Bus Travel Time Prediction Model Based on Profile Similarity.

In road-based mass transit systems, travel time is a key factor in providing quality of service. This article proposes a method of predicting travel time for this type of transport system. This method estimates travel time by taking into account its historical behaviour, represented by historical profiles, and the current behaviour recorded on the public transport vehicle for which the prediction is to be made. The model uses the k-medoids clustering algorithm to obtain historical travel time profiles. A relevant feature of the model is that it does not require recent travel time data from other vehicles. For this reason, the proposed model may be used in intercity transport contexts in which service planning is carried out according to timetables. The proposed model has been tested with two real cases of intercity public transport routes and from the results obtained we may conclude that, in general, the average error of the predictions is around 13% compared to the observed travel time values.

Applying Time-Dependent Attributes to Represent Demand in Road Mass Transit Systems.

The development of efficient mass transit systems that provide quality of service is a major challenge for modern societies. To meet this challenge, it is essential to understand user demand. This article proposes using new time-dependent attributes to represent demand, attributes that differ from those that have traditionally been used in the design and planning of this type of transit system. Data mining was used to obtain these new attributes; they were created using clustering techniques, and their quality evaluated with the Shannon entropy function and with neural networks. The methodology was implemented on an intercity public transport company and the results demonstrate that the attributes obtained offer a more precise understanding of demand and enable predictions to be made with acceptable precision.

System Proposal for Mass Transit Service Quality Control Based on GPS Data.

Quality is an essential aspect of public transport. In the case of regular public passenger transport by road, punctuality and regularity are criteria used to assess quality of service. Calculating metrics related to these criteria continuously over time and comprehensively across the entire transport network requires the handling of large amounts of data. This article describes a system for continuously and comprehensively monitoring punctuality and regularity. The system uses location data acquired continuously in the vehicles and automatically transferred for analysis. These data are processed intelligently by elements that are commonly used by transport operators: GPS-based tracking system, onboard computer and wireless networks for mobile data communications. The system was tested on a transport company, for which we measured the punctuality of one of the routes that it operates; the results are presented in this article.

Systematic Development of Intelligent Systems for Public Road Transport.

This paper presents an architecture model for the development of intelligent systems for public passenger transport by road. The main objective of our proposal is to provide a framework for the systematic development and deployment of telematics systems to improve various aspects of this type of transport, such as efficiency, accessibility and safety. The architecture model presented herein is based on international standards on intelligent transport system architectures, ubiquitous computing and service-oriented architecture for distributed systems. To illustrate the utility of the model, we also present a use case of a monitoring system for stops on a public passenger road transport network.

An Intelligent System Proposal for Improving the Safety and Accessibility of Public Transit by Highway.

The development of public transit systems that are accessible and safe for everyone, including people with special needs, is an objective that is justified from the civic and economic points of view. Unfortunately, public transit services are conceived for people who do not have reduced physical or cognitive abilities. In this paper, we present an intelligent public transit system by highway with the goal of facilitating access and improving the safety of public transit for persons with special needs. The system is deployed using components that are commonly available in transport infrastructure, e.g., sensors, mobile communications systems, and positioning systems. In addition, the system can operate in non-urban transport contexts, e.g., isolated rural areas, where the availability of basic infrastructure, such as electricity and communications infrastructures, is not always guaranteed. To construct the system, the principles and techniques of Ubiquitous Computing and Ambient Intelligence have been employed. To illustrate the utility of the system, two cases of services rendered by the system are described: the first case involves a surveillance system to guarantee accessibility at bus stops; the second case involves a route assistant for blind people.

Using massive vehicle positioning data to improve control and planning of public road transport.

This study describes a system for the automatic recording of positioning data for public transport vehicles used on roads. With the data provided by this system, transportation-regulatory authorities can control, verify and improve the routes that vehicles use, while also providing new data to improve the representation of the transportation network and providing new services in the context of intelligent metropolitan areas. The system is executed autonomously in the vehicles, by recording their massive positioning data and transferring them to remote data banks for subsequent processing. To illustrate the utility of the system, we present a case of application that consists of identifying the points at which vehicles stop systematically, which may be points of scheduled stops or points at which traffic signals or road topology force the vehicle to stop. This identification is performed using pattern recognition techniques. The system has been applied under real operating conditions, providing the results discussed in the present study.

Data mining methodology for obtaining epidemiological data in the context of road transport systems.

Millions of people use public transport systems daily, hence their interest for the epidemiology of respiratory infectious diseases, both from a scientific and a health control point of view. This article presents a methodology for obtaining epidemiological information on these types of diseases in the context of a public road transport system. This epidemiological information is based on an estimation of interactions with risk of infection between users of the public transport system. The methodology is novel in its aim since, to the best of our knowledge, there is no previous study in the context of epidemiology and public transport systems that addresses this challenge. The information is obtained by mining the data generated from trips made by transport users who use contactless cards as a means of payment. Data mining therefore underpins the methodology. One achievement of the methodology is that it is a comprehensive approach, since, starting from a formalisation of the problem based on epidemiological concepts and the transport activity itself, all the necessary steps to obtain the required epidemiological knowledge are described and implemented. This includes the estimation of data that are generally unknown in the context of public transport systems, but that are required to generate the desired results. The outcome is useful epidemiological data based on a complete and reliable description of all estimated potentially infectious interactions between users of the transport system. The methodology can be implemented using a variety of initial specifications: epidemiological, temporal, geographic, inter alia. Another feature of the methodology is that with the information it provides, epidemiological studies can be carried out involving a large number of people, producing large samples of interactions obtained over long periods of time, thereby making it possible to carry out comparative studies. Moreover, a real use case is described, in which the methodology is applied to a road transport system that annually moves around 20 million passengers, in a period that predates the COVID-19 pandemic. The results have made it possible to identify the group of users most exposed to infection, although they are not the largest group. Finally, it is estimated that the application of a seat allocation strategy that minimises the risk of infection reduces the risk by 50%.

Architecture of a framework for providing information services for public transport.

This paper presents OnRoute, a framework for developing and running ubiquitous software that provides information services to passengers of public transportation, including payment systems and on-route guidance services. To achieve a high level of interoperability, accessibility and context awareness, OnRoute uses the ubiquitous computing paradigm. To guarantee the quality of the software produced, the reliable software principles used in critical contexts, such as automotive systems, are also considered by the framework. The main components of its architecture (run-time, system services, software components and development discipline) and how they are deployed in the transportation network (stations and vehicles) are described in this paper. Finally, to illustrate the use of OnRoute, the development of a guidance service for travellers is explained.