Transfer Penalty Research Articles

Beyond the last mile: different spatial strategies to integrate on-demand services into public transport in a simplified city

AbstractIntegrating on-demand services into public transport networks might be the best way to face the current situation in which these new technologies have increased congestion in most cities. When cooperating with on-demand services rather than competing with them, public transport would not risk losing users, and could attract some passengers from private modes thanks to an increased quality of service. This fact has engendered a growing literature discussing how to design such an integrated system. However, all of that research has imposed that on-demand mobility is to solve the so-called “last-mile problem”, serving only as a feeder that connects the exact origins/destinations with the traditional public transit network. As it induces a large number of transfers and it precludes some scale-effects to be triggered, in this paper we challenge that imposition and investigate if this is the best spatial integration strategy. To do so, we study a simplified linear city in a morning peak situation, where we propose seven different line structures, all of them combining a traditional fixed line with on-demand ride-pooling (ODRP): three direct structures, where ODRP can serve full trips, three semi-direct, where a single ODRP vehicle can serve the largest part of a trip, and a base case in which ODRP is restricted to the first and final legs only. Our results show that the base case is optimal only under very specific demand patterns, or when transfer penalties are disregarded. Our analytical approach reveals relevant operational aspects of such integrated systems: namely, that the base case can help increase directness (diminishing detours), and that ODRP can help shorten the routes of the fixed services to decrease operator costs.

Analysis of multimodal passenger terminal design and its effects on transfer penalty

With public transit systems becoming more and more sophisticated, the likelihood of passengers having to transfer between modes or lines increases. Transferring burdens, the passenger with added mental and physical effort, referred to as transfer penalty, which has a negative effect on the travel experience. The physical design of multimodal transfer terminals, such as train stations, can have a significant effect on transfer penalties but has received relatively little attention in practice. The aim of this review essay is to create an understanding of the ways in which the physical design of terminals can influence the transfer penalties experienced by passengers. Our findings show that terminal design can influence transfer penalties through actual and perceived walking and waiting times at transfer as well as the pure transfer penalty. We identified distinct categories of design elements and of other parts of the transit systems that urban designers can employ in influencing these aspects.

Solving fuzzy shortest path problem with vertex transfer penalties under type-2 fuzzy environment

Solving fuzzy shortest path problem with vertex transfer penalties under type-2 fuzzy environment

Public Transit Itinerary Choice Analysis Considering Various Incentives

Previous research on public transport user preferences has primarily focused on mode choice, transfer penalties, and preferences related to service qualities. Individuals are generally found to not like waiting, transfers, or overcrowding. However, from an operation perspective, it could be important to convince users to change to a different itinerary to reduce overcrowding, which can cause problems such as bunching. This would most likely require convincing the user to voluntarily accept a longer wait or to take a different route potentially with a transfer. One means of doing so might be through incentives or offering a shared taxi. Incentivizing transit users to switch to a less-crowded itinerary, which could potentially be a way to manage bus overcrowding, is not well-studied. Through a discrete choice experiment, this study thus filled this research gap by testing people’s willingness to change routes or to take a shared taxi in exchange for one of three distinct types of incentives. A mixed binary logit model and a mixed nested logit model were used to investigate how different factors influence public transit user itinerary choice behavior. The results suggested that some people would be willing to change to the proposed alternative. Heterogeneity within the responses was seen as transit users were found to have distinct preferences for different incentives, and the value of the incentives influenced users differently. Our models and results could be helpful for public transit agencies to improve efficiency and reduce costs by better balancing the travel demand.

Solving fuzzy shortest path problem with vertex transfer penalties under type-2 fuzzy environment

Solving fuzzy shortest path problem with vertex transfer penalties under type-2 fuzzy environment

Transit Network Sensitivity Analysis

Transit network conditions change everyday. While those changes should be considered for modifying a transit network, formulating and optimizing the whole transit network may be a costly, difficult task.This article uses a developed transit network design model to examine how optimal transit networks should be developed based on changes in input elements of the transit network. Three major inputs—demand, travel speed, and transfer penalty—are chosen for the sensitivity analysis. Different optimal transit networks and their characteristics are generated, and the relationship between inputs and outputs is discussed. Using the sensitivity analysis, three typical transit networks—transfer-oriented transit, transfer-avoidance transit, and directly-connected transit—are introduced. Optimal types of transit networks are suggested based on transit network situations.

