Demand Responsive Transport Services Research Articles

A multi-task deep learning framework for forecasting sparse demand of demand responsive transit

Demand Responsive Transport (DRT) is a flexible transportation service that adjusts its routes and schedules in real time based on user requests. Real-time demand forecasting has become a timely issue since the resolution of transportation planning has increased to provide flexible services to users. Although many previous studies have been conducted to design the demand forecasting model using the mobility service historical data, there are more considerations when predicting real-time demand for on-demand mobility services. Therefore, this study aims to develop a demand forecasting model for DRT services addressing two issues. First, repeated calls for service and abandoned requests are converted into the unit of demand to capture served and potential demand. Second, given the sparsity of DRT historical data, a multi-task deep learning model is proposed to predict both the probability and volume of demand simultaneously to mitigate the data sparsity problem. A zero-inflated loss function was introduced to consider the skewed demand distribution. To attain our goals, the historical dataset of the I-MOD service trial in Yeongjongdo, Inchon, Korea was used. The performance of the proposed model is compared against several baseline models, demonstrating its superiority in forecasting demand for mobility services. The results show that a multi-task structure can alleviate the data sparsity problem by excluding where demand is not generated from the forecasting problem. The proposed models can provide insights into real-time demand forecasting with sparse demand.

Sustainability Trade-Offs in Climate Change Geographies in England

The evidence that climate change is the result of human actions is becoming increasingly stronger, as is the need to take action to limit the worst effects of climate change on the planet. However, politicians continue to equivocate and fail to address the trade-offs which are needed to deliver effective action. In this paper, we report on the potential of bottom-up approaches to transport planning to address the trade-offs between the need to reduce car-based travel and the social consequences of poor mobility options in rural areas. Using the theories of Sustainable Communities and Communities of Practice, we analyse the implementation of the Robin demand-responsive transport service in the West of England, presenting new data relating to the effectiveness of this service in providing low-carbon transport alternatives to rural residents. We find that the Robin is indeed effective, and that it has worked better in one location, where engagement with potential new users of the service has been prioritised. We conclude that such bottom-up transport options can be transformative, subject to the support of key stakeholders and integration with top-down systems of governance.

En-route transfer-based dynamic ridesharing

In the last few decades, the continuous growth of travel demand increased urban traffic congestion. The situation demands an efficient way of handling resources like roads and vehicles. Parallelly, the advancement of technology proposed more innovative forms of shared mobility, such as flexible pods. These autonomous, modular vehicles possess the unique capability to autonomously merge and split during their journeys, facilitating en-route transfer of passengers. Formulating a precise mathematical model that comprehensively captures the intricacies of such a complex system is quite challenging. This paper introduces an alternative approach rooted in Reinforcement Learning that circumvents the need for a predetermined mathematical model. Instead, it enables the system to learn its dynamics through interactions with the environment. The primary goal of this research is to acquire an optimized passenger transfer policy. The experimental findings indicate that the transfer-based pod service outperforms private taxi services and Demand Responsive Transport services, boosting vehicle utility by 18% and 8%, respectively. Remarkably, the pod service achieves these gains with only one-third of the vehicles required for private taxis and 13% fewer vehicles than the Demand Responsive Transport service. However, it's important to note that the transfer-based pod service does result in a modest increase in passenger travel time, ranging from 2% to 13%. Therefore, the transfer-based pod service offers a promising opportunity to alleviate traffic congestion while accepting a compromise on passenger travel times.

A phase-based perspective on urban demand responsive transport: A case study of Viavan pilot in Helsinki Capital Region

Despite decade-long efforts, real-time demand-responsive transport (DRT) services are still struggling to be a part of sustainable urban mobility systems. Among various approaches to understand the changing nature of DRT service, a longitudinal perspective has been found very effective. This study presents the case of an urban DRT pilot operated by Viavan in the Helsinki Capital Region, Finland. Research develops and implements a framework for longitudinal analysis, while drawing from the multi-level perspective (MLP) on sustainability transitions. DRT trip data is used to understand the service trajectory longitudinally over specific phases, delineated through key changes in service design. In addition, the framework uses three-level analysis of service parameters, to provide in-depth analysis in studying inter-phase and intra-phase changes. Case study approach also provides a comprehensive service area background and service implementation context to support the analysis. Results shows the complementary role of the urban DRT to fixed public transport services and points out certain challenges for efficient coordination. In addition, the analysis points towards a set of regime factors that should be considered for further understanding of the service trajectory. In practice, special emphasis should be paid on developing the mobility ecosystem and managerial practices. Future DRT case studies should develop additional frameworks, indicators and methods by drawing from sustainability transitions theory, to further understand service trajectories over time.

