Dynamic Optimal Traffic Assignment and Signal Time Optimization Using Genetic Algorithms

Abstract:   In this article a dynamic system-optimal traffic assignment model is formulated for a congested urban road network with a number of signalized intersections. A simulation-based approach is employed for the case of multiple-origin-multiple-destination traffic flows. The artificial intelligence technique of genetic algorithms (GAs) is used to minimize the overall travel cost in the network with fixed signal timings and optimization of signal timings. The proposed method is applied to the example network and results are discussed. It is concluded that GAs allow the relaxation of many of the assumptions that may be needed to solve the problem analytically by traditional methods.     

Optimal Road User Charging Cordon Design: A Heuristic Optimization Approach

Abstract:   This article addresses the optimal design problem of selecting a charging cordon in a general traffic network. A charging cordon is a set of tolled links surrounding a designated area so that all travelers entering or passing through this area will be tolled. Travelers in the network are assumed to respond to the tolls imposed by adjusting their behaviors to achieve a new equilibrium following Wardrop's equilibrium condition. The necessity of this equilibrium condition is imposed as one of the constraints in the optimal charging cordon design problem. This problem can be categorized as a Mathematical Program with Equilibrium Constraints (MPEC). This article presents an innovative Genetic Algorithm (GA) based method to tackle the problem. A new framework, called branch–tree framework, is developed to represent a closed charging cordon so that the method of GA can be used. The method is tested with a network of Edinburgh. Although the proposed algorithm is a heuristic-based method, the optimization result in the test is very promising. The optimal closed charging cordon as found by the algorithm produces a significantly higher benefit than that of judgmental cordons .     

Bi-level Programming Formulation and Heuristic Solution Approach for Dynamic Traffic Signal Optimization

Abstract:   Although dynamic traffic control and traffic assignment are intimately connected in the framework of Intelligent Transportation Systems (ITS), they have been developed independent of one another by most existing research. Conventional methods of signal timing optimization assume given traffic flow pattern, whereas traffic assignment is performed with the assumption of fixed signal timing. This study develops a bi-level programming formulation and heuristic solution approach (HSA) for dynamic traffic signal optimization in networks with time-dependent demand and stochastic route choice. In the bi-level programming model, the upper level problem represents the decision-making behavior (signal control) of the system manager, while the user travel behavior is represented at the lower level. The HSA consists of a Genetic Algorithm (GA) and a Cell Transmission Simulation (CTS) based Incremental Logit Assignment (ILA) procedure. GA is used to seek the upper level signal control variables. ILA is developed to find user optimal flow pattern at the lower level, and CTS is implemented to propagate traffic and collect real-time traffic information. The performance of the HSA is investigated in numerical applications in a sample network. These applications compare the efficiency and quality of the global optima achieved by Elitist GA and Micro GA. Furthermore, the impact of different frequencies of updating information and different population sizes of GA on system performance is analyzed.     

Genetic Algorithm-Simulation Methodology for Pavement Maintenance Scheduling

Abstract:   Pavement maintenance activities often involve lane closures, leading to traffic congestion and causing increases in road users' travel times. Scheduling of such activities should minimize the increases in travel times to all the travelers at network level. This article presents a hybrid methodology for scheduling of pavement maintenance activities involving lane closure in a network consisting of freeways and arterials, using genetic algorithm (GA) as an optimization technique, coupled with a traffic-simulation model to estimate the total travel time of road users in the road network. The application of this scheduling method is demonstrated through a hypothetical problem consisting of assigning three maintenance teams to handle 10 job requests in a network in 1 day. After 10 generations of genetic evolution with a population size of four, the hybrid GA-simulation model recommended a schedule that reduced the network total travel time by 5.1%, compared to the initial solution.     

The Optimization of Signal Settings on a Signalized Roundabout Using the Cross-entropy Method

Abstract:   The cross-entropy method (CEM) is a newly developed approach to the solution of complex combinatorial optimization problems. It is an iterative process that consists of generating solutions from some probability distribution whose parameter values are updated in each iteration using information from the best solutions found in that iteration. The article applies the method to the problem of the optimization of signal settings on a signalized roundabout. The performance of any given set of timings is evaluated using the cell transmission model, a deterministic macroscopic traffic flow model that permits the modeling of the spatial extent of queues and the possibility of "blocking back." The results from the investigations are encouraging, and show that the CEM has the potential to be a useful technique for tackling global optimization problems.     

