2013 Session: 393

• Evaluating Beat Structure and Truck Allocation for the Tarrant County Courtesy Patrol
  Abstract: Non-recurring congestion poses a significant concern to urban freeway drivers. Effective incident management relies on many tools to lessen the overall impact of crashes, road debris, and stalled/disabled vehicles. Many urban areas have adopted freeway service patrol (FSP) programs that patrol the freeway network searching for incidents, providing aid to motorists, and assisting with incident management and clearance. FSP management must consider the beat structure and fleet allocation. This study uses both deterministic and probabilistic response time estimations for each beat to assess different fleet allocations. The goal of the research is to consider whether the urban network should be segmented into as many beats as possible with individual trucks assigned to each beat or if additional trucks should be allocated to fewer beats.In an effort to explore the truck allocation problem with field data, the study uses the Tarrant County Courtesy Patrol (CP) as a case study. The Tarrant County CP typically uses a one-beat, two-beat or three-beat configuration with a single truck allocated to each beat. This study explores the merits of two-beat configuration. This study indicates that the incident distribution must be considered during both beat and tour design.
  Authors: Daneshgar, Farzad; Mattingly, Stephen P.; Haghani, Ali
  Authors: Daneshgar, Farzad; Mattingly, Stephen P.; Haghani, Ali
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-0725
• Continuum Approximation Approach to Competitive FacilityLocation Design Under Facility Disruption Risks
  Abstract: This paper presents game-theoretical models based on a continuous approximation (CA) scheme to optimize service facility location design under spatial competition and facility disruption risks. The share of customer demand in a market highly depends on the functionality of service facilities and the presence of nearby competitors, as customers normally seek the nearest functioning facility for service. Our game-theoretical models incorporate these complicating factors into an integrated framework, and use continuous and differentiable density functions to represent discrete location decisions. We first analyze the existence of Nash equilibria in a symmetric two-company competition. Then we build a leader-follower Stackelberg competition model to derive the optimal facility location design when one of the companies have first-move advantage over its competitor. Both models are solved efficiently, and closed-form analytical solutions can be obtained for special case. Numerical experiments (with hypothetical and empirical data) are conducted to show the impacts of competition, facility disruption risks and transportation cost metrics on the optimal design. Interesting properties of the models cast managerial insights to real-world problems.
  Authors: Wang, Xin; Ouyang, Yanfeng
  Authors: Wang, Xin; Ouyang, Yanfeng
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-0797
• Network Design of Logistics Systems with Expedited Shipment Services
  Abstract: This paper studied an integrated logistics network design problem that optimizes supplier locations and their assignments to terminal facilities, expedited shipment configurations and inventory management strategies in an uncertain environment. We first formulated elementary models for certain special case problems and discuss their model properties and solution methods. Built on these developments, we proposed a mathematical programming model that minimizes the sum of supplier set-up costs, expected shipment costs for both regular and expedited services, and expected inventory holding cost under stochastic demand rates and transportation lead times for a general network logistics system throughout the planning horizon. Due to the consideration of interdependent operational components including inventory management and shipment expedition, the proposed model is of high non-linearities and involve complex discrete structures. After studying problem element properties, we developed a customized solution approach based on the Lagrangian relaxation technique that can solve the model efficiently and accurately. This approach has been tested on a number of instances on three logistics networks of different scales, and their results uniformly showed that this approach can solve a realistic problem instance to its exact optimum or a near-optimum solution in a short time. We presented extensive experimental results to draw managerial insights into how problem settings and key parameter values affect the optimal design results, including cost component magnitudes, transportation configurations, inventory management strategies and network layouts. We noted that under the optimal design, utilizing expedited shipment services actually does not incur too much extra cost overall while guaranteeing service reliability. We also found under the integrated design, all planning and operational components complement each other in the optimal way, and thus it is important and necessary to consider all the planning and operational decisions all together in a holistic design framework.
