Keywords

1 Introduction

The civil aviation transportation area has consistently grown in recent years in China, leading to a considerable expansion of fleet sizes. In conjunction with these changes, a shortage of professionals in China has become an increasingly prominent problem. Civil aviation personnel including pilots, air traffic controllers and maintenance technicians are trying to handle greater task-loads, and this is causing a strain, especially in the area of maintenance [11]. Researches show that 20%–30% of engine failure in air, 50% of delays, and 50% of flight cancellations are caused by human errors in maintenance areas [13]. In a study on human errors, errors were found to be caused by the volume of work accounts for a large proportion of errors, and this indicated that excessive task-loads may be one of the most important factors inducing or leading to flight accidents. The high task-load does not only affect the physical and mental health of the crew but also creates potential risks to civil aviation operation [2]. Therefore, with flight safety in mind, studying the evaluation criteria of civil aviation personnel task-loads becomes vital.

To solve the series of problems caused by excessive task-loads effectively, research on task-load evaluation methods has become a key subject in the field of aviation. Task-load is a part of workload. Till date, research on workload has mainly focused on controllers, pilots, and dispatchers. For example, in 1984, ICAO proposed evaluating the workload of controllers and dispatchers through the DORATASK theory [9]. Moreover, in 1992, Corwin evaluated pilot workload by using a subjective load assessment method, which proved that the flight scores of high-load workers was higher than that of workers with lower workload under the same conditions [15]. In 1995, Tofukuji [14] discussed the workload of controllers based on airspace tension, checked whether the workload was acceptable, and then calculated the maximum actual traffic capacity. In 2000, Han [7] conducted several observational experiments on practical controls for air traffic approaches and area control airspaces in Guangzhou, China. He adopted the DORATASK method and analyzed the human factors that affected the capacity of sections. In 2012, Ren [12] analyzed the workload of the release dispatcher using single machine sorting to analyze the peak load and adopted a neighborhood variable to solve the assignment problem of dispatcher’s seat assignment. In 2014, Sun proposed a control worker’s workload measurement model that could represent the actual operation of Chinese air traffic controllers using a DORATASK load assessment method framework [16].

Based on the current research, the existing workload evaluation and measurement methods can be divided into two categories: subjective methods and objective methods. According to their characteristics and applicable scope, these methods can be divided into four categories: subjective evaluation methods, main task evaluation methods, auxiliary task evaluation methods, and physiological and biochemical index evaluation methods [18]. The application of these evaluation methods provides a basis for the evaluation and measurement of workloads, but they also have some limitations. The subjective evaluation method measures the workload according to the subjective feeling of the evaluation target. This evaluation result is accurate but it is too subjective. The main task and auxiliary task evaluation methods are highly dependent on the operational analysis ability of the researcher, so they are difficult to implement. The physiological and biochemical index evaluation method requires that the subjects wear various measurement instruments to observe physiological functions. Thus, the test cost is high, the study time consuming, the operation environment and experimental equipment are high demanded, and the laboratory requirements are strict.

2 Methodology

2.1 Research Framework

Based on an in-depth investigation and analysis of the maintenance work of civil aviation maintenance staff, this paper analyzes the impact of different workloads on the performance of the human body by using the JACK 8.4 engineering software to introduce how JACK is used in maintenance workload evaluations to provide a novel and portable system. The framework is as follow as Fig. 1 shows.

Fig. 1.
figure 1

The research framework of this paper

Full size image

2.2 JACK Simulation Software Introduction

Jack is a human factors engineering analysis software from Siemens Industry Software (formerly UGS). Originally developed at the Center for Human Modeling and Simulation at the University of Pennsylvania in 1995, then commercialized by Siemens Industry Software more than 10 years. After more than ten years of research and improvement, the software has become a high-end simulation software integrating 3D simulation, digital human modeling and human performance analysis.

As a real-time visual simulation system, JACK can import CAD 3D model created by users and construct simulation environment. It can create a 3D human body model with biomechanical properties, assign tasks to digital human and obtain valuable information through the digital human behavior simulation analysis. This digital human model has been widely used in many fields of scientific research and engineering optimization design. The Jack Task Analysis Toolkit (TAT) enables researchers to evaluate human performance from an in-depth ergonomics perspective early in the product lifecycle, before designs are frozen and changes require costly rework. The toolkit can evaluate tasks using the Jack (male) and Jill (female) human models, without ever putting real workers at risk.

2.3 Maintenance Task-Load Evaluation Model

Civil maintenance work is characterized by high levels of time pressure, a complex working environment, standardized procedures and a substantial physical workload [5, 6]. Based on the definition of working loads, this paper proposes a basic measurement called the “unit time task-load” as the maintenance task-load.

The “task-load” is one of the representation of workload. Workload is the quantity and quality of work tasks. Civil aviation engineering maintenance is a complicated and difficult work, so it is difficult to measure it directly. ICAO and CAAC used the DORATASK theory during the workload evaluation of air traffic controllers and dispatchers. First, they classified the work into specific categories. Then, through the observation and measurement of each sub-task in the task process, they used the “actual time consumed” to measure the workload [1, 8].