Understanding the behavioral intention of the rural population to use demand-responsive transport services

Global efforts to reduce carbon emissions include endeavors to raise the use of public transport means. In rural areas, this poses a distinctive challenge. Demand-responsive transport (DRT) services are a form of public transport that is regarded as a potential contribution to the required mobility behavior change. They mainly utilize small floor buses and require a pre-booking of each trip (via internet, app or telephone). Despite the growing number of active DRT services and trial operations, occupation numbers remain small. To understand this gap, this study investigates latent constructs that influence the behavioral intention of inhabitants living in rural regions to use DRT services. For this purpose, a questionnaire based on the Unified Theory of Acceptance and Use of Technology was adapted, tested independently through an exploratory factor analysis and distributed in Austria. The sample (n = 186) was subjected to a structural equation model. Results show that performance expectancy and attitude towards public transport have a significant impact on the behavioral intention of inhabitants to use DRT services in rural regions, whereas the attitude towards private cars was insignificant. Inter alia, we suggest concentrating on the determination of people’s specific everyday mobility needs in rural areas, analyzing and optimizing the public transport offer as an interrelated system, and communicating its benefits more efficiently.

Demand-responsive transport for students in rural areas: A case study in Vulkaneifel, Germany

In rural areas with low population density, demand-responsive transport (DRT) is considered a promising alternative to conventional public transport (PT). With a fleet of smaller vehicles, DRT provides a much more flexible and convenient service. This characteristic makes DRT also a potential mode of transport to serve students in rural areas. If DRT vehicles are used to serve students, then the funding for conventional school buses (or adapted public transport schedules) can be reinvested in the DRT system. This may help to relieve the financial burden experienced by DRT operators and enable the operation of a large-scale DRT service in rural areas. In this study, a demand model for school commutes based on real-world, open-source data for Landkreis Vulkaneifel, a rural region in Germany, is built. Then a feasibility study is carried out using an agent-based transport simulation framework. In the feasibility study, various setups and operational schemes are explored, which are followed by a systematic cost analysis. Based on a conservative estimation, an annual budget of around 1600 Euro per student will be needed to maintain and operate a fleet of DRT vehicles that can transport all the students in the region from home to school on time in the morning. During the remaining time of the day and on school holidays, the vehicles can be used for conventional DRT service for the public.

Multi-day fair collaboration in demand-responsive transportation

In this paper, we consider the case of companies that offer a demand-responsive transportation service, such as a shared-taxi service, and are engaged in a horizontal collaboration initiative. The goal of the coalition is to optimize the transportation operations in such a way that no company is penalized, in terms of customers served and/or working time. We present an optimization model for a multi-day planning horizon that includes constraints aimed at guaranteeing a level of fairness to all companies that can be controlled over the planning horizon and day-by-day, if beneficial. An adaptive large neighborhood search heuristic is then presented for its solution. The computational experiments show that, although the model constraints the optimization space, it still guarantees substantial savings. Moreover, they show that the model is flexible and can guarantee the sustainability in the long term of the collaboration initiative.

