The concept of workload essentially reveals the degree of adaptation between individuals and positions (or tasks). When the workload brought by a position is too high, it can lead to the inability to complete work tasks or cause huge physical and mental pressure on workers in a short period of time, resulting in decreased work performance and increased error rates. When the workload is too low, it can also cause adverse consequences such as mental neglect, decreased alertness, and waste of homework resources among the workers. Workload is the concept of interaction between individuals and positions, which affects the configuration plan of personnel on positions through interactive mechanisms. There is no unified consensus in the academic community regarding the connotation of workload, and its definition and evaluation methods are often diverse. In most situations, the definition of workload is mainly qualitative, usually referring to the amount of work that is applied or prepared to be applied to an objective carrier within a certain period of time. There are three main methods for evaluating workload. One is the subjective evaluation method, which studies workload by examining the operator’s subjective perception and feelings; The second method is task analysis, which studies workload by examining the amount of work tasks; The third is physiological measurement method, which studies workload by examining the physiological changes of operators ( Chu et al., 2025 ).

This work will use the DO-RATASK task analysis method to analyze the configuration of cabin crew. DO-RATASK is a workload assessment method proposed by the Operations Research and Analysis Council for air traffic controllers in the United States. Due to the similar nature and methods of work between high-speed train crew members and air traffic controllers, using the DO-RATASK method to quantitatively analyze the workload of crew members has high reliability ( Stoilova , 2020 ).

Through on-site tracking and investigation, this section will collect statistics on the working time of Guangzhou-Qingyuan intercity train conductors and cleaners under different passenger load conditions, and analyze their workload. It should be noted that as the work of mechanics mainly revolves around trains and is less affected by passenger flow, we only analyze the workload of train conductors and cleaners. In addition, considering that the current intercity trains in Guangzhou-Qingyuan include two types: eight formation and four formation, we also analyzed the workload of train conductors and cleaners under different formation conditions ( Van Rossum et al., 2024 ).

According to the train conductor’s work process and the required time attributes of the work tasks, we can divide the train conductor’s task units into fixed work tasks and variable work tasks. Among them, fixed work tasks (such as receiving trains at the platform, inspecting trains, organizing train capacity, etc.) refer to the work that the train conductor must complete according to the operation process, and the required time is generally fixed; And variable work tasks (such as carrying elderly and child care, serving key passengers, business acceptance, etc.) refer to the work tasks that the train conductor needs to adjust according to the needs of passengers, and the required time is variable, generally increasing with the increase of passenger flow. By using task analysis method and analyzing survey data and statistics, we can obtain the average passenger load factor and conductor workload rate of each train, as shown in Table 2 ( Wang et al., 2024 ; Wang et al., 2025 ).

According to existing research, there is a certain proportional relationship between the workload rate of train conductors and the passenger load factor of trains. Therefore, based on the data in Table 2 , we used a power function to fit the workload rate of the train conductor and the passenger load factor of the train, in order to obtain the functional relationship between the workload rate of the train conductor and the passenger load factor of the train. The fitting process is as follows:

According to Figure 2 , we can obtain the regression equation as W = 14.379K^0.4501, where R² = 0.9906, indicating that the reliability of the regression equation is high and can also be used to predict the future workload of train conductors.

Based on the functional relationship between W and K, we can predict the workload results of the train conductor under different passenger load factors, as shown in Table 3 .

Similarly, by using task analysis method to analyze survey data and statistical data, we can obtain the average passenger load rate and cleaner workload rate of each train, as shown in Table 4 .

We used a exponential function to fit the workload rate of cleaners and the passenger load factor of the train, in order to obtain the functional relationship between the workload rate of cleaners and the passenger load factor of the train. The fitting process is as follows:

According to Figure 3 , we can obtain the regression equation as W = 12.797e^0.0493K, where R² = 0.9993, indicating that the reliability of the regression equation is high and can also be used to predict the future workload of cleaners.

Similarly, based on the functional relationship between W and K, we can predict the workload results of cleaners under different passenger load factors, as shown in Table 5 .

The British Rail and Standards Agency divides workloads into three levels based on time occupancy rate ( Table 6 ):

According to Table 3 and Figure 2 , it can be seen that the passenger load factor ( $$ \text{K}$$) of the train has a significant impact on the workload of the train conductor. At that time, 0% ≤ K < 17%, and the train conductor was in a low load working condition. At this time, it is possible to consider the situation of the train conductor and the mechanic taking on both roles, but the prerequisite for taking on both roles is that the train conductor (or mechanic) must not only ensure the smooth completion of their own work unit, but also be proficient in each other’s work requirements; When 17% ≤ K < 31%, the conductor is in a medium load state; When 31% ≤ K < 37%, the conductor is in a high load state. It should be noted that if the conductor is in a high load state for a long time, the quality of service provided will decrease. At this time, a feasible solution is to arrange mechanics (or other positions) to assist the conductor in completing some of the work; When 37% ≤ K < 53%, one conductor is completely unable to complete the corresponding task and needs to designate one conductor to assist the conductor. At this time, both the conductor and the conductor are in a medium load state; When 53% ≤ K < 64%, the configured one conductor and one conductor are both in a high load state; When 64% ≤ K < 75%, one train conductor and one train attendant are no longer able to complete the job requirements, and an additional train attendant is needed to assist the conductor. At this time, the one train conductor and two train attendants are both in a medium load state; When 75% ≤ K < 86%, the configured train conductor and two train attendants are both in a high load state.

Similarly, by comparing Table 5 with Figure 3 , we can analyze the workload of the cleaners. When 0% ≤ K < 28%, the configured one cleaner is in a low load state; When 28% ≤ K < 37%, one assigned cleaner is in a medium load state; At that time, 37% ≤ K < 44%, and one assigned cleaner was in a high load state; At that time, 44% ≤ K < 58%, one cleaner was unable to complete the work task and an additional cleaner was needed. At this time, both cleaners called for a medium load state; When 58% ≤ K < 65%, the two assigned cleaners are in a high load state; At that time, 65% ≤ K < 72%, and 2 cleaners were unable to complete the task, so an additional cleaner was needed. At this time, all three cleaners are in a medium load state; When 72% ≤ K < 80%, the three assigned cleaners are in a high load state.

Based on the calculation and analysis of the appeal, we can obtain the configuration plan of the high-speed train crew under different passenger load factors, as shown in Table 7 .

The Guangqing intercity railway belongs to the commuting type of line. Due to its early opening, the current passenger occupancy rate is relatively low. Therefore, the personnel configuration plan adopted is “1 train conductor + 1 mechanic + 1 cleaner”. With the completion of the intercity railway network and the increasing maturity of operation, it is necessary to dynamically evaluate the selection of the crew configuration plan. Inter city railway line managers should dynamically predict the next stage’s passenger load factor and select the corresponding crew configuration plan based on the expected passenger load factor (refer to Table 2 ). In the early stage when passenger flow is not yet mature, it is considered to set up train conductors and mechanics simultaneously, but this requires adjustments to the organizational structure of management and employee training.