  Authors: Li, Xiaopeng
  Authors: Li, Xiaopeng
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-1208
• Game Theory-Based Optimization of Taxi Fare in Taxi Market of China
  Abstract: The hearing system provides a game platform for dealing with the optimization problem of taxi fare due to the taxi market regulations and the variation of operating cost in taxi market of China. In this paper, a bi-level programming model is proposed for optimization of taxi fare in monopoly market, as simultaneously considers the equilibrium between the social welfare and profit of taxi firms. The upper-level problem is a set of formulations ensuring maximization of social welfare under constrain on taxi fare restriction of government in taxi industry. The lower-level model aims to maximize the profit of taxi firms from fare revenue and maintains the positive value of firm profit and supply-demand equilibrium of taxi market. The Lagrangian approach is used to transform the lower model into upper model with K-K-T conditions, and the bi-level programing model becomes a single-level programing model. The Genetic Algorithm and Simulated Annealing algorithm are respectively designed to solve the model. A numerical calculation is presented to illustrate the accuracy and efficiency of proposed model and algorithms in a real urban road network of Harbin.
  Authors: Wang, Jian; Sun, Guanglin; Hu, Xiaowei
  Authors: Wang, Jian; Sun, Guanglin; Hu, Xiaowei
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-3243
• Transit Bottleneck Model and Its Application to a Transit Line in Paris
  Abstract: The paper addresses the issue of passenger waiting and being stored at a station platform, from which point they plan to board transit services towards egress stations. Each transit service has a specific set of downstream egress stations and is operated at given frequency using homogeneous vehicles of limited available capacity. The model yields individual waiting time by egress station and the assignment of vehicle capacity to the flows by egress station.Two cases are distinguished, unsaturated versus saturated. The unsaturated case is addressed by standard line combination, where service frequency is added up among the routes that service a given egress station. The saturated case is addressed by making explicit the average number of passengers waiting on platform for a given egress station. From these passenger stocks is derived the individual probability to board a vehicle of limited capacity that service a given route hence a given subset of egress stations. Waiting passengers are assumed to be mingling on the origin platform. The subset of routes that service a given egress station, their vehicle capacities and the boarding probabilities induce a line capacity for that destination: to this is faced the passenger flow demanded during the assignment period, in a bottleneck model that yields an average waiting time per passenger.The vector of passenger stocks by egress station is shown to satisfy a fixed point problem. The existence and uniqueness of the solution are demonstrated on the basis of an equivalent, convex minimization program.
  Authors: Leurent, Fabien M.; Chandakas, Ektoras
  Authors: Leurent, Fabien M.; Chandakas, Ektoras
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-3623
• Supporting Station Model for Reliable Infrastructure Location Design Under Interdependent Disruptions
  Abstract: This paper proposes a new modeling method that transforms an infrastructure system with interdependent and correlated facility failures into an equivalent one with an explicit supporting structure. This structure consists of a set of supporting stations that are subject to only independent disruptions with identified probabilities, and thus is much easier to characterize and formulate. Such a supporting structure framework is capable of not only providing a mathematical representation of facility failure mechanisms but also physically emulating interdependent infrastructures and their inter-connections in many real-world systems. We examine the properties of this structure and find that it can be used to model a range of heterogeneous and correlated facility failure patterns. A mathematical model built on the supporting structure is created to solve reliable facility location design problems under correlated facility failure risks. This model determines the optimal locations for supporting stations and service facilities to minimize the total system cost including infrastructure investment in the early planning stage and the expected transportation costs and service-loss penalties during the operational stage. This model is formulated into a compact integer linear program and can be efficiently solved by state-of-the-art solvers. A set of experiments and case studies are conducted to demonstrate the applicability of the proposed model and to draw managerial insights into the optimal system design.
  Authors: Li, Xiaopeng; Ouyang, Yanfeng; Peng, Fan
  Authors: Li, Xiaopeng; Ouyang, Yanfeng; Peng, Fan
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-3747
• Adaptive Transit Routing in Stochastic Time-Dependent Networks
  Abstract: An adaptive transit routing (ATR) problem in a stochastic time-dependent transit network is defined and is formulated as a finite horizon Markov Decision Process (MDP). Routing strategies are defined to be conditional on the arrival times at intermediate nodes, and real time information on the location and arrival times of other buses in the network. The objective is to find a strategy that minimizes the expected travel time, subject to constraints that guarantee that the destination is reached within a certain threshold. The problem inherits the curse of dimensionality and state space reduction through pre-processing is achieved by solving variants of the time dependent shortest path problem. An interesting analogy between the state space reduction techniques and the concept of light cones is discussed. A dynamic program framework to solve the problem is developed and numerical results on a small instance of the Austin transit network are presented to investigate the extent of state space reduction using the proposed methods.