Hybrid Meta-Heuristic Algorithm for the Simultaneous Optimization of the O–D Trip Matrix Estimation

Abstract:   In the present article, the origin–destination (O–D) trip matrix estimation is formulated as a simultaneous optimization problem and is resolved by employing three different meta-heuristic optimization algorithms. These include a genetic algorithm (GA), a simulated annealing (SA) algorithm, and a hybrid algorithm (GASA) based on the combination of GA and SA. The computational performance of the three algorithms is evaluated and compared by implementing them on a realistic urban road network. The results of the simulation tests demonstrate that SA and GASA produce a more accurate final solution than GA, whereas GASA shows a superior convergence rate, that is, faster improvement from the initial solution, in comparison to SA and GA. In addition, GASA produces a final solution that is more robust and less dependent on the initial demand pattern, in comparison to that obtained from a greedy search algorithm.     

A Neuro-Fuzzy Logic for ATIS Stand-Alone Control Systems: Structure, Calibration, and Analyses

Abstract:   An Advanced Traveler Information System (ATIS) stand-alone operation scheme is formulated as a bi-level optimization problem. The scheme logic attempts to optimize the network overall travel time by adjusting the path proportions while guessing the signal phase split decisions. An approximate simulation-based optimization algorithm is devised as an example of the logic operating this scheme. The logic is then replicated by a fuzzy-logic control system. Neural nets are utilized to develop the knowledge base of the fuzzy system and to calibrate the fuzzy set parameters. The neural nets utilize data replicates generated by the approximate simulation-based optimization algorithm. The calibration and effectiveness results of the fuzzy control system are presented.     

Influence of Demand Uncertainty and Correlations on Traffic Predictions and Decisions

Abstract: Decisions to improve a regional transportation network are often based on predictions of future link flows that assume future travel demand is a deterministic matrix. Despite broad awareness of the uncertainties inherent in forecasts, rarely are uncertainties considered explicitly within the methodological framework due at least in part to a lack of knowledge as to how uncertainties affect the optimality of decisions. This article seeks to address this issue by presenting a new method for evaluating future travel demand uncertainty and finding an efficient technique for generating multiple realizations of demand. The proposed method employs Hypersphere Decomposition, Cholesky Decomposition, and user equilibrium traffic assignment. Numerical results suggest that neglecting correlations between the future demands of travel zone pairs can lead to improvement decisions that are less robust and could frequently rank improvements improperly. Of the six sampling techniques employed, Antithetic sampling generated travel demand realizations with the least relative bias and error.     

Surrogate‐based toll optimization in a large‐scale heterogeneously congested network

Toll optimization in a large‐scale dynamic traffic network is typically characterized by an expensive‐to‐evaluate objective function. In this paper, we propose two toll‐level problems (TLPs) integrated with a large‐scale simulation‐based dynamic traffic assignment model of Melbourne, Australia. The first TLP aims to control the pricing zone (PZ) through a time‐varying joint distance and delay toll such that the network fundamental diagram (NFD) of the PZ does not enter the congested regime. The second TLP is built upon the first TLP by further considering the minimization of the heterogeneity of congestion distribution in the PZ. To solve the two TLPs, a computationally efficient surrogate‐based optimization method, that is, regressing kriging with expected improvement sampling, is applied to approximate the simulation input–output mapping, which can balance well between local exploitation and global exploration. Results show that the two optimal TLP solutions reduce the average travel time in the PZ (entire network) by 29.5% (1.4%) and 21.6% (2.5%), respectively. Reducing the heterogeneity of congestion distribution achieves higher network flows in the PZ and a lower average travel time or a larger total travel time saving in the entire network.     

An optimization‐based traffic diversion model during construction closures

During major highway construction, when lanes or entire highway sections must be temporarily closed, traffic managers would like to inform motorists of alternative routes around the construction site well in advance of the project location. This study develops a traffic diversion model to propose an optimum alternate route to drivers during a construction activity. The models and algorithms developed in this study assess a potential diversion route to optimize network performance while considering the drivers’ behaviors in following the proposed alternate route during a closure. A bilevel optimization model is proposed to minimize the total travel time of the affected network considering the link closure and a proposed alternate route for the travelers. A travelers’ route choice decision is modeled based on the user equilibrium traffic assignment, whereas a certain percentage of drivers are assumed to divert to the recommended alternate route. A sufficiently large subnetwork is selected, and a path selection method is proposed to reduce the computational effort required to optimize the model. A set of simulation experiments is conducted using the Tarrant County network in north Texas. The results show the ability of the model to improve the overall network performance during hypothetical closure scenarios.     

Numerical experience with diagonalization algorithms for asymmetric demand traffic assignment

This paper reviews the concept of a diagonalization algorithm for use in solving traffic network equilibrium problems for which the arc cost and/or the origin-destination travel demand functions are asymmetric. Such functions are known to occur in realistic settings involving multiple modes or users. The computational performance of this algorithm for different degrees of travel demand asymmetry is then explored by a detailed numerical experiment since no previous results of this type have been reported. It is found that, through the use of progressive stopping tolerances, the impact of high degrees of travel demand function asymmetry on the computational burden associated with finding a traffic network equilibrium may be mitigated; in effect equilibrium problems with high degrees of demand asymmetry are little more difficult to solve than perfectly symmetric problems.     

