Keywords

1 Introduction

At present, passenger demand for urban rail information services is increasing. It is required the PIS to be updated more quickly, and at the same time, the PIS has the characteristics of a long development cycle and complex system. Therefore, the MBSE idea is used to design the PIS system.

MBSE is a modeling idea that drives systems with models, enabling modeling approaches to support activities, such as system requirements, design, analysis, verification, and validation [1]. This enables the faster development of systems and reduces the number of design iterations [2]. At this stage, it is widely used in large complex systems and software development projects for different complex systems and industrial areas with different methods, languages, and tools [3]. The relatively mainstream methods are the Object-Oriented Systems Engineering Method (OOSEM), Arcadia, Magic Grid, and other methods, modeling tools such as Cameo Systems Modeler, Visual Paradigm, M-Design, and common modeling languages such as Simulink, STATEMATE, UML, SysML [4].

Peng Kun et al. [5] adopted the MBSE idea for the forward design of a manned lunar spacecraft system. Mei Qian et al. [6] applied MBSE to design the functional architecture of a civil aircraft. Liu Chang et al. [7] demand analysis of intake air damper control based on MBSE. Chao Tang et al. [8] used the Raphsody method to develop the function of a civil aircraft system. Baomin Wang et al. [9] applied the MBSE idea to simulate the control process of a pantograph. Yuan Wenqiang [10] built a functional knowledge base and component base for a high-voltage system for rolling stock based on MBSE. Chen Bo et al. [11] used the SYSML language to model and simulate the end-change scenario of a moving train set.

This research uses M-Design system modeling tool and SysML modeling language, combined with the characteristics of IPIS, proposes the BW of IPIS_IPIS modeling method. M-Design, as a modeling tool, can achieve syntactic modeling of requirements, behaviors, structures, and parameters, and build a metamodel library for secondary development of IPIS.

2 Methodology Applicable to IPIS

In modeling IPIS, the whole system is divided into two layers, B-black and W-white, B-black refers to modeling at the black box level and W-white refers to modeling at the white box level.

Analysis of the B-Black layer. In this stage, the stakeholders of the system (B1) are studied in depth, and their crowd-related requirements (B2) are deduced based on the stakeholder analysis. The system's usage scenarios are analyzed in detail, the corresponding use case scenarios are identified (B3). These use case scenarios were modeled by the black box activity diagram modeling method (B4).

Analysis of the W-White layer. Convert some activities in a black box activity diagram into a white box activity diagram (W1). For the activities in the W1 layer, cluster analysis was performed to determine the required functionality of the system (W2). The functional analysis of the W2 layer is studied in depth, the requirements of the system (W3) are refined. After completing the functional analysis and system requirements, analyze the system architecture (W4). The physical architecture (W5) is obtained by analyzing the physical components and structures that can implement the system functions in the W2 layer.

The detailed research methodology flow is shown in Table 1.

Table 1. The framework of the BW_IPIS method.
Full size table

3 IPIS Modeling

To model IPIS, this study was developed using PA as an example. The stakeholders of IPIS, such as station staff, passengers, and drivers, were analyzed at the B1 layer, as shown in Fig. 1. Stakeholder requirements were analyzed at the B2 layer, 89 crowd-sourced requirements were obtained, and the requirements were itemized, as shown in Fig. 2.

Fig. 1.
figure 1

Stakeholder analysis.

Full size image
Fig. 2.
figure 2

Partial entry view of stakeholder requirements.

Full size image

The use case scenario was abstracted at the B3 layer by analyzing the stakeholder requirements. The PA applications was distinguished by IDs. The first layer uses case ID x.x, and the second layer uses case ID x.x.x. The first layer includes ID 1.1 Public address content, ID 1.2 Public address method, ID 1.3 Intercom, and ID 1.4 Recordings in the case of a passenger pressing the emergency button, as shown in Fig. 3.

Fig. 3.
figure 3

Use case of PA applications.

Full size image

The black box activities of the ID1.2 Public address method were analyzed in B4 layer. ATC, Driver, TCMS denote the external environment participants under this use case, and the external participants of the system are assigned to different activities, as shown in Fig. 4.

The triggered Fully auto-play public address in Fig. 4. Was modeled at the W1 layer as a white box activity diagram, as shown in Fig. 5. The activity starts from the external input of the Triggered fully automatic public address, Speed information, and GPS location (distance to station) signals. After receiving the signal, IPIS executes the activity in the following order: Pre-playing arrival information, Play the name of the next station, Play open door message, Play closing message, and finally the activity ends.

No functional Stereotype in the SysML predefined Stereotype was established for IPIS. Therefore, a functional metamodel was customized, as shown in Fig. 6.

Fig. 4.
figure 4

PA black box activity diagram.

Full size image
Fig. 5.
figure 5

Fully auto-play public address activity diagram.

Full size image
Fig. 6.
figure 6

Functional element model.

Full size image

Functional analysis of W2 layer yielded 12 PA functions as shown in Fig. 7. By analyzing the system requirements at the W3 layer, the PA needs to meet the system-level requirements for Storage Requirements, Intercom, PA output control, Radio broadcasting, Digital Annunciation/Closing Alarm, and Monitoring, as shown in Fig. 8.

Fig. 7.
figure 7

PA functions.

Full size image
Fig. 8.
figure 8

PA system requirements.

Full size image

The system architecture analysis of PA in the W4 layer is divided into Broadcast System, Intercom System, Control System, and, Emergency System.

After completing the architecture analysis of the PA, a physical architecture analysis was performed on W5 layer. The Control System of the PA was implemented to obtain the Driver Announcement Controller, Passenger Announcement Controller, and Carriage Controller, as shown in Fig. 9.

Fig. 9.
figure 9

PA physical architecture.

Full size image

4 Conclusions

In this study, the concept of MBSE was applied to the study of rail transit IPIS. The main contributions are as follows:

  1. (1)

    Adopting a positive design concept from the perspective of stakeholders, with a demand oriented approach, meeting the needs of stakeholders, combining PA, CCTV, and PID functions to form IPIS, fully reflecting the characteristics of low resource consumption and high integration.

  2. (2)

    Using M-Design for IPIS design, a MBSE based approach was proposed BW_IPIS modeling method for IPIS. IPIS is divided into two layers: first analyze the B-black layer, and then analyze the W-white layer. In the B-black layer, stakeholder and crowd participation needs, use case scenarios, and black box activities were analyzed. In the W-white layer, white box activities, functions, subsystem decomposition, system requirements, system architecture, and physical architecture were analyzed. This method belongs to top-down active design, and the results of each stage can actively drive the design of the next stage.

  3. (3)

    According to the characteristics of IPIS, function meta model, performance meta model and system requirement meta model are established to improve the reusability of modeling elements and provide reference for the subsequent development of IPIS.