Yuval Hadas

Public Transit Network Connectivity: Spatial-Based Performance Indicators

The service reliability of public transit operations is gaining increased attention as agencies face immediate problems in providing credible service and attempting to reduce operational cost. Unreliable service has been cited as the major deterrent element for existing and potential passengers. Because most public transit attributes are of a stochastic nature—travel time, dwell time, demand, and so forth—the passenger is likely to experience unplanned waiting and riding times. Missed transfers are major contributors to the reliability issues of any transit system; therefore, this work focuses on the connectivity of the transit network with the reliability and comfort of transfers in multilegged trips. Because of the physical and spatial characteristics of the transit routes, it is possible to categorize the transfers into three types: nonadjacent transfer points, adjacent transfer points, and shared transfer points. Each type of transfer contributes differently to the degree of connectivity of the transit system. On the basis of transit network of routes, spatial-based performance indicators are introduced and defined. That is followed by results of a case study of the transit networks of two Israeli midsize cities.

Improving Bus Passenger Transfers on Road Segments Through Online Operational Tactics

The use of transfers in public transit has the advantages of reducing operational costs and introducing more flexible, efficient route planning. The main drawback from the passengers’ point of view is the inconvenience of traveling multilegged trips. To diminish the waiting time caused by passenger transfers, synchronized timetables were introduced. The use of these timed transfers, however, creates uncertainty about the simultaneous arrival of two (or more) buses at an existing stop, which can lead to a deterioration in system reliability. To alleviate this uncertainty about simultaneous arrivals, a new passenger transfer concept was developed that extends the commonly used single-point encounter (at a single transit stop) to a road segment encounter (with any point along the road segment constituting a possible encounter point). The objectives of this work are as follows: (a) to define the bus-encounter probability along a road segment, (b) to introduce a simulation model to estimate the bus-encounter probability, (c) to model the bus-encounter probability upper bound, which is a major input to a dynamic programming model that optimizes the total travel time, and (d) to present simulation results that confirm the benefits of such a system. It is believed that the proposed concept will reduce the uncertainty of two buses meeting at a point, as well as reduce the average travel time and enable more flexibility in deploying online operational tactics (e.g., holding buses, skipping stops, slowing down).

Transfer Synchronization of Public Transport Networks

Transfers in public transport, especially in bus operations, are used to create a more efficient network by the reduction of operational costs and the allowance of more flexible route planning. However, because of the stochastic nature of traffic, scheduled transfers do not always occur; this situation increases the total passenger travel time and reduces the attractiveness of the public transport service. The use of selected operational tactics in public transport networks for increasing the actual occurrence of scheduled transfers was analyzed. A model was developed to determine the impact that instructing vehicles to either hold at or skip certain stops had on total passenger travel time and the number of simultaneous transfers. The model consisted of two components. First, a simulation of a public transport network examined the two tactics for maximizing the number of transfers. Second, an ILOG optimization model was used for optimal determination of the combination of the two tactics to achieve the maximum number of simultaneous transfers. A bus network was created as a case study, in Auckland, New Zealand, to verify the impact of the models application. Results showed that applying online operational tactics dramatically improved the frequency of simultaneous transfers by more than 100%. The concept has great potential for increasing the efficiency and attractiveness of public transport networks that involve scheduled transfers.

Public Transit Simulation Model for Optimal Synchronized Transfers

The use of transfers in public transit has the advantage of reducing operational costs and introducing more flexible and efficient route planning. In contrast, the main drawback for passengers is the inconvenience of traveling multilegged trips. To diminish the waiting time caused by transfers, synchronized (timed) timetables were introduced. Their use, however, suffers from uncertainty about the simultaneous arrival of two (or more) vehicles at an existing stop. This can lead to deterioration in system reliability. To alleviate the uncertainty of simultaneous arrivals, operational tactics can be deployed, such as hold, skip-stop, and short-turn, to increase the chances of simultaneous arrival of both vehicles at the same station. Each tactic has positive and negative effects on the total travel time. A dynamic programming model was developed for minimizing the total travel time, resulting in a set of preferred tactics to be deployed. An optimization model based on dynamic programming and a public transit simulation that validates the benefits of such a model are described. The results of the simulation confirm the benefits of the model with a 10% reduction of total travel time and more than a 200% increase of direct transfers (transfers in which both vehicles arrive simultaneously at the transfer point).

Multiagent Approach for Public Transit System Based on Flexible Routes

The use of transfers in public transit has the advantages of reducing operational costs and introducing more flexible and efficient route planning. In contrast, the main drawback, from the passenger point of view, is the inconvenience of traveling multilegged trips. To diminish the waiting time caused by transfers, synchronized (timed) timetables were introduced. Their use, however, suffers from uncertainty about the simultaneous arrival of two (or more) vehicles at an existing stop. This can lead to deterioration in system reliability. To alleviate the uncertainty of simultaneous arrivals, a new passenger transfer concept was developed that extends the commonly used single-point encounter (at a single transit stop) to a road-segment encounter (any point along the road segment constituting a possible encounter point). The new concept is based on a multiagent system incorporating new models: (a) simulation-based models that describe the encounter probability of public transit vehicles along a road segment under dependent travel time conditions, (b) an optimization model based on distributed dynamic programming, and (c) a simulation model for the validation of the model. This multi-agent concept incorporates a low-complexity, near-optimal model that reduces the average travel time by 7% and increases the number of direct transfers by 285%.

