Search

The climate anomalies caused by La Niña events have caused some economic losses and human casualties in several regions around the world. In this paper, we first visualize and analyze the climate data of four representative countries to obtain the impact of La Niña on different countries. Then a BP neural network model is built to predict the probability of La Niña events in different regions. Finally, taking six different provinces in southern China as examples, an indicator system is established, a fuzzy comprehensive evaluation algorithm is used to summarize the impacts of different regions and indicators under a La Niña event, and practical response strategies are proposed.

Many real-world service facilities are subject to probabilistic disruptions. Such disruptions often exhibit correlations that arise from shared external hazards or direct interactions among these facilities. This paper builds an overarching methodological framework for reliable facility location design under correlated facility disruptions. We first incorporate and extend the concepts of supporting station structure and quasi-probability from Li et al. (2013) and Xie et al. (2015), such that any correlated facility disruptions (positive and/or negative) can be equivalently represented by independent failures of a layer of properly constructed supporting stations, which are virtually added to the original facility system for capturing the effect of correlations among facilities. We then develop a compact mixed-integer mathematical model to optimize the facility location and customer assignment decisions in order to strike a balance between system reliability and cost efficiency. Lagrangian relaxation based algorithms, including modules for obtaining upper bound and lower bounds of relaxed subproblems, are proposed to effectively solve the optimization model. Numerical case studies are carried out to demonstrate the methodology, to test the performance of the framework, and to draw managerial insights.

In service systems with natural or anthropogenic barriers (e.g., rivers, railroads), customers who intend to visit facilities for service must first pass through certain network access points (e.g., bridges, railway crossings). Possible blockage or disruptions of these access points could change the customer-facility assignments or even affect reachability of various facilities, and thus introduce facility reliability and correlation issues. This paper incorporates network access points and their probabilistic failures into a joint optimization framework. A layer of network access points are added and connected to facilities to imply the real-world connections between facilities and access points. The access points are assumed to be subject to disruptions with site-dependent probabilities. We then develop a compact mixed-integer mathematical model to optimize the facility location and customer assignment decisions for the service systems design. Lagrangian relaxation based algorithms are designed to effectively solve the proposed model. Multiple case studies are constructed to test the model and the algorithm, and to demonstrate their performance and applicability.

Unconventional intersection designs have been used to increase the capacity of intersections that are over-saturated under conventional ones. However, existing unconventional designs typically require extra land space and their effectiveness often depends on drivers' familiarity with the uncommon operating rules. To overcome these challenges, we propose a new unconventional design, where movements that are mutually incompatible under the conventional design can be made compatible of each other by allocating exit lanes to them appropriately, thereby creating opportunities for capacity improvement. We develop a lane-based capacity optimization model that incorporates the allocation of exit lanes as decision variables. The model is formulated as a Binary Mixed Integer Linear Programming problem, which can be efficiently solved by standard branch-and-bound algorithms. Numerical experiments show that significant capacity improvement can be obtained under our design. Besides proposing a new unconventional design, we also contribute to the literature of lane-based signal optimization methods by providing a novel linear formulation for the latest, yet nonlinear, model described in Wong and Heydecker [Transportation Research Part B 45(4):667–681]. This improvement is methodologically beneficial as linear models are computationally more convenient than nonlinear ones. Copyright © 2017 John Wiley & Sons, Ltd.

Railroad companies rely on good call centers to reliably handle incoming crew/resource call demands so as to maintain efficient operations and customer services in their networks. This paper formulates a reliable caller districting problem which aims at partitioning an undirected network into a fixed number of districts. The demand of each district is assigned to crew caller desks (one primary desk, multiple backups) under possible desk disruption scenarios. We simultaneously take into account several operational criteria, such as district contiguity and compactness, workload balance, and caller desk service reliability. The resulted districting problem is modeled in the form of a challenging mixed-integer program, and we develop a customized heuristic algorithm (based on constructive heuristic and neighborhood search) to provide near-optimum solutions in a reasonable amount of time. Hypothetical and empirical numerical examples are presented to demonstrate the performance and effectiveness of our methodology for different network sizes and parameter settings. Managerial insights are also drawn.

