Saeed Asadi Bagloee

A heuristic methodology to tackle the Braess Paradox detecting problem tailored for real road networks

Adding a new road to help traffic flow in a congested urban network may at first appear to be a good idea. The Braess Paradox (BP) says, adding new capacity may actually worsen traffic flow. BP does not only call for extra vigilance in expanding a network, it also highlights a question: Does BP exist in existing networks? Literature reveals that BP is rife in real world. This study proposes a methodology to find a set of roads in a real network, whose closure improve traffic flow. It is called the Braess Paradox Detection (BPD) problem. Literature proves that the BPD problem is highly intractable especially in real networks and no efficient method has been introduced. We developed a heuristic methodology based on a Genetic Algorithm to tackle BPD problem. First, a set of likely Braess-tainted roads is identified by simply testing their closure (one-by-one). Secondly, a seraph algorithm is devised to run over the Braess-tainted roads to find a combination whose closure improves traffic flow. In our methodology, the extent of road closure is limited to some certain level to preserve connectivity of the network. The efficiency and applicability of the methodology are demonstrated using the benchmark Hagstrom–Abrams network, and on a network of city of Winnipeg in Canada.

An efficient hybrid heuristic method for prioritising large transportation projects with interdependent activities

Transportation projects are generally large, with limited resources and highly interdependent activities. The complexities and interdependencies apparent in large transportation projects have prohibited effective application of management science and economics methods to these problems. We propose a heuristic method with several hybrid components. We formulate the problem as a Travelling Salesman Problem (TSP). A Neural Network (NN) is used to cope with the interdependency concerns. An algorithm with an iterative process is confined to search for the longest path (most benefit or most reduction in the user-time) in the NN as a solution to the TSP. The solution from each iteration step is utilised to update and train the NN and enhance its prediction. A search engine inspired by the concept of Ant Colony (AC) and hybridised with Genetic Algorithm (GA) is developed to find a suitable solution to the TSP. The hybrid heuristic method proposed in this study is applied to the real data for the city of Winnipeg in Canada to demonstrate the applicability of the proposed framework and exhibit the efficacy of the procedures and algorithms.

A Hybrid Branch-and-Bound and Benders Decomposition Algorithm for the Network Design Problem

Given a set of candidate road projects associated with costs, finding the best subset with respect to a limited budget is known as the network design problem NDP. The NDP is often cast in a bilevel programming problem which is known to be NP-hard. In this study, we tackle a special case of the NDP where the decision variables are integers. A variety of exact solutions has been proposed for the discrete NDP, but due to the combinatorial complexity, the literature has yet to address the problem for large-size networks, and accounting for the multimodal and multiclass traffic flows. To this end, the bilevel problem is solved by branch-and-bound. At each node of the search tree, a valid lower bound based on system optimal SO traffic flow is calculated. The SO traffic flow is formulated as a mixed integer, non-linear programming MINLP problem for which the Benders decomposition method is used. The algorithm is coded on a hybrid and synchronized platform consisting of MATLAB optimization engine, EMME 3 transport planning module, MS Access database, and MS Excel user interface. The proposed methodology is applied to three examples including Gaos network, the Sioux-Falls network, and a real size network representing the city of Winnipeg, Canada. Numerical tests on the network of Winnipeg at various budget levels have shown promising results.

A hybrid meta-heuristic algorithm for solving real-life transportation network design problems

The network-design problem (NDP) has a wide range of applications in transportation, telecommunications, and logistics. The idea is to efficiently design a network of links (roads, optical fibres, etc.) enabling the flow of commodities (drivers, data packets, etc.) to satisfy demand characteristics. Various exact and heuristic methods such as branch and bound, Tabu search, genetic algorithm (GA), ant system (AS) have been developed to address the NDP which is a highly intractable combinatorial problem. The literature has yet to address the NDP in real-size networks. In this study, we propose a new meta-heuristic algorithm for solving large NDPs by hybridising GA and AS methods. The applicability of the proposed meta-heuristic approach to real-size networks is demonstrated at two different sites. First, we use a large real-life problem for the city of Winnipeg, Canada and show that our heuristic method produces exact solutions very efficiently. Second, we evaluate the performance of the proposed algorithm using the data of Sioux Falls (a benchmark in the literature). While the proposed approach produces solutions similar to the other available methods in the literature, it is superior for developing solutions in large-size NDPs.

Methodology for Parking Modeling and Pricing in Traffic Impact Studies

The need to park a car is a key consideration in any trip. The need can affect or determine travel mode, departure time, and even the entire trip chain, as well as impose pressure on the network; drivers can create traffic jams as they seek places to park. Given the complexity of motorist parking behavior, no reliable methodology has been developed to model parking activities realistically. To date, the studies have focused on the effects of parking on travel demand (especially on mode choice). The study reported here developed a convenient and easy-to-use methodology, which integrated parking choice with the traffic assignment. The methodology responds in particular to the needs of practitioners who carry out traffic impact study projects in which a detailed analysis is sought of a confined area (study area). The proposed methodology splits a study areas zonal trips into two main parts: (a) walking trips to and from parking lots and (b) vehicular trips between origin parking lots and destination parking lots. A logit model was adapted to model parking choice, which could accommodate factors that influenced motorist behavior (e.g., a lots price, security, protection from the elements). The methodology was tested through its application to a part of the central business district of the city of Abu Dhabi, United Arab Emirates. A pragmatic approach to the problem of how to price parking to alleviate its shortage and to use its supply efficiently was proposed.