  Authors: Rambha, Tarun; Boyles, Stephen; Waller, S. Travis
  Authors: Rambha, Tarun; Boyles, Stephen; Waller, S. Travis
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-4210
• A Multiobjective, Stochastic, and Capacity-Constrained Static Location Model for Ambulances
  Abstract: In this study, a new static ambulance location model is presented. To develop this model, we first introduce measures that an ideal ambulance location model is expected to include. Seven measures are defined for the ideal model. Then previous models are adapted to these seven measures to see which model is logically the best among models presented up until now. Our analysis shows that at best only five out of these seven measures have been addressed by the current models. We propose a new ambulance location model which can cover all the seven measures. This model is a multi-objective, probabilistic, and capacity-constrained location model (MPCLM). The MPCLM is the first model that can determine simultaneously location of ambulances and their under-coverage regions. To show its applicability in real world, the MPCLM is run for a large city of approximately 3 million people. The Fuzzy Goal Programming (FGP), as an approach to solve multi-objective problems, is employed. Sensitivity analysis of its variables shows that the model is flexible and works very well in different situations.
  Authors: Karimi, Behzad; Shafahi, Yousef; Mohammadian, Abolfazl; Sturm, Karl
  Authors: Karimi, Behzad; Shafahi, Yousef; Mohammadian, Abolfazl; Sturm, Karl
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-4972
• Continuum Approximation of Discrete Spatial Data for Optimal Facility Location Design
  Abstract: The paper proposes methodologies to convert discrete spatial data into continuous functions in a metric space so that the continuous approximation (CA) optimization framework can be applied to a general class of discrete facility location problems. We integrated a range of computational geometry methods to estimate the optimal system costs and the optimal number of facilities, and then developed an enhanced disk model to compute near optimal facility location design, both based on discrete input data. The results from the proposed CA method can be further improved by neighborhood search algorithms. Numerical experiments show that the proposed CA framework effectively finds near-optimum solutions to very large problem instances within a short time.
  Authors: Peng, Fan; Ouyang, Yanfeng; Wang, Xin
  Authors: Peng, Fan; Ouyang, Yanfeng; Wang, Xin
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-0791
  Practice-Ready: Yes
• Optimization Model for Inventory Routing Problem with Third-Party Logistics Provider Consideration
  Abstract: In this research, the inventory control and transportation of syrup concentrate and final products for one bottling company working for a beverage company is studied. Bottling companies get syrup concentrate from a beverage company to mix it with different ingredients to produce various products and distribute them to retailers. Unsatisfied orders have several harmful effects on the bottling company. The main reason for not serving one retailer is that the number of vehicles is not sufficient for delivery to all retailers. One method for preventing missed orders is sending orders to some retailers in advance to hold for future use. Another way to deal with this problem is by renting vehicles, which increases the fleet size. The last is outsourcing shipping and/or warehousing. This problem belongs to Inventory Routing Problem (IRP) with some new features such as options for rental vehicle and Third Party Logistics Providers (TPLP). The proposed model in this study includes several time steps in which a decision in one time step can affect future time steps. The proposed model is a multi-tier, multi-plant, multi-warehouse, and multi-product model which considers non-homogeneous fleet. No model in the literature considers all of these characteristics simultaneously.
