Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

WD32

risk analyses in China’s rail transportation projects. Then an improved fuzzy comprehensive risk evaluation model is established in order to evaluates the risks of projects from the perspective of economy and efficiency, which combining economical sensitivity analysis method with expert scoring method to determine the weight of each level’s index. Then the model verification is carried out with two examples of China’s rail transportation projects. The results from multiple angles is consistent with the actual situation and the risk control measures are put forward according to it. 3 - Railway Crew Scheduling with Semi-flexible Timetables Ulrich Thonemann, Universitat zu Koln, Wirtschafts und Sozialwissenschaftliche, Fakultat Albertus Magnus Platz, Koeln, D-50923, Germany, Christian Raehlmann We investigate the impact of coordinating the timetable and the crew schedule in an operational freight railway system. Usually both problems are solved sequentially - resulting in suboptimal schedules with long idle times for the train drivers. We coordinate the timetable and the crew schedule by adding flexibility to the timetable. We introduce small time windows, that allow to shift entire trains forwards and backwards by discrete time periods. We solve our model with a column generation heuristic and test it on three real datasets. Our results show large reductions in idle time and cost. 4 - Integrated Crew Scheduling & Crew Rostering Model for Rapid Transit Networks Manuel Fuentes, Universidad Rey Juan Carlos, Camino del Molino, 5, Fuenlabrada, 28943, Spain, Luis Cadarso Solving Crew Planning Problem sequentially, by tackling first the Crew Scheduling and then the Crew Rostering problems independently, usually leads to suboptimal solutions. Instead, facing the problem without splitting it into two may improve the global solution. However, this is often impractical when using exact methods, due to the complexity of the problem.We propose a heuristic approach for solving the integrated crew planning problem for rapid transit networks, where the schedules are usually daily repeated and the crews are sequenced in rotating rosters. n WD32 North Bldg 222B Transportation-Freight Contributed Session Chair: Peng Sun, K hne Logistics University, Gro er Grasbrook 17, Hamburg, 20457, Germany 1 - Freight Demand Synthesis with Mode Choice: Combined Estimation Procedure Freight Demand Synthesis (FDS) is the process of estimating freight demand from the available data such as traffic counts, cost matrix, productions, and attractions; bypassing the need for an extensive data collection efforts. This research develops a combined model for FDS that incorporates the estimation of modal split between rail and truck. A gravity model is adopted for the estimation of trip distribution; a binary logit model for modal split between rail and truck; and Noortman and van E’s model is used for empty trips. This research serves as a potential tool for transportation planners in evaluating various policy outcomes. 2 - Stochastic Drayage Scheduling Problem with Time Windows Samaneh Shiri, UPS, Baltimore, MD, United States, Nathan Huynh, ManWo Ng This work studies the drayage problem which involves transporting containers in the hinterland of an intermodal terminal. It relaxes the assumption of the deterministic duration of activities in drayage operation. Two chance-constrained programming models were develops with the following assumption: 1) only the mean and variance of activities are available, and 2) the mean as well as the lower and upper bounds of activities are available. To keep the problem tractable, the chance constraints are converted to their deterministic equivalents using Cantelli’s and Hoeffding’s inequalities. The models were solved using optimization software CPLEX and reactive tabu search algorithm. 3 - The Impact of Mass Transport on the Design of Multimodal Two-tier City Logistics Systems Pirmin Fontaine, Technical University of Munich, Arcisstrasse 21, LST Minner, Munich, 80333, Germany, Teodor Gabriel Crainic, Ola Jabali, Walter Rei We build on the service network design problem of a two-tier city logistics system where the goods are transported from distribution centers at the city border to satellites in the city from where the final distribution is done. Additionally, to classical trucks, we focus on the impact of mass transportation (e.g., trams) in such a system. We use a service-based integer programming formulation for the resulting multi-mode service network design problem with resource management. Decomposition techniques are used to solve the problem efficiently. The numerical results give insights on the advantages and disadvantages of the different transportation modes and what influences on the fleet structure. Lokesh Kumar Kalahasthi, Graduate Student, Rensselaer Polytechnic Institute, Troy, NY, 12180, United States, Jose Holguin-Veras, John E. Mitchell