Heuristic Approach to Capacitated Traffic Assignment Problem for Large-Scale Transport Networks

Although capacity constraints in traffic assignment can represent many realistic features, these constraints are largely ignored in practice because of mathematical complexities in applying the methods proposed in the literature. In this study such complexities are relaxed by the adoption of an intuitive interpretation for the Lagrange values of the capacity constraints, that is, the amount of penalty added to the travel time of the oversaturated links to discharge the excessive flow to the extent to which they become saturated. This penalty term bears some similarity to the marginal cost of the system optimal. Hence the capacitated traffic assignment problem (TAP) becomes a normal uncapacitated TAP in which the aforementioned additional penalty is updated iteratively. The proposed provision is flexible to accommodate TAPs solution algorithms such as Frank–Wolfe. The main motivation of this study is to address the needs of the industry; hence, the proposed method is coded in a leading commercial transport planning software product, and a large-scale network of Winnipeg, Manitoba, Canada, is used for numerical evaluations. Furthermore the benchmark network of Hearn is also used for comparative evaluations with respect to other methods. Results suggest that in regard to the reliability of the outcomes and computational efficacy, the proposed algorithm is as good as other methods. Unlike other methods, there is no additional parameter to be calibrated, and the convergence behavior of the algorithm is promising.

A hybrid machine-learning and optimization method to solve bi-level problems

Bi-level optimization has widespread applications in many disciplines including management, economy, energy, and transportation. Because it is by nature a NP-hard problem, finding an efficient and reliable solution method tailored to large sized cases of specific types is of the highest importance. To this end, we develop a hybrid method based on machine-learning and optimization. For numerical tests, we set up a highly challenging case: a nonlinear discrete bi-level problem with equilibrium constraints in transportation science, known as the discrete network design problem. The hybrid method transforms the original problem to an integer linear programing problem based on a supervised learning technique and a tractable nonlinear problem. This methodology is tested using a real dataset in which the results are found to be highly promising. For the machine learning tasks we employ MATLAB and to solve the optimization problems, we use GAMS (with CPLEX solver).

A logit-based model for measuring the effects of transportation infrastructure on land value

ABSTRACT Mutual interactions between transportation and land use have long been debated. Despite progress made in computational technology, the study of these interactions is not adequately developed. The most important aspect of such interactions is given by the changes in land values due to changes in transportation infrastructures. We consider the behavioural features of these interactions along with the constraints on the land and/or zoning restrictions and propose a reliable model for the first time to predict land value changes with respect to changes in transportation facilities and accessibility. The proposed model is a logit-based mathematical programming methodology where the relative price of land is predicted with respect to transportation accessibility, neighbourhood amenities, location premium, availability of land, and zoning regulations. A real-world case study is used to exhibit the applicability of the proposed methodology and demonstrate the efficacy of the algorithms and procedures.

Transit priority lanes in the congested road networks

Given the advances in communication technologies and real-time traffic management, transit priority lanes are emerging as an indispensable component of intelligent transport systems. This scheme calls for giving priority to public transport. In this study, the question of interest is: Which roads can be nominated to give an exclusive lane to transit modes? Due to computational and theoretical complexities, the literature has yet to address this problem comprehensively at the network level considering various modes (public and private). Additionally, taking space away from private modes in favor of public transport may adversely affect the congestion level. To this end, inspired by the Braess Paradox, we seek mis-utilized space used by private modes to be dedicated to transit modes mainly on congested roads. To find such candidate roads, we define a merit index based on transit ridership and congestion level. The problem then becomes to find the best subset of these candidate roads to cede a lane to transit mode. It is formulated as a bilevel mixed-integer, nonlinear programming problem in which the decision variables are binary (1: to cause the respective road to have an exclusive transit lane or 0: not). The adverse effects are minimized on the upper level represented by total travel time (public and private modes) spent on the network. The lower level accounts for a bimodal traffic assignment, to consider the impact of transit priority on private modes. We then develop an efficient low-RAM-intensity branch and bound as a solution algorithm. The search for the subset is made in such a way that improved public transport is achieved at zero cost to the overall performance of the network. A real dataset from the city of Winnipeg, Canada is used for numerical evaluations.

A logit–based model for facility placement planning in supply chain management

The facility placement in supply chain management entails suppliers and consumers along with the terminals in between for distributing commodities. This study seeks to find the best terminal placement by taking into consideration the costs for both transportation and terminal construction. We call this a supplier-terminal-consumer (STC) problem and show that the STC is an NP-hard quadratic assignment problem. The NP-hard problems in real-size are proven to be intractable; hence, we develop a two-fold heuristic method for solving the STC problems. First, we identify the commodity flow by using a logit-based mathematical programming (Logit-MP) methodology based on the demand for the commodity and the locations of the candidate-terminals. We apply Logit-MP in an iterative process and specify the maximum utilisation of the candidate-terminals. Second, the best possible locations for the terminals are identified by analysing the utilisation rates in a geographic information system interface and using an interpolation method for converting the point-based utilisation rates into spatial data. We present numerical results of a large-size transportation case study for the city of Chicago where the commodity, terminals and consumers are interpreted as wheat, silos and bakeries, respectively.