A Bi‐Level Model for Planning Signalized and Uninterrupted Flow Intersections in an Evacuation Network

Abstract: The problem to be addressed in this paper is the lack of an advanced model in the literature to locate the optimal set of intersections in the evacuation network for implementing uninterrupted flow and signal control strategies, respectively, which can yield the maximum evacuation operational efficiency and the best use of available budgets. An optimization model, proposed in response to such needs, contributes to addressing the following critical questions that have long challenged transportation authorities during emergency planning, namely: given the topology of an evacuation network, evacuation demand distribution, and a limited budget, (1) how many intersections should be implemented with the signals and uninterrupted flow strategies; (2) what are their most appropriate locations; and (3) how should turning restriction plans be properly designed for those uninterrupted flow intersections? The proposed model features a bi‐level framework. The upper level determines the best locations for uninterrupted flow and signalized intersections as well as the corresponding turning restriction plans by minimizing the total evacuation time, while the lower level handles routing assignments of evacuation traffic based on the stochastic user equilibrium (SUE) principle. The proposed model is solved by a genetic algorithm (GA) ‐based heuristic. Extensive analyses under various evacuation demand and budget levels have indicated that the location selection of uninterrupted flow and signalized intersections plays a key role in emergency traffic management. The proposed model substantially outperforms existing practices in prioritizing limited resources to the most appropriate control points by significantly reducing the total evacuation time (up to 39%).     

Integrated Planning of Supply Chain Networks and Multimodal Transportation Infrastructure Expansion: Model Development and Application to the Biofuel Industry

Abstract: As the biofuel industry continues to expand, the construction of new biorefinery facilities induces a huge amount of biomass feedstock shipment from supply points to the refineries and biofuel shipment to the consumption locations, which increases traffic demand in the transportation network and contributes to additional congestion (especially in the neighborhood of the refineries). Hence, it is beneficial to form public‐private partnerships to simultaneously consider transportation network expansion and biofuel supply chain design to mitigate congestion. This article presents an integrated mathematical model for biofuel supply chain design where the near‐optimum number and location of biorefinery facilities, the near‐optimal routing of biomass and biofuel shipments, and possible highway/railroad capacity expansion are determined. The objective is to minimize the total cost for biorefinery construction, transportation infrastructure expansion, and transportation delay (for both biomass/biofuel shipment and public travel) under congestion. A genetic algorithm framework (with embedded Lagrangian relaxation and traffic assignment algorithms) is developed to solve the optimization model, and an empirical case study for the state of Illinois is conducted with realistic biofuel production data. The computational results show that the proposed solution approach is able to solve the problem efficiently. Various managerial insights are also drawn. It shall be noted that although this article focuses on the booming biofuel industry, the model and solution techniques are suitable for a number of application contexts that simultaneously involve network traffic equilibrium, infrastructure expansion, and facility location choices (which determine the origin/destination of multi‐commodity flow).     

Optimal design of double-layer barrel vaults using genetic and pattern search algorithms and optimized neural network as surrogate model

This paper presents a combined method based on optimized neural networks and optimization algorithms to solve structural optimization problems. The main idea is to utilize an optimized artificial neural network (OANN) as a surrogate model to reduce the number of computations for structural analysis. First, the OANN is trained appropriately. Subsequently, the main optimization problem is solved using the OANN and a population-based algorithm. The algorithms considered in this step are the arithmetic optimization algorithm (AOA) and genetic algorithm (GA). Finally, the abovementioned problem is solved using the optimal point obtained from the previous step and the pattern search (PS) algorithm. To evaluate the performance of the proposed method, two numerical examples are considered. In the first example, the performance of two algorithms, OANN + AOA + PS and OANN + GA + PS, is investigated. Using the GA reduces the elapsed time by approximately 50% compared with using the AOA. Results show that both the OANN + GA + PS and OANN + AOA + PS algorithms perform well in solving structural optimization problems and achieve the same optimal design. However, the OANN + GA + PS algorithm requires significantly fewer function evaluations to achieve the same accuracy as the OANN + AOA + PS algorithm.     

