The paper presents applied aspects of the design and development of an automated passenger transportation management system for any transport company. A flexible architecture of the transport system (TS) is proposed, which is built according to the microservice methodology and simplifies the synchronization processes between drivers and operators. Vehicle information is updated via GPS. The proposed design approaches allow to customize the TS to the individual needs and initiatives of customers and quickly expand functionality and add new features and services. In addition, the system can be used as an intermediate node for embedding in an existing system to collect information and provide a graphical user interface for operators.
Transport is one of the most important branches of social production and it is designed to meet the needs of the population in transportation. In Ukraine, most bus services are provided by small private enterprises. Their fleet of vehicles provides urban, long-distance and international transportation in Europe and the CIS. A feature of the work of private operators is the difficulty of control and the lack of a single system for monitoring vehicles on routes, which is often an obstacle to making operational decisions to optimize the operation of transport. In this regard, there is a need to create an automated passenger traffic management system with the following capabilities: 2 • Storage and management of information about vehicles and their technical characteristics; • Driver information management: current position during the trip, contact details, the vehicle to which they are assigned; • Ability to expand locations on the map and update them: location is a material point, which is added by the operator to establish a point of repair, parking, stopping, etc.; • Keeping records of reference data on emergency repair and synchronization of data from external resources; • Creating routes using external APIs, using software to store information in internal repositories and caching this data for the fastest response from the server; • Systematize available information for planning future transportation; • Automatic processing of the vehicle’s arrival to the final stop and the ability to adjust the time in case of trip delays.
In recent years, researchers have shown great
interest in the public transport system. This is
largely due to the growth of cities and transport
development of territories [
However, despite the large number of publications related to the development of transport management systems, it is important to consider the applied aspects of the development of their infrastructure and implementation stages of design solutions.
In addition, one of the main problems associated with TS are external and internal integration processes, such as: the implementation of data dependencies and tracking their changes, tracking cars using intermediate services, and so on.
Thus, the main goal of this project is to develop an automated passenger transportation management system of the transport company with the ability to: systematize data obtained during trips, receive instructions for creating and installing new services, infrastructure preparation (ease and speed of deployment, integration of customer data, etc.), the ability to use the system on any platform and support for gadgets from phone to computer.
The main non-functional requirements (NFR) for the project are: • completeness of information and technical specifications; • availability of the service regardless of time; • the ability to expand the content; • dynamic cartography, independence from the map provider; • confidentiality of customer data; • security of obtaining data on synchronization of movements of system objects; • service of operators and drivers in real time with minimal delays; • design flexibility and ease of use by end customers; • the ability to scale the system and the number of services during support; • the possibility of the system deployment process on different platforms; • support for data integration using resources and sources implemented by the customer; • possibility of constant logging and data recovery; • simplification of visualization on mobile devices and tablets, to maintain the full consistency of data, as well as the use of the platform by operators in cases of inability to use a computer; • fast conversion of content with translation into other languages; • dynamic specifications (vertical expansion): CPU speed, memory, disk space, network performance.
In addition, to determine the progress of the project, functional sections were introduced, which in turn build the system as a whole: Models, Vehicles, Drivers, Contacts, Locations, Cities / Regions / Countries, Routes, Schedules, Trips, System Settings.
With each passing day, the implementation or
sequential integration of artificial intelligence
(AI) according to [
Most modern systems use the Global
Positioning System (GPS), which allows you to
get the coordinates of vehicles and place marks on
a virtual map [
It should be noted that the developed system also implements this approach and deepens it. This is an optimization model, when information is provided in small portions, and the user while navigating the virtual map loads updated or changed data. This reduces the load on the server and facilitates display on the client application.
Additionally, the system provides the ability to automatically calculate and build routes, simplify the workflow for operators and synchronize statuses for system users. In addition, the system can be used as an intermediate node to integrate into an existing system and used to gather information and provide a graphical interface for operators. This approach saves time on reporting and allows you to use the data recorded in the system as automatically generated reports. This increases the level of security and also reduces the risk of error on the part of the operator. All current data: vehicle location, rerouting, trip status update, adding or deleting locations and system users, are stored at the information store level.
Note that most of the existing logistics systems are tied only to the construction of vehicle tracking without the possibility of integration into this process of other users of the system, such as operators, administrators, persons who provide parking and repair facilities, etc. This project solves the main problem of transport companies: combining all operators, drivers and intermediate users of the system in a single application and saving current data for sequential analysis.
Project consumers can choose their own configuration and use integration processes to enter existing data (drivers, locations, automotive objects, routes, etc.) using an external integration service. This allows you to not disrupt the technological processes of the business and make it easy to combine two or more projects in one ecosystem.
It is assumed that the consumer is a selfemployed person or a commercial group who is interested in saving time and money on late shipments, and instead wants to use systematization, storing important data on travel and drivers for further analysis. Data analysis of this type can provide an understanding of how employees perform their work and will allow them to track problems without direct contact with operators.
There are several basic approaches to
obtaining information from external resources
[
All of them allow you to get up-to-date information using external systems.
WebSocket is a computer communication protocol that provides full-duplex communication channels over a single TCP connection. It is synchronous, which gives a high level of correctness, but connecting a large number of connections can lead to poor performance of the module.
Server-sent events (SSEs) are a one-way communication channel where events are transmitted only from the server to the client. Events sent by the server allow browser clients to receive the stream of events from the server over an HTTP connection without constant polling.