The effect of data science on urban sustainability through the optimization of demand-responsive transportation

The study investigated the viability of implementing a demand-responsive transportation (DRT) system within the premises of Al-Baha University to facilitate the punctual arrival of students to their classes and ensure their convenient departure from the campus. This study aims to demonstrate and elucidate how data mining enhances the efficiency of DRT systems. The case study investigates the potential improvements in DRT operations through the utilization of data mining techniques. The research employs descriptive and prescriptive techniques for mining trip planning data. The database maintained by the Deanship of Admission and Registration et al.-Baha University in Al-Baha City, Saudi Arabia, has identified 50 designated locations for student pick-up. The research establishes a comprehensive database that classifies pick-up points as spatial vectors, incorporating longitude, latitude, and counts of students. Cluster analysis and multidimensional scaling techniques are employed to minimize travel times and enhance operational efficiency by consolidating pick-up points. A completely adaptable DRT system prioritizes user satisfaction over operational efficiency, facilitating superior flexible services with fewer buses. The University Transport Department can implement Fixed Route systems in regions with significant population density and substantial demand. Conversely, in areas with a lower population, where Fixed Route systems may be less efficient, the department can opt for DRT systems. Additionally, the findings demonstrate the potential cost-saving benefits of implementing a hybrid system that integrates the university and off-campus student housing. The results indicate that providing fully flexible DRT services to students’ harms efficiency and user experience. A comparative analysis between the Fixed route system and the DRT system reveals that the latter has the potential to offer enhanced service to students, utilizing an equivalent number of vehicles while reducing the number of rides required, all while maintaining a higher degree of flexibility in service provision. According to the recommendation of this study, it is advisable to implement a hybrid fixed route and DRT system. However, to assess the feasibility of such a system in a university setting in the Kingdom of Saudi Arabia (KSA), it is necessary to conduct a comprehensive viability analysis to identify any potential challenges or concerns.

Assessing spatial characteristics to predict DRT demand in rural Switzerland

The niche market segment of demand responsive transport (DRT) services is meant to overcome structural economic problems of currently cost ineffective public transport (PT) services in rural areas. Simulation studies for mainly urban DRT services showed that demand for DRT trips is correlated with spatial characteristics. More knowledge of spatial characteristics of rural settings and their influence on DRT trips is necessary.In this study, trip data of a rural DRT service called mybuxi is used. Machine learning is applied for a better understanding of spatial characteristics of DRT demand in two different rural settings of the mybuxi service. Here in, the transferability from one mybuxi setting to the other is then tested.Results show that the number of inhabitants is the most important spatial characteristic for the prediction of DRT demand, followed by the distance to a train station and the presence of a restaurant in a given zone. The quality indicator of PT had low or no predictive power. The study shows that both DRT service areas experienced an increase in accessibility. For future transport planning, the increase in accessibility by DRT services in different rural areas must be taken as a legitimation for these services to be implemented instead of line-bound PT services.

The Role of DRT in European Urban Public Transport Systems—A Comparison between Tampere, Braunschweig and Genoa

Demand-Responsive Transport (DRT) is one of the most valid solutions to tackle the problems affecting public transport today, both in urban and rural areas. Despite its undoubted advantages, it still remains underdeveloped compared to its great potential. The purpose of this paper is to understand the role that on-demand transport plays in the strategic choices of public transport authorities (PTAs): to this end, this study examined the DRT services of three geographically distant European cities, in order to test different social, cultural and regulatory backgrounds, examining their main characteristics. Tampere, Braunschweig and Genoa were selected for the purposes of this work; data and information were collected by viewing the official websites of public transport companies and by direct contact via mail/telephone with the managers responsible for on-call transport. The data collected were then analyzed based on specific Key Performance Indicators (KPIs) identified in academic literature. The results of this paper show that the role of on-call service in the strategic decisions of PTAs depends on the cultural context of reference; some cities focus more on urban services, others on rural transport. In all three case studies examined, on-demand transport is an important aspect of local mobility and with wide room for growth.