Accelerating turbulent reacting flow simulations on many-core/GPUs using matrix-based kinetics

The present work assesses the impact - in terms of time to solution, throughput analysis, and hardware scalability - of transferring computationally intensive tasks, found in compressible reacting flow solvers, to the GPU. Attention is focused on outlining the workflow and data transfer penalties associated with “plugging in” a recently developed GPU-based chemistry library into (a) a purely CPU-based solver and (b) a GPU-based solver, where, except for the chemistry, all other variables are computed on the GPU. This comparison allows quantification of host-to-device (and vice versa) data transfer penalties on the overall solver speedup as a function of mesh and reaction mechanism size. To this end, a recently developed GPU-based chemistry library known as UMChemGPU is employed to treat the kinetics in the flow solver KARFS. UMChemGPU replaces conventional CPU-based Cantera routines using a matrix-based formulation. The impact of i) data transfer times, ii) chemistry acceleration, and iii) the hardware architecture is studied in detail in the context of GPU saturation limits. Hydrogen and dimethyl ether (DME) reaction mechanisms are used to assess the impact of the number of species/reactions on overall/chemistry-only speedup. It was found that offloading the source term computation to UMChemGPU results in up to 7X reduction in overall time to solution and four orders of magnitude faster source term computation compared to conventional CPU-based methods. Furthermore, the metrics for achieving maximum performance gain using GPU chemistry with an MPI + CUDA solver are explained using the Roofline model. Integrating the UMChemGPU with an MPI + OpenMP solver does not improve the overall performance due to the associated data copy time between the device (GPU) and host (CPU) memory spaces. The performance portability was demonstrated using three different GPU architectures, and the findings are expected to translate to a wide variety of high-performance codes in the combustion community.

Understanding the ground access and airport choice behavior of air passengers using transit payment transaction data

We investigate the ground access and airport choice behavior of individual air passengers that use rail transit in their ground transport to airports. Using a novel dataset of payment transaction records of rail transit trips, we calibrate a binomial logit model of individual's choice between two airports in Beijing, China to examine air passengers' preferences for ground rail service attributes including travel time, travel cost and number of transfers and compute the implied value of travel time and transfer penalty. We further visualize the spatial distribution of access and egress air ridership across the city at the rail transit station level with a set of heat maps and develop a gravity model to understand factors that affect the air ridership at individual stations. Our findings have significant implications for airport planner and transit authorities in assessing the potential benefit of airport rail link projects and improving the efficiency of regional airport systems.

Transfer Penalty Estimation of Bus Passengers Considering Travel Frequency in Chengdu, China

The paper aims to examine bus passengers’ route choice behavior in the context of transfer penalty estimation. Methodologically, it leverages passengers’ stop-level trip information inferred from automatic vehicle location (AVL) and automated fare collection (AFC) data and compares the stop-level origin–destination (O-D) pairs with direct paths and transfer paths. The route choice models were established with the inclusion of normalized travel frequency as a key variable. A case study was performed using the trip data of adult card holders from the bus systems in Chengdu, China. The results show that the normalized travel frequency is a variable that makes an informative contribution to the explanation of passenger route choices. Specifically, passengers with higher travel frequency prefer to choose the transfer path. Moreover, the transfer penalty function in the context of the monthly travel frequency was obtained and shows that the travel frequency can reduce a passenger’s transfer penalty. Finally, the stabilities of the normalized travel frequency function and transfer penalty function were assessed using the trip data from the weekdays in four weeks with the results showing that the normalized travel frequency function is similar in shape. In addition, the relationship between a passenger’s monthly travel frequency and their familiarity with the bus network was discussed using the second derivative of the normalized travel frequency function.

