Informs Annual Meeting 2017

SC55

INFORMS Houston – 2017

2 - Special Event Transit Demand Estimation using AFC Data Pramesh Kumar, University of Minnesota, Minneapolis, MN, 55414, United States, kumar372@umn.edu, Alireza Khani An improved trip chaining algorithm is proposed for transit origin-destination matrix estimation using Automatic Fare Collection (AFC) data. An optimization model is proposed to decompose the OD matrix ta regular and irregular demand, to capture the effect of special events on transit demand. Results from the Twin Cities transit network are presented. 3 - Observability Quantification of Dynamic Transit Systems with Multi-source Sensor Data Jiangtao Liu, Arizona State University, 2026 S.Hammond Drive, Apt 205, Tempe, AZ, 85252, United States, jliu215@asu.edu, Xuesong Zhou A transit network with low observability creates barriers for effective and reliable transit service supply. By incorporating the passenger-based trip time from smart card and flow count data from surveillance system, we propose an agent-based arc-based mixed integer programming (MIP) to estimate the whole transit system state in a time-discretized space-time network. Further, an agent-based path- based binary integer programming (IP) model is offered to quantify the flow range of targeted paths to evaluate the system observability under the estimated transit state before. Chair: Shrikant Jarugumilli, Monsanto, 12447 Bennett Springs Ct, St. Louis, MO, 63146, United States, shrikant.jarugumilli@monsanto.com 1 - Field Production Planning Based on Operational Rogerio Lenza, Monsanto Company, Rua Joaquim Nabuco, 163 ap 1501, Maringa, Brazil, rogerio.p.lenza@monsanto.com This project has made use of analytics through operational research built with a novel technology in order to deliver a Crop Placement solution for the corn business in Brazil. This solution allows the generation of multiple what-if scenarios focused on minimizing production cost considering the entire supply chain. Additionally, it has transformed the overall crop planning process, improved the quality of the production plan and increased the productivity of the planning team for all corn plants. 2 - Integrated Supply Chain Optimization at Monsanto Utilizing Advanced Analytics Barry Surber, Monsanto, Creve Coeru, MT, United States, barry.l.surber@monsanto.com The flow of material through the production and distribution processes within a Supply Chain is far too often managed through a series of individual decisions that are analyzed independently of the processes that proceed or follow them. In order to efficiently manage these processes in an optimal manner a series of user interfaces, processes, algorithms and an optimization engine was developed to integrate the decision making processes into one system that considers the costs and constraints from these individual decisions in one integrated solution. This solution was developed in-house and has met with outstanding acceptance due to its ease of use and logical process flow for creating scenarios. 3 - Machine Learning Based Simulation and Optimization of Soybean Variety Selection Durai Sundaramoorthi, Washington University in Saint Louis, 10352 Conway Road, Saint Louis, MO, 63131, United States, dsundaramoorthi@gmail.com, Lingxiu Dong, Yu Li, Xiao Tan, Piruthiviraj Sivaraj Humanity is facing the greatest challenge of feeding itself. The World Food Programme estimates that about 795 million people do not have adequate food to have a healthy life. About 3.1 million children die every year because of poor nutrition. As a small part of addressing this great humanitarian challenge, this research proposes an analytics framework for growing the optimal soybean varieties. Selecting soybean varieties for planting is an important decision that has significant implications for the yield of the farm. We formulated and solved a simulation-based optimization problem to determine the optimal soybean-mix to minimize the risk associated with the yield. SC54 362A Operations Research in Agriculture Invited: Agricultural Analytics Invited Session

4 - Application of Vehicle Routing with Time Windows and Split Deliveries in Agriculture Hadi Panahi, Monsanto, St, Louis, MO, United States, hadi.panahi@monsanto.com, Naveen Singla Every year Monsanto’s testing pipeline tests new seeds for their yield and traits. Seeds are planted in the fields, and harvested for data collection. Advancement decisions can be made once data is collected throughout the harvest season. However, the timing of harvest has a direct effect on the quality of the data. To collect the highest data quality to enable making reliable advancement decisions, first the optimal harvest time is determined via predictive modeling and next harvesters (combines) are deployed from hubs to fields within the recommended harvest time window. In the literature, this problem is studied as Vehicle Routing Problem with time windows and split deliveries (VRPTWSD) as multiple combines can visit the same field. A mixed-integer programming model (MIP) along with a heuristic providing feasible solutions to the MIP as warm start are developed for this problem. Agricultural business rules and objectives are also captured by the MIP model and the heuristic algorithm. This model has been used to provide operational routes during harvest season and to determine the fleet size. 362B Stochastic Systems in Finance Sponsored: Financial Services Sponsored Session Chair: Alexandra Chronopoulou, achronop@illinois.edu 1 - Indifference Pricing of XVA in Stochastic Volatility Models Stephan Sturm, WPI, Department of Mathematical Sciences, 100 Institute Road, Worcester, MA, 01609, United States, ssturm@wpi.edu We consider the pricing of XVA in markets with stochastic volatility. Assuming that the hedger’s risk preferences are given by a convex dynamic risk measure, we study the hedger’s indifference price for entering the underlying contract with value adjustments due to credit risk, collateralization and differential rates. 2 - Trading Illiquid Goods: Market Making as a Sequence of Sealed-Bid Auctions Andrew Papanicolaou, NYU, Tandon, Brooklyn, NY, United States, ap1345@nyu.edu We provide analytic results for the optimal control problem faced by a market maker who can only obtain and dispose of inventory via a sequence of sealed-bid auctions. Under the assumption that the best competing response is exponentially distributed around a commonly discerned fair market price we examine properties of the market maker’s optimal behavior. We show that simple adjustments to skew and width accommodate customer arrival imbalance. We derive a straightforward relationship between the market marker’s fill probability and direct holding costs. We propose a stochastic delay financial model which describes influences driven by uncertain historical events. The underlying is modeled by stochastic delay differential equation (SDDE). 4 - Designing Clearinghouse Default Funds: the Role of Risk-taking Incentives Agostino Capponi, Columbia University, 500 W. 120th Street, New York, NY, 10027, United States, ac3827@columbia.edu, Jessie Wang, Hongzhong Zhang Central counterparties (CCPs) are mandated to reduce counterparty risk in the over-the-counter derivatives markets. We study the optimal design of the default fund contributions, serving the purpose of mutualizing losses of defaulted clearing members. While the default funds allow members to effectively share counterparty risk ex post, we highlight a novel mechanism related to loss mutualization and inducing members to take excess risk ex ante due to an inherent externality among them. We show that the CCPs can mitigate such an inefficiency by optimally choosing a default fund level, balancing the ex post risk- sharing and ex ante risk-taking. SC55 3 - A Financial Market of a Stochastic Delay Equation Kiseop Lee, Purdue University, kiseop@purdue.edu

87