Comparing passengers’ satisfaction with fixed-route and demand-responsive transport services: Empirical evidence from public light bus services in Hong Kong

Conventional public transport services are usually highly regulated with fixed-route and fixed-schedule, which are considered less flexible and efficient to serve low-density and dispersed settled areas. Demand-responsive transport is possibly a more economically and operationally viable solution, which offers dynamically allocated routes and schedules and serve the widespread passenger demand. There have been numerous simulation studies comparing the effectiveness of these two types of transport services in different settings, but which type is more favorable to passengers and what service aspects play the key role in shaping the overall satisfaction with the service quality from the passengers’ point of view are largely unknown. Using fixed-route and demand-responsive public light buses in Hong Kong as a case study, this research conducted an on-street questionnaire survey with 547 samples regarding the passengers’ perception of the performance of ten service aspects and the overall performance. An ordered logit model was accordingly developed to identify the relative importance and contribution of these service aspects to the overall service quality. An importance-performance analysis was further carried out to visualize the ratings of importance and performance of the service aspects, and to prioritize the improvement areas. The findings reveal that the passengers were less satisfied with the demand-responsive public light bus services in general. Among all the service aspects, we find that the in-vehicle environment, waiting time, driver’s attitude, and safety hold a higher improvement priority. Policy implications for improving and regulating both fixed-route and demand-responsive public light bus services are suggested.

The journey of demand responsive transportation: Towards sustainable services

The concept of Demand Responsive Transport (DRT) describes a technology-enabled shared mobility service that has a flexible operating schedule and/or provides virtual stops and/or flexible routes. While these on-demand services are not new, the developments in communication and tracking technology (vehicle positioning systems) have revived these services in the past decade. The benefits of adopting demand responsive transport services for intra-community and FLM travel needs are widely accepted, but there is still a cautious approach towards their implementation due to the failure of many promising demand responsive transport schemes in the past. This article 1) creates an overview of the various on-demand services introduced across the world, to understand the factors that may have contributed to the failure of these services in the past 2) identifies the progress made towards sustainable demand responsive transport ventures through analysis of global case studies 3) provides an overview of the flexibility of vehicle and deployment technologies in the demand responsive transport sphere. A bibliometric analysis, where the top keywords were further categorised using VOSviewer’s default clustering algorithm, highlighted the importance of sustainability in demand responsive transport ventures. By the progress made towards sustainable demand responsive transport ventures, it can be concluded that environmentally sustainable demand responsive transport ventures can be achieved through the adoption of electric and autonomous vehicles for demand responsive transport services, by reducing mileage of the vehicle and/or adjusting the length of route. The study concludes by reviewing existing research gaps regarding performance expectation, and recommending policy and practice implications, based on the case study of the Bus-on-Demand in Dubai, UAE.

PGDRT: Prediction Demand Based on Graph Convolutional Network for Regional Demand-Responsive Transport

To provide an efficient demand-responsive transport (DRT) service, we established a model for predicting regional movement demand that reflects spatiotemporal characteristics. DRT facilitates the movement of restricted passengers. However, passengers with restrictions are highly dependent on transportation services, and there are large fluctuations in travel demand based on the region, time, and intermittent demand constraints. Without regional demand predictions, the gaps between the desired boarding times of passengers and the actual boarding times are significantly increased, resulting in inefficient transportation services with minimal movement and maximum costs. Therefore, it is necessary to establish a regional demand generation prediction model that reflects temporal features for efficient demand response service operations. In this study, a graph convolutional network model that performs demand prediction using spatial and temporal information was developed. The proposed model considers a region’s unique characteristics and the influence between regions through spatial information, such as the proximity between regions, convenience of transportation, and functional similarity. In addition, three types of temporal characteristics—adjacent visual characteristics, periodic characteristics, and representative characteristics—were defined to reflect past demand patterns. With the proposed demand forecasting model, measures can be taken, such as having empty vehicles move to areas where demand is expected or encouraging adjustment of the vehicle’s rest time to avoid congestion. Thus, fast and efficient transportation satisfying the movement demand of passengers with restrictions can be achieved, resulting in sustainable transportation.