An international time equivalency of the pure transfer penalty in urban transit trips: Closing the gap

The perception of transfers in urban transit trips plays a key role when choosing an appropriate design of a public transport network for a given city, as there are lines structures that involve significantly more connections than others, e.g. hub-and-spoke or feeder-trunk. Besides additional walking and waiting, a transfer involves the interruption of a trip, whose value, called pure transfer penalty (PTP), has not received the same attention from a behavioral viewpoint. In this paper we contribute to find whether there is an equivalency with as general validity as walking and waiting regarding in-vehicle time. We do this by reviewing available evidence – all in relatively large cities - and by estimating the PTP in the very small Spanish city of Vitoria adapting a generic methodology that has been applied only to a metropolitan area (Madrid). Although Vitoria is much smaller in size, with fewer shares of multimodal trips, harder climate and shorter trip distances, results reveal that PTP is perceived as an increase of 18.4 Equivalent-In-Vehicle-Minutes (EIVM) when it does not rain or snow, very close to the values obtained in Madrid (15.2–17.7) and other reported meta-analysis (17.6). This figure drops when bad weather happens, which yields a weighted annual average of 13.9 EIVM. We propose 13–18 as a reasonable equivalence range for planning purposes.

Choices of intercity multimodal passenger travel modes

In recent years, transportation problems have evolved from city transportation problems to urban agglomeration problems. Thus, the development of Chinese transportation cannot neglect the role of Inter-City transportation being the infrastructural framework. Keep abreast of Inter-City passengers’ choice behaviors can support the operation of multi-transportation systems. First, to better address the large deviation between the impedance function and the travelers’ perceived cost, we construct the generalized cost function with seven terms, which are car time, waiting time, walking time, access/egress times, ticket price, transfer penalty and comfort level respectively, with reliability and security being taken into considerations. Second, pointing at the problem of low accuracy in calculating the sharing rate of various transportation modes under different traffic systems, we proposed a Two-Stage Path-Size Weibit (TS-PSW) model to calculate the sharing rate of multi-mode passenger flow between cities respectively. After that, we decomposed the TS-PSW model into two-layer Path-Size Weibit (PSW) model by the Two-Stage estimation method, and used the maximum likelihood estimation to solve PSW model respectively. Finally, using the intercity transportation networks (Chengdu–Chongqing, Guangzhou–Qingyuan, Beijing–Zhangjiakou) as background, compare six discrete choice models’ calculation results. Result shows that the passenger flow sharing rate curve calculated by TS-PSW Model is evidently the closest to the actual transportation modes, the maximum error from Chengdu to Chongqing is 6.60%, the minimum is 3.80%, the maximum error from Beijing to Zhangjiakou is 10.20%, and the minimum is 3.30%. Results verify the effectiveness of the proposed generalized cost function and mode choice model. The upcoming Winter Olympics will be held during the Spring Festival, other events will not be arranged at the same time. Therefore, travel demand will increase significantly. Improving the mode choice model and verifying its effectiveness on multiple intercity networks, especially the Beijing–Zhangjiakou network, can provide theoretical support for mastering passenger travel mode selection during the future Winter Olympics and further improving transportation planning and related deployment.

The Gatekeeper’s Dilemma: “When Should I Transfer This Customer?”

When Should I Transfer This Customer? “Please hold while I transfer you to next level of support.” Most of us have been on the receiving end of this message. In this study, the authors look at transfers from the service worker’s perspective. They create an online experiment in which participants play the role of call center agents who need to decide whether to transfer a virtual service request or continue attempting to resolve it. Consistent with compensation schemes common in call centers, participants receive a bonus for each successful resolution and may pay a penalty if they transfer. The authors find that these incentives generally work well; however, agents appear to overreact to transfer penalties by handling more requests than they should and transferring too few requests. Although this may be good news for customers who dislike being transferred, such behaviors may be costly for the call center; thus, managers need to be careful when rolling out complex compensation schemes.

Refined choice set generation and the investigation of multi-criteria transit route choice behavior