It was decided to use a combined approach to connect the system to the constant updating of vehicles via the GPS system: to create an intermediate module that will use web sockets, and to receive synchronous information.
As soon as the socket sends a new message from the vehicle tracker, this message is transmitted to the asynchronous queue. Thus, the queue will use the SSE approach, which sends messages to all system subscribers (authenticated users) through other communication channels (Fig. 1).
The project is based on the principles of the structure of micro-services, as shown in Fig. 2. The first service is a graphical user interface (GUI) layer what is a web application. The next mechanism in the system that provides most of the data changes and external resource providers is called the micro-service pool. A microservice is a small piece of system that can reside on different servers or even be embedded in a partition of an application, and works with external APIs, data warehouses, or prepares scheduling events.
The basic access service (BDS) is a common
component for connecting to the data access level
(DAL) and some necessary plug-ins: mapping,
service data, etc. An internal database is an
abstract layer that can be switched between
providers and has a single interface. The only
limitation is the need to have entities that are used
in the main application and stored procedures.
This type of abstraction makes it possible to
choose the most convenient provider for the
database and absolutely provides a connection to
the structure, as separate as possible from
dependencies. This principle is the principle of
GRASP, Low Coupling & High Cohesion [
External datasets will be processed by the Data Integration Service (DIS), which will then be passed to a mapping process, where the data is converted to an internal model. Authentication service (AS) is a modern generation of token systems called JWT (JSON Web Token). This service allows you to authenticate and then identify moderators using only a hash with metadata sealed in it. The Geolocation Service (GS) is the "core" of the project and its main part. This means that the GS directly affects the data, in addition, it must build linked lists and include some additional data about places and objects.
It should be noted another pattern that was
used during the implementation of the system: the
Routing Service (RS) is based and configured
using the TomTom API. To reduce requests, a
Proxy template [
For example, if a route has the same location, the system will use the saved data instead of making a new request.
The scheduler is the only module that can be either a separate service or a built-in part of the developed system. For the data warehouse, SQL Server was chosen, which provides the best connection experience and many features, such as triggers to schedule or even save data analysis if necessary.
To explain the implementation, consider the Use-Case diagram in the Fig. 5, which presents the two main users of the system: the administrator and the driver.
As can be seen from the diagram, the driver has the following functionalities: go through the process of authorization and authentication, has automatic logging of its location and display in the application for administrators, can create requests to solve problems during trips, search for parking or repair; register intermediate statuses and terminate trips automatically or manually.
In turn, the administrator monitors the capabilities and responsibilities of drivers, confirming actions, as well as coordinates and solves problems as needed: creates new administrators and moderators, drivers, logs in the system, creates locations, fills in new information for system users, sets stop for repairs and parking, solves problems and problems of drivers during trips, creates new routes and manages trips.
The next step of implementation is the functions of creating new entities of the system, as shown below in the activity diagrams: the processes of creating a new driver, creating and making changes to routes and locations, creating a trip and accompanying processes during its execution.
As shown in the Fig. 6 - the process of creating a driver begins with filling out a questionnaire and then it branches into two parallel processes, which can be asynchronous, which means the possibility of their delayed completion. If necessary, a separate entry is created for the driver with a new bus or an existing one is selected. After filling out the questionnaire, you can additionally fill in the contact details. Finally, all processes are synchronized and sent to the database.
Creating and updating routes begins with a modal window that offers to specify or redefine the starting point, if this has been done or the route already has a starting position, the user can select new intermediate stops or change existing ones. The final step is the process of specifying the last location, which cannot be changed after. The system automatically offers estimates of time and route length based on data obtained during route calculation.
In order to start a trip, you need to create a new schedule or choose an existing one, after which the algorithm branches into two parallel processes: creating a system of stops with time intervals, setting metadata for a specific trip, as in Fig. 7. The system also has an automatic time calculation, which determines how much real time is required to travel the distance based on the geolocation of data. The process of creating metadata takes responsibility for specifying which bus, driver will perform the trip, as well as adjusting such parameters as date and time, etc.
Finally, we note the life cycle of the entire system through a state diagram (Fig. 8), which indicates the process, starting from the creation of the user, ending with the trip process and logging its final destination.
The developed system and the conducted researches introduce the service methodology for creation or integration into the existing systems for simplification of processes of synchronization between drivers and operators. The proposed approaches allow you to customize the system to individual customer needs and allow you to quickly expand the functionality regardless of existing ones.
As mentioned in the sections, each function or micro-service corresponds to an independent micro-service, which helps to increase the stability of the system. Despite the fact that the system already has a lot of services to address the basic needs of customers, it has and will expand in commercial projects. This project solves the problem of bureaucracy when generating reports, analyzing trips, creating plans, etc. But the most important aspect it covers is the need to synchronize information between operators and drivers.
As every professional employer wants to provide good working conditions, and in the transport industry there is a growing demand for personnel management, and the staff is expanding every year, there is a need to quickly form trip schedules to avoid conflicts. Conflict situations mean overlapping schedules of routes, as well as emergencies. .
It should also be noted that iterative operations were performed to increase optimization and correctness of the system, which led to an increase in possible connections to the tracking. All locations are stored in chunks, and when the client moves, it receives a new image with new data and cached previous ones, which saves memory and makes the application faster. Each route created by the user will be saved in the repository, and when the user tries to build a new route with multiple repetitive routes, this will result in retrieving the saved data instead of calling the API. This saves the client money and affects the performance of the application.