  Authors: Sadrsadat, Hadi; Haghani, Ali; Sharifi, Elham
  Authors: Sadrsadat, Hadi; Haghani, Ali; Sharifi, Elham
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-1193
  Practice-Ready: Yes
• Limited Information Sharing Strategy for the Taxi-Customer Searching Problem in Non-Booking Taxi Service
  Abstract: One of the issues existed in the current taxi service is the imbalance between supply and demand. In response to this issue, the automatic taxi dispatching approach in which customers can book taxis through phones or mobile devices is widely used in many large cities worldwide. However, the utilization of this approach is not satisfied, since most customers still prefer the Non-Booking Taxi Service (NBTS) that taking the taxi by either waiting at taxi stand or hailing on the street. One important reason for this phenomenon is that the customer takes lower risk in NBTS: they are free from complicated booking procedures and have no commitment to any yet-arrived taxis. In order to facilitate the taxi-customer matching process in NBTS, this paper has proposed a novel control strategy namely the Limited Information Sharing Strategy (LISS) for the Taxi-Customer Searching Problem (TCSP) in NBTS, in which both the taxi and the customer are equipped with mobile devices that can communicate with each other within limited searching ranges. The proposed LISS is based on the game-theoretical formulation in which a learning algorithm is developed to find the pure Nash-Equilibrium (NE). A microscopic traffic simulation model is developed for the evaluation of the LISS. The simulation results show that the proposed LISS is an effective control strategy when taxi supply is low, and it will not increase the risk of taxi in terms of losing the total occupied time.
  Authors: Lee, Der-Horng; Wu, Xian; Sun, Lijun
  Authors: Lee, Der-Horng; Wu, Xian; Sun, Lijun
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-1981
  Practice-Ready: Yes
• Optimization of Transit Operation Strategies: Case Study of Guangzhou, China
  Abstract: This paper aims to optimize transit operation strategies including fare structure and service frequency to obtain demand-supply equilibrium in transit systems such as bus, BRT and metro. Transit fare structure and service level have a significant impact on passenger mode choice and system welfare. Considering these impacts in objective function, we propose an optimization model with constraints on fare control, capacity, budget and flow to meet both the agencies' and users' expectation to transit services. The penalty function method is adopted to simplify the optimization model into a general programming model with linear constraints. The Genetic Algorithm (GA) and the Simulated Annealing algorithm (SA) are used to obtain near-optimum solutions. Finally, the optimal transit operation strategies are applied to a real-scale network of Guangzhou to test the model and the suggested algorithms.
  Authors: Wang, Jian; Sun, Guanglin; Hu, Xiaowei
  Authors: Wang, Jian; Sun, Guanglin; Hu, Xiaowei
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-5002
  Practice-Ready: Yes
• U.S. Postal Airmail Routing Optimization
  Abstract: This paper studies an airmail routing problem at the Unites States Postal Services (USPS). Each day, millions of collected airmail of different priorities is shipped between local mail Processing and Distribution Centers (P&DCs) of the USPS through commercial air. The air carriers provide flight services according to the contracts. Each available flight has a finite, maximum capacity for airmail. In addition, the shipping between P&DCs incurs a cost that is determined by the mail type, volume and shipping distance. The research objective is to decide airmail allocation onto the flights subject to the contracted air capacity to minimize the system shipping cost while maintaining service quality. An integrated mathematical model is developed to design air routes and allocate airmail to routes optimally, as compared with a practice of isolated scheduling at each P&DC. A column generation method is proposed for this model. A numerical test is conducted with operational data. The test not only shows a significant increase in the total volume that can be shipped through the limited capacities but also provides a better service in terms of in time and early deliveries.