Robust Transportation Network Design Under Demand Uncertainty

Abstract:   This article addresses the problem of a traffic network design problem (NDP) under demand uncertainty. The origin–destination trip matrices are taken as random variables with known probability distributions. Instead of finding optimal network design solutions for a given future scenario, we are concerned with solutions that are in some sense "good" for a variety of demand realizations. We introduce a definition of robustness accounting for the planner's required degree of robustness. We propose a formulation of the robust network design problem (RNDP) and develop a methodology based on genetic algorithm (GA) to solve the RNDP. The proposed model generates globally near-optimal network design solutions, f, based on the planner's input for robustness. The study makes two important contributions to the network design literature. First, robust network design solutions are significantly different from the deterministic NDPs and not accounting for them could potentially underestimate the network-wide impacts. Second, systematic evaluation of the performance of the model and solution algorithm is conducted on different test networks and budget levels to explore the efficacy of this approach. The results highlight the importance of accounting for robustness in transportation planning and the proposed approach is capable of producing high-quality solutions.     

End‐to‐End Deep Learning Methodology for Real‐Time Traffic Network Management

This article presents a novel real‐time traffic network management system using an end‐to‐end deep learning (E2EDL) methodology. A computational learning model is trained, which allows the system to identify the time‐varying traffic congestion pattern in the network, and recommend integrated traffic management schemes to reduce this congestion. The proposed model structure captures the temporal and spatial congestion pattern correlations exhibited in the network, and associates these patterns with efficient traffic management schemes. The E2EDL traffic management system is trained using a laboratory‐generated data set consisting of pairings of prevailing traffic network conditions and efficient traffic management schemes designed to cope with these conditions. The system is applied for the US‐75 corridor in Dallas, Texas. Several experiments are conducted to examine the system performance under different traffic operational conditions. The results show that the E2EDL system achieves travel time savings comparable to those recorded for an optimization‐based traffic management system.     

基于IPSO-DBN 的管道故障诊断方法

针对城市燃气管道故障诊断效果不佳的问题，提出了一种基于改进粒子群算法优化深度信念网络（IPSO-DBN）的管道故障诊断方法。该方法首先对粒子群算法（PSO）中的惯性权重ω、加速因子C1 和C2 进行修正，得到改进粒子群优化算法（IPSO），并采用两种基准函数对比测试PSO 与IPSO 的网络性能，证明所选改进方法的优越性。其次利用IPSO 优化深度信念网络（DBN）的初始权重，建立合适的DBN 网络，将4 种不同燃气管道工况下的实验数据用于IPSO- DBN 网络训练及预测。最后将实验所得的故障诊断准确率与BP、DBN、PSO-DBN 方法进行对比分析。实验结果表明，对于燃气管道不同工况下的故障分类识别，IPSO- DBN 方法的平均测试集诊断准确率高达94.5%，诊断效果优于传统的BP、DBN 以及PSO-DBN 方法。     

Sensitivity analysis for a continuum traffic equilibrium problem

This paper develops a sensitivity analysis for the continuum traffic equilibrium problem of a city with several competing facilities. In the city, the customers’ origins are continuously dispersed. We assume that the customer demand is dependent on the total cost of patronizing these facilities. Specific travel cost–flow relationships are considered. The choice of facility in the continuum transportation system follows a user equilibrium principle in which from each origin, no customer can reduce their individual cost to patronize any of the facilities by unilaterally changing route or facility. The problem can be formulated as a minimization problem that is subject to a set of constraints and solved with a finite element method. The sensitivity analysis is based on the implicit function theorem at the equilibrium solution. A numerical example is presented to illustrate the applications of these sensitivity analysis results.     

沪南公路现状拥堵分析及改善对策

对上海沪南公路区域路网,交通流量,交叉口和出入口影响以及现状交通构成进行拥堵分析。在拥堵原因分析基础上,提出了完善区域路网;取消支路交叉口信控,采用右进右出交通组织;增加车道规模,局部路段采用"主路+辅路"的交通组织形式等改善对策。最终指出解决沪南公路交通拥堵问题的最佳策略是增加车道规模,局部路段采用"主路+辅路"的交通组织形式,可满足周围居民出行需求。     

Time‐Varying Lane‐Based Capacity Reversibility for Traffic Management

Abstract: This article presents a new bi‐level formulation for time‐varying lane‐based capacity reversibility problem for traffic management. The problem is formulated as a bi‐level program where the lower level is the cell‐transmission‐based user‐optimal dynamic traffic assignment (UODTA). Due to its Non‐deterministic Polynomial‐time hard (NP‐hard) complexity, the genetic algorithm (GA) with the simulation‐based UODTA is adopted to solve multiorigin multidestination problems. Four GA variations are proposed. GA1 is a simple GA. GA2, GA3, and GA4 with a jam‐density factor parameter (JDF) employ time‐dependent congestion measures in their decoding procedures. The four algorithms are empirically tested on a grid network and compared based on solution quality, convergence speed, and central processing unit (CPU) time. GA3 with JDF of 0.6 appears best on the three criteria. On the Sioux Falls network, GA3 with JDF of 0.7 performs best. The GA with the appropriate inclusion of problem‐specific knowledge and parameter calibration indeed provides excellent results when compared with the simple GA.