Modeling Bus Dwell Time with Decision Tree–Based Methods

A large proportion of transit travel time is made up by dwell time for passengers boarding and alighting. More accurate modeling and estimation of bus dwell time (BDT) can enhance the efficiency and reliability of the public transportation system. Multiple linear regression (MLR) has been the most commonly used method in the literature for modeling and estimating BDT. However, the underlying assumptions of the MLR method, such as multicollinearity and normality of random error, cannot always be satisfied for real applications. This study developed and implemented two methods based on decision trees (DTs), namely, classification and regression tree and chi-squared automatic interaction detector, for the first time for BDT modeling and estimation. The models were compared with the traditional MLR model after calibrating and validating the new models against the data collected from four bus stops in Auckland, New Zealand. Various error measurements were used to evaluate the accuracy of the models. The DT-based methods eliminated the limitations of the MLR method and provided reliable and accurate estimation of BDT.

Developing a model for the stochastic time-dependent vehicle-routing problem

Vehicle-routing problems (VRP) have been studied in depth. Many variants of the problem exist, most of them trying to find a set of routes with the shortest distance possible for a fleet of vehicles. This paper combines two important variants, the stochastic VRP and the time-dependent VRP, to form and define the Stochastic Time-Dependent VRP. An efficient heuristic that is a new variant of the well-known saving algorithm is introduced. The algorithm incorporates simulation that enables an estimate of each routes probability of being the quickest. This new algorithm yields fast results that are 10% higher than optimal solutions. Such results are similar to the performance of the saving algorithm when compared to the capacitated VRP.

Optimal Connected Urban Bus Network of Priority Lanes

This paper presents a new approach and modeling for selecting an optimal network of public transport (PT) priority lanes. Bus priority schemes and techniques on urban roads and highways have proved effective for almost half a century. Many bus priority studies have been published and demonstrated worldwide, but none has dealt with optimal connected networks of PT priority lanes. The approach used in this study was based on a system wide concept to obtain optimal PT network coverage. Such a PT priority lane network would enable fast and less interrupted vehicle movement, would increase the reliability of transfers, and would provide better adherence to schedule performance. The study developed a model for the optimal selection of a set of PT priority lanes that maximized the total travel time savings and, at the same time, maintained balanced origin and destination terminals, given a budget constraint. An efficient CPLEX model was developed and tested. The model was used in a case study of Petah Tikva, a midsize city in Israel, and produced a successful, optimal network of priority lanes.

Tool to Assess Regional Public Transport Plans for Integrated Systems

This study developed a tool that can be used by planners to assess the quality of their regional public transport plan and to produce a well-integrated system. The five attributes of integration—network integration, integrated timed transfer, integrated physical connection for transfers, information integration, and fare and ticketing integration—were adopted to develop a framework that illustrated the relationship between them. The tool was composed of the developed framework and a multicriterion decision-making approach; three steps were required to complete the assessment. In Step 1, each integration attribute was scored at an upper level and for each subattribute based on the information provided in the plan. Step 2 involved the analytical hierarchy approach, a multicriterion decision-making tool, to determine the weight of each attribute on the basis of its importance. The final step was to calculate the total weighted score of the plan. The novelty of the approach was the integration of a well-defined and easy-to-use framework with a systematic, hierarchical weighting model. Two case studies were undertaken to demonstrate the use of the tool: the regional plans of New Zealand’s two most metropolitan cities, Auckland and Wellington. The results demonstrate that the tool can be used to assess the quality of public transport plans for integration. The tool’s ability to indicate deficiencies can assist decision makers to improve their planning toward a better public transport system.

Caffeinated Chewing Gum as Countermeasure to Drivers’ Passive Task-Related Fatigue Caused by Monotonous Roadway

This study analyzed driver passive task-related fatigue caused by a monotonous environment and the effectiveness of caffeinated chewing gum as a countermeasure. Data collected by a driving simulator in the laboratory were used to measure changes in driving performance. A self-perceived measure of fatigue was also analyzed. Seventy-two subjects were asked to drive for 70 min along a straight road after receiving one of the following substances (treatments): caffeinated chewing gum, a cup of coffee, or placebo chewing gum. The 72 subjects were subdivided into three groups of 24 each, and all participants were asked to take part in two driving sessions: one control drive without administration (no treatment) and one with administration (one of the treatments). The negative effects on driving performance of prolonged driving and the effectiveness of the standard deviation of the lateral position in representing worsening driving performance were demonstrated. This analysis indicated that intake of caffeine in the form of caffeinated chewing gum (100 mg caffeine) improved driving performance in less than 10 min. Drinking an ordinary cup of coffee (with the same caffeine content) did not improve driving performance in the same short time interval.