  Authors: Li, Yihua; Miao, Qing; Wang, Bruce
  Authors: Li, Yihua; Miao, Qing; Wang, Bruce
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-4708
  Practice-Ready: Yes
• Evaluating Beat Structure and Truck Allocation for the Tarrant County Courtesy Patrol
  Authors: Daneshgar, Farzad
  Authors: Daneshgar, Farzad
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-0725
• Limited Information Sharing Strategy for the Taxi-Customer Searching Problem in Non-Booking Taxi Service
  Authors: Lee, Der-Horng
  Authors: Lee, Der-Horng
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-1981
• Continuum Approximation of Discrete Spatial Data for Optimal Facility Location Design
  Authors: Ouyang, Yanfeng
  Authors: Ouyang, Yanfeng
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-0791
• Continuum Approximation Approach to Competitive Facility Location Design Under Facility Disruption Risks
  Authors: Ouyang, Yanfeng
  Authors: Ouyang, Yanfeng
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-0797
• Network Design of Logistics Systems with Expedited Shipment Services
  Authors: Li, Xiaopeng
  Authors: Li, Xiaopeng
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-1208
• Optimization Model for Inventory Routing Problem with Third-Party Logistics Provider Consideration
  Authors: Sharifi, Elham
  Authors: Sharifi, Elham
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-1193
• Limited Information Sharing Strategy for the Taxi-Customer Searching Problem in Non-Booking Taxi Service
  Authors: Sun, Lijun
  Authors: Sun, Lijun
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-1981
• Limited Information Sharing Strategy for the Taxi-Customer Searching Problem in Non-Booking Taxi Service
  Authors: Wu, Xian
  Authors: Wu, Xian
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-1981
• Game Theory-Based Optimization of Taxi Fare in Taxi Market of China
  Authors: Sun, Guanglin
  Authors: Sun, Guanglin
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-3243
• U.S. Postal Airmail Routing Optimization
  Authors: Miao, Qing
  Authors: Miao, Qing
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-4708
• Supporting Station Model for Reliable Infrastructure Location Design Under Interdependent Disruptions
  Authors: Li, Xiaopeng
  Authors: Li, Xiaopeng
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-3747
• Transit Bottleneck Model and Its Application to a Transit Line in Paris
  Authors: Chandakas, Ektoras
  Authors: Chandakas, Ektoras
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-3623
• Optimal Allocation of Service Frequencies over Transit Network Routes and Time Periods: Formulation, Solution, and Implementation Using Bus Route Patterns
  Authors: Verbas, Ismail Omer
  Authors: Verbas, Ismail Omer
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-5267
• Optimal Allocation of Service Frequencies over Transit Network Routes and Time Periods: Formulation, Solution, and Implementation Using Bus Route Patterns
  Authors: Mahmassani, Hani
  Authors: Mahmassani, Hani
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-5267
• A Multiobjective, Stochastic, and Capacity-Constrained Static Location Model for Ambulances
  Authors: Karimi, Behzad
  Authors: Karimi, Behzad
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-4972
• Adaptive Transit Routing in Stochastic Time-Dependent Networks
  Authors: Rambha, Tarun
  Authors: Rambha, Tarun
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-4210
• Optimization of Transit Operation Strategies: Case Study of Guangzhou, China
  Authors: Sun, Guanglin
  Authors: Sun, Guanglin
  Year: 2013
  Document Type: Presentation
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-5002
• Optimal Allocation of Service Frequencies over Transit Network Routes and Time Periods: Formulation, Solution, and Implementation Using Bus Route Patterns
  Abstract:

  This study proposes a formulation for the Transit Network Frequency Setting Problem that provides an optimal allocation of resources over space and time, while recognizing the existence of multiple service patterns along each bus route. Transit agencies have to allocate their limited resources optimally in order to maximize user benefits, operator benefits or a combination of the two. The coupling between the routes and the service patterns provided along (all or portions of) these routes is effectively captured, and thereby the user perspective and the operator perspective are merged into one formulation. The service patterns may be scheduled with different subsets of stops for a given route. On the other hand, users see the resulting combined route headways at the stops. The number of riders is elastic to the prevailing number of bus trips at a given stop, which is the combination of different pattern dispatch frequencies. Two main formulations are introduced. The first one extends previous work of Furth and Wilson, and seeks to maximize the number of riders and the total wait-time savings, under budget, fleet, policy headway and bus loading constraints. The second one minimizes the net cost under fleet, policy headway, bus loading, minimum ridership and minimum wait time savings constraints. In both formulations, pattern headways in different time of day intervals are the decision variables. This study provides the mathematical formulation underlying the proposed methodology, describes solution method and implementation, and uses it to demonstrate, by example, important properties of the frequency setting problem in this context, including some that may at first seem counterintuitive.

  Authors: Verbas, Ismail Omer; Mahmassani, Hani S.
  Authors: Verbas, Ismail Omer; Mahmassani, Hani S.
  Year: 2013
  Document Type: Paper
  Subject: Planning and Forecasting
  Session: 393
  Paper Number: 13-5267