Detailing DRT users in Europe over the last twenty years: a literature overview

Travel choices in terms of means of transport and frequencies have changed during the recent pandemic period due to mobility restrictions, the growing fear of contagion and, especially in some months, the reduction of public transport capacity during the phases of the pandemic (especially for low demand areas). These trends must be analysed in order to optimize the implementation of possible complementary solutions to fill the deficit of local public transport (TPL) by introducing for example the Demand Responsive Transport services (DRT). A preliminary analysis is useful to identify the most efficient, effective and sustainable solutions in the various contexts, taking into account users and their motivation to travel. A growing need for “on-demand” mobility is linked to the increase in the number of elderly and disabled people. With a lack of alternative services and a reluctance to bear the burden and cost of ownership of vehicles, transport infrastructure will be particularly important to this aging population. Therefore, the improvement of transport services must consider some main characteristics of this modal choice are: being user-oriented; guarantee the accessibility of the service via the web, on specific platforms available on fixed and mobile devices and also enjoy the versatility of use with reference to the areas and users to be served. The present work, therefore, focuses on an evaluation of the literature, defining the main characteristics of DRT in Europe over the last twenty years. The results lay the foundations for a better planning of the service in the post-pandemic phase and a diffusion of bottom-up approaches for the calibration of the service itself through the dissemination of survey campaigns.

Designing demand responsive transport services in small-sized cities using an agent-based model

This paper presents an agent-based model (ABM) to simulate and compare two different operation strategies of a public transport service in small-sized cities, namely a fixed-route transit (FRT) and a demand-responsive transport (DRT) service, under varying demand rates and supply configurations. The ABM builds upon a previous work by the Authors, where flexible and feeder services of a Mass Rapid Transit system were simulated. In this paper, instead of a many-to-one pattern typical of a feeder service, we considered a many-to-many one. The objective is to investigate the conditions that make a DRT more attractive than a FRT in small-sized cities and guide its design considering the demand fluctuation, land-use pattern, service constraints and passenger preferences. A dispatching algorithm for the DRT allows to assign each new trip request to a vehicle, and a couple of origin and destination stops, updating the vehicle schedule in real time. The service includes fixed and virtual stops, allowing request consolidation and balancing operator-related (cost of the service) and user-related (quality of service) needs. The model is applied to Vittoria (Italy), a small city with 60,000 residents in Southern Italy where most trips are made by car, also due to the absence of an urban public transport service. First results highlight the benefits of providing a flexible service compared to a fixed one to minimize detours, waiting times and walking distances experienced by passengers while allowing for a higher shareability and efficiency of the service.

Possible Congestion-Mitigating Carrot-and-Stick Transport Policies for Jerusalem

We employ calibrated and validated MATSim multi-modal model of the Jerusalem Metropolitan Area traffic for assessing the effectiveness of Demand-Responsive-Transport (DRT) with ridesharing as a possible game-changer of the existing equilibrium between public transport and private cars. We investigate the combined effect of congestion charges, parking prices, and shared DRT services on the modal split and demonstrate that the DRT is effective when introduced together with congestion charges or parking prices. The policy effectiveness depends on the choice of the deterrence mechanism – given the overall payment for entering the controlled area, the effect of congestion charges depends on the size of the area. At the same time, parking prices are equally effective for an area of any size and, thus, seem advantageous.

Demand responsive transport: New insights from peri-urban experiences

Despite great differentials between urban, peri-urban and rural experiences, little research on demand responsive transport (DRT) has been conducted in peri-urban settings. This paper explores the influence of peri-urban area (PUA) conditions on the design and the operation of DRT and the resulting demand patterns and modal shifts. It uses a combination of a trip dataset and a survey of users and non-users. Key findings suggest that the provision of public transport services within the PUA and to nearby towns and cities, the high car ownership rate and the low demand have a significant influence in designing the DRT scheme. An integrated DRT and public transport (PT) system has potentials to be successful in a context of PUA. Our empirical evidence indicates that such a DRT service design can improve the accessibility and generate a stable demand while reducing the risk of substituting PT. However, there is a danger of the unwanted substitution of walking and cycling. We provide recommendations for DRT planners and operators in developing future schemes in PUA settings.