To best make informed public transportation decisions, agencies must be aware of how individuals interact with the transit system. The key question to consider is how do passengers get from Point A to Point B and which attributes of the transit trip have the largest influence on that choice? To more accurately understand passengers’ interaction with transit systems, a new method for transit route choice estimation is proposed. This method modifies and builds upon a schedule-based forward label setting multi-criteria shortest path algorithm by introducing an iterative trip elimination methodology. This new methodology yields high quality transit path choice sets with exact regeneration of 70.5% of the observed paths. Moreover, detailed temporal information on all types of network links (in-vehicle, walking, and waiting) are produced with reasonable computational time of 0.5 s per path calculation and 6.9 s per individual’s choice set generation. This increased specificity, in turn, heightens the validity and accuracy of the route choice model. Passenger information is sampled from a transit on-board survey containing origin–destination locations, demographic details, and trip-specific attributes. A multinomial logit model with stop-level path size correction term is estimated. This estimation yields a 67% recovery rate between the path with the highest estimated likelihood and the surveyed (taken) transit path. Furthermore, a transfer penalty of 28.8 min was estimated while the marginal rates of substitution for the waiting and walking time coefficients are in close alignment with similar values in the literature. Express bus routes were found to have a statistically significant negative impact on path utility for the lowest income thresholds while transitways (light rail, bus rapid transit, or commuter rail) had a positive associated perception for the highest household income class. Thus, support is found for the claim that transitways can potentially attract higher-income “choice” riders to the transit network. The results of this research can be used to improve ridership projections and highlight areas for policy improvements that could have the largest impact on retaining and attracting new passengers to the transit system.

Relevance of detailed transfer attributes in large-scale multimodal route choice models for metropolitan public transport passengers

Given the aim of increasing public transport patronage, it is important to understand how passengers perceive different trip characteristics. Most of the existing studies about public transport demand and route choice assigned a higher value of time to transfers than in-vehicle time and used a general transfer penalty to capture an average increase in the travel disutility because of the amount of transfers. However, it is likely that there are nuances to the transfer behaviour depending on specific transfer conditions that existing models do not capture and hence it is difficult to evaluate measures aimed at improving transfers to make public transport more attractive.

An Energy-Efficient Edge Computing Paradigm for Convolution-Based Image Upsampling

State-of-the-art deep learning solutions for image upsampling are currently trained using either resize or sub-pixel convolution to learn kernels that generate high fidelity images with minimal artifacts. However, performing inference with these learned convolution kernels requires memory-intensive feature map transformations that dominate time and energy costs in real-time applications. To alleviate this pressure on memory bandwidth, we propose a novel energy-efficient edge computing paradigm that confines the use of resize or sub-pixel convolution to training in the cloud by transforming learned convolution kernels to deconvolution kernels before deploying them for inference as a functionally equivalent deconvolution. These kernel transformations, intended as a one-time cost when shifting from training to inference, enable a systems designer to use each algorithm in their optimal context by preserving the image fidelity learned when training in the cloud while minimizing data transfer penalties during inference at the edge. We compare the inference properties of these convolution-based image upsampling algorithms and introduce a novel deconvolution inference algorithm, which we refer to as REVD2. To demonstrate the benefits of our approach, we upsample images selected from the BSD300 dataset using a pre-trained single-image super resolution network provided by the PyTorch model zoo. Using quantitative models of incurred time and energy costs to analyze this deep neural network, we estimate that using REVD2 for inference at the edge improves system latency by $2.1\times $ or $2.8\times $ and energy efficiency by $2.1\times $ or $2.7\times $ when respectively compared to sub-pixel or resize convolution counterparts.

Algorithms for Finding Shortest Paths in Networks with Vertex Transfer Penalties

In this paper we review many of the well-known algorithms for solving the shortest path problem in edge-weighted graphs. We then focus on a variant of this problem in which additional penalties are incurred at the vertices. These penalties can be used to model things like waiting times at road junctions and delays due to transfers in public transport. The usual way of handling such penalties is through graph expansion. As an alternative, we propose two variants of Dijkstra’s algorithm that operate on the original, unexpanded graph. Analyses are then presented to gauge the relative advantages and disadvantages of these methods. Asymptotically, compared to using Dijkstra’s algorithm on expanded graphs, our first variant is faster for very sparse graphs but slower with dense graphs. In contrast, the second variant features identical worst-case run times.