n WD30 North Bldg 221C Practice – Shipping & Transportation Operations & Management I Contributed Session Chair: Qingcheng Zeng, Dalian Maritime University, School of Transportation Management, Dalian, 116026, China 1 - Revenue Management in Ocean Shipping Andres Iroume, Revenue Analytics, Atlanta, GA, United States, Michael Seelhorst Container shipping companies face many of the same issues addressed by traditional revenue management (RM): perishable inventory, demand uncertainty and customers with different price sensitivities, among others.Historically, operations research based decision support systems have been implemented in this industry, including empty container repositioning, terminal operations, disruption management and others.Today, some of the first revenue management systems are being implemented. We discuss nuances and similarities with previous applications of RM. 2 - Bi-level Optimization Method to Minimize Externalities by Means of Eco-transfer Staging Areas in Urban Cores Mario E. Arrieta-Prieto, Graduate Research Assistant, Rensselaer Polytechnic Institute, Troy, NY, 12180, United States, Abdelrahman Ismael, Carlos Rivera-González, John E. Mitchell Changes in urban population combined with the rise of e-commerce have had tremendous impacts on sustainable supply chains at urban cores. This paper proposes a methodology to decide optimal location of on-street staging areas in urban cores. By using a bi-level optimization problem the authors were able to capture how the agents involved in the decision pursue different objectives. In freight systems the public sector aims to maximize social welfare while the private sector looks for maximizing profit. The result of the research exhibits a formulation and a case study scenario in Manhattan that allows the evaluation of on-street staging areas from both the private and the public sector perspective. 3 - Modelling the AGV and ALV System in Automated Container Terminals Qingcheng Zeng, Dean, Dalian Maritime University, Linghai Road 1, Dalian, Dalian, 116026, China This paper addresses the impact of the number of operations on the terminal efficiency based on AGV and ALV transport systems. Performance indicators of AGV and ALV transport system is obtained by queuing models, where AGV transport system is modelled by a closed queuing network and ALV transport system is modelled by a mixed queuing network. We also developed a cost model to investigate the present values of total cost of AGVs and ALVs, including operation cost, capital cost and labour cost. Numerical experiments indicate that the efficiency of ALVs and AGVs. Sensitive analysis shows the impact of the number of buffers and vehicle travelling speed on terminal efficiency. n WD31 North Bldg 222A Practice- Rail Transportation I Contributed Session Chair: Manuel Fuentes, Universidad Rey Juan Carlos, Camino del Molino, 5, Fuenlabrada, 28943, Spain 1 - A Capacity Estimation System for High-speed Railway: An Integrated Approach Considering Railway Traffic and Rolling Stock Utilization Haiying Li, Prof., Beijing Jiaotong University, Shangyuan Cun No 3, Beijing, 100044, China, Zhengwen Liao, Ying Wang, Xinyi Li A capacity estimation approach considering both the passing capacity of railway open track segments and stations, as well as the holding and the circulation of rolling stocks is introduced in the presentation. A MIP model which jointly optimizes the timetabling, track assignment and rolling stock circulation problems is proposed to generate a saturated integrated plan. For solving the MIP model, a pressure testing method is embedded in a time-rolling algorithm framework for generating a whole-day train timetable. Based on the algorithm, a decision support system with the user interface is developed and applied in a case study on Beijing-Tianjin intercity railway. 2 - Research on the Risk Fuzzy Evaluation and Control Measures of China’s Rail Transportation Projects Based on Economic Sensitivity Xu Yan, Southwest Jiaotong University, Chengdu, China National United Engineering Laboratory of Integrated and Intelligent Transportation, Chengdu, China, Qiyuan Peng, Jie Liu, Yong Yin, Yongxiang Zhang This paper construct a multi-level risk evaluation system based on the sources of

505