Mobility support for disadvantaged and disabled travelers during pandemic or similar situations

The COVID-19 lockdown has reduced public transportation service to the disadvantaged and disabled people who urgently need adequate mobility to obtain essential suppliers. This paper aims to improve the life quality of people with disabilities and elderly people by addressing social exclusion, accessibility, and mobility issues. Demand responsive transport services are frequently offered in the context of door-to-door transportation of the elderly and persons with disabilities. We study and compare two frameworks. We apply both Sample average approximation (SAA) and Rolling Horizon (RH) to optimize a car sharing system for the total cost, including initiation cost and operation cost after fleet size is determined. The model is implemented with given geographic conditions and other local information to be tailored for specific applications for local communities. Given that no historical data is available, random sample data is generated to simulate expected demands. We consider three types of probability distributions for daily demand data, and the results generated using three different distributions are being examined and compared. The research shows that the demand data following a normal distribution results in the minimum total cost. Additionally, we study the impact of several factors on total cost, including demand fulfillment rates and operation hours. Our results suggest that the impact of fulfillment rate on fleet size is exponential after a threshold under all three types of daily demand data, and extended operation hours can significantly reduce the total cost. Finally, the paper provides applicable frameworks for city planners, NPOs, and policymakers to better allocate limited resources to implement the carsharing system when little to no historical travel information is available for low-density population areas. It is anticipated that the outcome from this research would benefit disadvantaged and disabled travelers during COVID-19 or similar difficult situations in the future.

Performance evaluation for demand responsive transport services: A two-stage bootstrap-DEA and ordinary least square approach

Demand responsive transport (DRT) services have often been introduced to provide flexible public transport services, especially for areas with dispersed demand. However, objectively evaluating the performance of DRT services is challenging. This study uses a two-stage bootstrapped DEA and Ordinary Least Square (OLS) approach to measure performance using the Taiwanese DRT services as the empirical setting. In the first stage, bias-corrected Data Envelopment Analysis (DEA) efficiency scores are calculated by employing the smoothed homogenous bootstrapped procedure (Simar & Wilson, 2000). In the second stage, a OLS regression model is used to explore the impact from external factors with bias-corrected DEA efficiency scores as the dependent variable. Model results confirm that more than half of DRT services have increasing return to scale. The model results also show heavy rail services bring significant negative impact to the DRT service performance. This finding implies the main users of DRT services are local residents with fewer intercity travellers who typically would use heavy rail for access. The model results can lead to an improvement strategy for DRT services, for example, having a strategy to attract tourists to use DRT services to access tourist attraction sites would be one way to increase demand which would have the added advantage of improving sustainability. This study evaluates how this might improve public transport network accessibility and concludes with recommendations of a way forward to provide more sustainable DRT services.

Perceived accessibility of employment sites by jobseekers and the potential relevance of employer-subsidised demand responsive transport to enhance the commute

The constraints placed by the transport options available to job-seekers are key factors for the accessibility of employment locations and therefore social inclusion. The present paper investigates the importance of these constraints and the potential appeal of an employer-subsidised Demand Responsive Transport (DRT) service to job-seekers at risk of social exclusion. Mainly quantitative questionnaire data were obtained from a survey (n = 254) of jobseekers attending three ‘Jobcentre Plus’ government agency offices in Bristol (UK) during September 2017. The offices, which integrate the provision of social security benefits with support to secure work, were in inner-city, intermediate, and peripheral locations. Comparative spatial analysis was conducted both within and between the locations. The respondents emerged as having high public transport dependence for the commute, and transport-related perceived barriers emerged as second in importance only to ‘qualifications and skills’ and were reported as having inhibited attendance at job interviews and jobs. The preferences identified from the literature for finding work near home or in the city centre was confirmed. Reaching employment locations on the periphery of the city was particularly problematic. Job-seekers interviewed at the intermediate location reported the widest geographical scope of search. Logistic regression modelling confirmed the perceived options for public-transport commuters were somewhat different. Gender and the type of work sought also influenced spatial perceptions. Respondents were more willing to share the commute with ‘people they knew’, and strongly supported the concept of employer-subsidised DRT, with some statistically significant gender differences in attractiveness regarding the specific nature of the service offer. It is concluded that employer-subsidised DRT services would be most appropriate for remote sites, in situations in which the labour force is likely to be drawn from areas hard to connect with public transport, and where car use is either low, or being reduced by car use restraint policies. Future research into the context of real-world applications is required to examine whether benefits to employers, including staff recruitment and retention, would be sufficient to justify employer subsidies.