In North America, thousands of locomotives operating in the railroad network require various types of mechanical work every day. This work may be unscheduled, periodic, or on demand and could be conducted at either a fixed facility or a movable facility. Each facility is characterized by the type of work it can provide. The long-term infrastructure planning of these facilities is vital to the efficiency of the railroad. In this paper, we develop a large-scale mixed-integer mathematical model for infrastructure planning. The model integrates and optimizes decisions about (1) Locations, capabilities, and capacities of fixed facilities, (2) Home locations and routing plans of movable facilities, and (3) Assignments of locomotive work demands to facilities. We propose a decomposition-based heuristic solution framework consisting of several underlying algorithms to solve this large-scale optimization model. Computational results on numerous scenario studies using field data from the railroad industry show that the proposed model and algorithms are capable of providing solutions that are significantly superior to the current practice. The successful application of our methodology to real-world railroad planning has clearly demonstrated the substantial cost-saving benefits of our models on infrastructure planning solutions.

Infrastructure facilities may be subject to probabilistic disruptions that compromise individual facility functionality as well as overall system performance. Disruptions of distributed facilities often exhibit complex spatial correlations, and thus it is difficult to describe them with succinct mathematical models. This paper proposes a new methodological framework for analyzing and modeling facility disruptions with general correlations. This framework first proposes pairwise transformations that unify three probabilistic representations (i.e., based on conditional, marginal, and scenario probabilities) of generally correlated disruption profile among multiple distributed facilities. Then facilities with any of these disruption profile representations can be augmented into an equivalent network structure consisting of additional supporting stations that experience only independent failures. This decomposition scheme largely reduces the complexity associated with system evaluation and optimization. We prove analytical properties of the transformations and the decomposition scheme, and illustrate the proposed methodological framework using a set of numerical case studies and sensitivity analyses. Managerial insights are also drawn.

It is well recognized that the left turn reduces the intersection capacity significantly, because some of the traffic lanes cannot be used to discharge vehicles during its green phases. In this paper, we operationalize the phase swap sorting strategy (Xuan, 2011) to use most, if not all, traffic lanes to discharge vehicles at the intersection cross-section to increase its capacity. We explicitly take into consideration all through, left- and right-turning movements on all arms and formulate the capacity maximization problem as a Binary-Mixed-Integer-Linear-Programming (BMILP) model. The model is efficiently solved by standard branch-and-bound algorithms and outputs optimal signal timings, lane allocations, and other decisions. Numerical experiments show that substantially higher reserve capacity can be obtained under our approach.

Unconventional intersection designs have been used to increase the capacity of intersections that are over-saturated under conventional ones. However, existing unconventional designs typically require extra land space and their effectiveness often depends on drivers' familiarity with the uncommon operating rules. To overcome these challenges, we propose a new unconventional design, where movements that are mutually incompatible under the conventional design can be made compatible of each other by allocating exit lanes to them appropriately, thereby creating opportunities for capacity improvement. We develop a lane-based capacity optimization model that incorporates the allocation of exit lanes as decision variables. The model is formulated as a Binary Mixed Integer Linear Programming problem, which can be efficiently solved by standard branch-and-bound algorithms. Numerical experiments show that significant capacity improvement can be obtained under our design. Besides proposing a new unconventional design, we also contribute to the literature of lane-based signal optimization methods by providing a novel linear formulation for the latest, yet nonlinear, model described in Wong and Heydecker (2011). This improvement is methodologically beneficial as linear models are computationally much more convenient than nonlinear models.

It is well recognized that the left turn reduces the intersection capacity significantly, because some of the tra ffic lanes cannot be used to discharge vehicles during its green phases. In this paper, we operationalize the phase swap sorting strategy (Xuan, 2011) to use most, if not all, traffi c lanes to discharge vehicles at the intersection cross section to increase its capacity. We explicitly take into consideration all through, left- and right-turning movements on all arms and formulate the capacity maximization problem as a Binary-Mixed-Integer-Linear-Programming (BMILP) model. The model is e fficiently solved by standard branch-and-bound algorithms and outputs optimal signal timings, lane allocations, and other decisions. Numerical experiments show that substantially higher reserve capacity can be obtained under our approach.