biophysical, and economic data—coupled with the expansion of computational capacity—is creating new opportunities for data-driven scientific discovery. This talk will discuss recent advances in the emerging fields of digital agriculture and agricultural analytics, including applications in policy analytics and risk assessment. 3 - Digital Agriculture and the Future of Farming Raj Khosla, Colorado State University, Fort Collins, CO, United States, Raj.Khosla@colostate.edu Precision Agriculture has been around for over two decades. The first decade had a strong focus on quantifying spatial variability in soils, the second decade spent significant time on science and technology of precision management of nutrients. Now with increasing adoption of Precision Ag there is interest in harnessing the power of data for making management decision based on evidence. The success of future farming practices, output, efficiency and sustainability, would rely on “farming the data” as much as “farming the ground”. This presentation will empower audience with research based information on how agriculture is embracing information technologies to feed the growing population. 362B Methods in Financial Engineering Sponsored: Financial Services Sponsored Session Chair: Lingfei Li, Chinese University of Hong Kong, Shatin, N T, Hong Kong, lfli@se.cuhk.edu.hk Co-Chair: Qi Wu, Chinese University of Hong Kong, Shatin, Hong Kong, qwu@se.cuhk.edu.hk 1 - Dynamic Mean CVaR Portfolio Selection and Time Consistency Induced Term Structure of CVaR MB54 MB55

Duan Li, The Chinese University of Hong Kong, Faculty of Engineering, Dept of Systems Engineering & Engineering Mgmt. Shatin, Hong Kong, dli@se.cuhk.edu.hk We derive in this research optimal dynamic mean-CVaR portfolio policy for general incomplete discrete-time markets. As the dynamic mean-CVaR problem formulation is time inconsistent, how to update investor’s VaR level and the trade-off between the mean and CVaR measures dynamically and adaptively is an important issue to address. We provide a complete answer to this question by deriving analytically the time-consistency induced term structure of CVaR in this research. 2 - Dark Pool Trading a Hawkes Process Approach Xuefeng Gao, Chinese University of Hong Kong, William M.W. Mong Engineering Building,, Room 606, Shatin, Hong Kong, xfgao@se.cuhk.edu.hk, Lingjiong Zhu, Xiang Zhou Dark pools are automated trading facilities which do not display bid and ask quotes to the public. In this talk, we use the Hawkes process to model the clustered arrival of trades in a dark pool and analyze various performance metrics including time-to-first-fill, time-to-complete-fill and the expected fill rate of a resting dark order. 3 - Negative Term Structure via Density Splitting Shingfan Chan, Chinese University of Hong Kong, William M.W. Mong Engineering Building, Rm 507, Shatin, NT, Hong Kong, qwu@se.cuhk.edu.hk, Qi Wu Persistent low interest rate environment demands a term structure model to have high conditional probabilities near zero point. We construct a “density splitting” technique to model this “sticky” behavior of yield curves. By breaking the continuity of conditional densities of risk factors that many existing models assume, we gain control of how much probability mass is allocated at a particular level. The method is flexible in that it applies to existing models, and allows sticky behavior at both negative and positive region. We apply this approach to the Nelson-Seigel family and calibrate to ECB and JPY curves. We found that bond and swap data prefer our model over existing shadow rate models. 4 - Analysis of Markov Chain Approximation for Option Pricing and Hedging: Grid Design and Convergence Behavior Lingfei Li, Chinese University of Hong Kong, 608 William M.W. Mong Engineering Building, Shatin, N.T, Hong Kong, lfli@se.cuhk.edu.hk, Gongqiu Zhang Continuous time Markov chain (CTMC) approximation is an intuitive and powerful method for pricing options in general Markovian models. This paper analyzes how grid design affects the convergence behavior of barrier and European options in general diffusion models. We obtain sharp estimates for the convergence rate of option price, delta and gamma. Based on our theoretical results, we propose a novel class of non-uniform grids, which enable the CTMC approximation method to price and hedge a large number of options with different strikes fast and accurately. Applicability of our results to jump models is discussed through numerical examples. 362C Fueling Station Location and Routing Decisions for Alternative Fuel Vehicles II Sponsored: Location Analysis Sponsored Session Chair: Ismail Capar, xas A&M University, College Station, TX, 77843-3367, United States, capar@tamu.edu 1 - Supply Chain Planning in a Cross-Dock Environment Manoj Vanajakumari, Texas A&M.University, 4318 Toddington Ln, College Station, TX, 77845, United States, manojuv@tamu.edu, Chelliah Sriskandarajah, Haoying Sun We consider a supply chain planning problem via efficient cross-docking for an oil field services Company. The supply chain problem involves minimizing the total inbound and outbound transportation costs and the inventory management costs at the cross-docking facility and the plants. We analyze two problems: (i) the company ships all products bound to each plants periodically, and (ii) the company outsources the delivery of each product to a 3PL, on an individual basis, to the plants. We develop a decision support systems for the company. MB56

184