On the heat transfer characteristics of a Lamilloy cooling structure with curvatures with different pin fins configurations

PurposeThe pin fin is applied into a Lamilloy cooling structure which is broadly used in the leading edge region of the modern gas turbine vane. The purpose of this paper is to investigate effects of the layout, diameter and shape of pin fins on the flow structure and heat transfer characteristics in a newly improved Lamilloy structure at the leading edge region of a turbine vane.Design/methodology/approachA numerical method is applied to investigate effects of the layout, diameter and shape of pin fins on the flow structure and heat transfer characteristics in a newly improved Lamilloy structure at the leading edge of a turbine vane. The diverse locations of pin fins are Lp = 0.35, 0.5, 0.65. The diameter of the pin fins varies from 8 mm to 32 mm. Three different ratios of root to roof diameter for pin fins are also investigated, i.e. k = 0.5, 1, 2. The Reynolds number ranges from 10,000 and 50,000. Results of the flow structures, heat transfer on the target surface and pin fin surfaces, and friction factor are studied.FindingsThe heat transfer on the pin fin surface gradually decreases and then increases as the location of the pin fins increases. Increasing the diameter of the pin fins causes the heat transfer on the pin fin surface to gradually increase, while a lower value of the friction factor occurs. Besides, the heat transfer on the pin fin surface at a small root diameter increases remarkably, but a slight heat transfer penalty is found at the target surface. It is also found that both the Reynolds analogy performance and the thermal performance are increased compared to the baseline whose diameter and normalized location of pin fins are set as 16 and 0.5 mm, respectively.Social implicationsThe models provide a basic theoretical study to deal with nonuniformity of the temperature field for the turbine vane leading edge. The investigation also provides a better understanding of the heat transfer and flow characteristics in the leading edge region of a modern turbine vane.Originality/valueThis is a novel method to adopt pin fins into a Lamilloy cooling structure with curvature. It presents that the heat transfer of the pin fin surface in a pin-fin Lamilloy cooling structure with curvature can be significantly increased by changing the parameters of the pin fins which may lead to various flow behavior. In addition, the shape of the pin fin also shows great influence on the heat transfer and flow characteristics. However, the heat transfer of the target surface shows a small sensitivity to different layouts, diameter and shape of pin fin.

Analysis of the Passenger Flow Transfer Capacity of a Bus-Subway Transfer Hub in an Urban Multi-Mode Transportation Network

In the context of a rapid developing urban economy and the increasing number of motor vehicles, urban commuting transportation has witnessed a serious mismatch between the supply of and the demand for transportation network resources. In developing an urban multi-mode traffic network, using a urban traffic transfer hub to coordinate the transportation capabilities among different traffic networks is perceived to be highly effective for exploring the network transportation capacity of an entire transportation system, and improving travel efficiency and experiences for the public. Based on the super-network model, this paper develops a topological structure for a multi-mode traffic network, in which two typical combined travel modes are selected to establish the path impedance function for that network. Moreover, the multi-mode traffic allocation model and the solving algorithm are constructed in the research. The paper studies the impact of two types of factors related to the transfer capacity of passenger flows based on the regular traffic network of a bus-and-subway transfer hub using a sensitivity analysis of the transfer time and the associated penalty. The findings suggest that both changes in transfer walking time and the transfer penalty have no significant effect on the bus passenger flow.

Running the STEM Gauntlet: The Complicity of Four-Year Universities in the Transfer Penalty

Students who transfer between institutions of higher education often experience a transfer penalty or a decrease in the odds of degree completion. While the transfer pathway is deemed a responsibility shared between 2- and 4-year institutions, failures in the transfer process are typically attributed to community colleges. The aim of this study was to examine the post-transfer experiences of STEM community college transfer students with particular emphasis placed on how institutional policies and practices at 4-year institutions contributed to the transfer penalty and poorer attainment rates of this population. Implications for future research and interventions are discussed.

Improving Synchronization in an Air and High-Speed Rail Integration Service via Adjusting a Rail Timetable: A Real-World Case Study in China

Air and high-speed rail (AH) integration services are gaining ground with the development of the high-speed railway and airline industries. A well-designed feeder train timetable with good synchronization is of great significance in an AH integration service, because it can improve the connectivity at transfer nodes and offer more opportunities for intermodal passengers to travel. In this study, we propose a multi-objective model of a feeder railway timetable problem in an AH integration service to improve synchronization. The aims of the optimization model are to maximize the number of synchronizations and the coverage of synchronized flights, as well as to minimize the transfer penalties of passengers. We focus on a scenario of a partial subnetwork in which one direction of a two-direction railroad line with one transfer station is considered. The model is applied to Shijiazhuang Zhengding International Airport, China. The results illustrate the effectiveness of the approach developed in the paper.