In this paper, the basic idea is to measure the task-load based on the rate of time occupancy. Using the above approach, we proposes the following task-load evaluation measurement model for maintenance personnel:

$$ {\text{TLI}} = \frac{{{T_{\text{s}}}}}{T_a} $$
(1)

In Eq. (1), TLI is the task-load index; \( {T_s} \) is the simulated working time of a standard maintenance operation process through JACK; and \( {T_a} \) is available time provide from maintenance management of a maintenance task.

Here TLI are classified six grades:

  • Grade 1: TLI ≤ 0.2, that the task-load is almost non-existent, and the available maintenance man-hours (MMH) needs to be re-planned;

  • Grade 2: 0.2 < TLI ≤ 0.4, that the task-load is low, and the currently scheduled MMH may be redundant for the actual maintenance task. The maintenance management needs to reduce the scheduled time of this task, in order to optimize the efficiency of maintenance;

  • Grade 3: 0.4 < TLI ≤ 0.6, that the task-load is relatively low, the MMH can be adjusted according to the actual situation;

  • Grade 4: 0.6 < TLI ≤ 0.8, that the task-load is good, the current MMH is basically the same as that required in the actual work, do not need to modify;

  • Grade 5: 0.8 < TLI ≤ 1, that the task-load is high, the maintenance management needs to increase the time required for maintenance, for reducing the task-load of maintenance personnel;

  • Grade 6: TLI > 1, that the MMH is not reasonable, this maintenance task is impossible to be achieved and must be reformulated.

2.4 Maintenance Task-Load Evaluation Based on JACK Simulation

On account of the specific nature of the maintenance work, the existing task-load evaluation method has a series of disadvantages in both its applicability and evaluation methods. Finding a task-load evaluation technique, which is characterized by high operational and low process invasiveness, and is also customized for the characteristics of civil aviation maintenance, is essential. In this paper, we put forward a method to evaluate the task-load using the Task Simulation Builder (TSB) system and the Task Analysis Toolkit (TAT) in the JACK 8.4 software.

Civil aviation maintenance work involves a large number of project types, and the various maintenance programs differ to a large extent. To prevent the formation of maintenance errors and ensure the quality of maintenance, the maintenance management has standard operating procedures for each maintenance project. Strict rules about the order of the maintenance and the size of the tools exist [10]. These rules provide good conditions for the simulation analysis of the standard maintenance operation processes.

MTM-1 Time Prediction Method

For the simulation hours, this article uses data analyzed using JACK’s Predetermined Time System (PTS). It is based on the Time Measurement Method-1 (MTM-1) theory. PTS is an internationally recognized advanced technology for time standards. Its most significant characteristic is that it is used for all kinds of actions with varying operating standards to determine the time required to complete tasks rather than only through observation or measurement. It can accurately describe the action and add the predetermined time, avoiding randomness and uncertainty that come from tests or statistical sampling. The data thus obtained are more consistent and objective than the data obtained by other methods [4, 15].

The MTM theory establishes a standard time based on carrying our repeated “basic actions”. It is the most advanced and practical time measurement technology in the field of international industrial engineering. It can not only obtain accurate and objective time standards but also establish and improve working methods. This method is especially suitable for short period and high repeatability operations. The PTS system in the JACK software is mainly used as the base system of MTM theory, and it is called MTM-1. It analyzes each of the tasks of the overall process, identifies the tasks of the workers, uses the corresponding time in the basic action classification table, and then works out the time required to complete the entire process.

In civil aviation maintenance, by breaking down a set of maintenance tasks into multiple steps, we can analyze the variety and quantity of maintenance tasks. Then, by using the basic task time, the necessary time to complete the whole maintenance process can be calculated. In this way, predicting the time required to complete all the maintenance tasks during a maintenance operation and simulating the working time (\( {{\text{T}}_{\text{s}}} \)) needed to complete a maintenance task become possible.

Acquisition of the Simulation Work Time (\( {{\mathbf{T}}_{\mathbf{s}}} \))

Figure 2 shows the PTS operation interface in JACK 8.4 software.

Fig. 2.
figure 2

The PTS time prediction tool interface in JACK

Full size image

The system automatically calculates the \( {{\text{T}}_{\text{s}}} \) through an analysis of the dynamic simulation process and the MTM-1 theory. Then, combined with the available time which the maintenance management gave (\( {T_a} \)), the maintenance task-load index (TLI) is calculated as \( \frac{T_s}{T_a} \).

3 Case Study: Application of JACK in Removal of the Gear Wheel (for Airbus A320)

3.1 Static Environment and Dynamic Simulation Creation

JACK’s static simulation creation system is also a simulation of the working environment. The creation of JACK’s simulation environment includes the creation of 3D models, digital human body models and models for the relative position of each maintenance object.

For maintenance work, the entities related to the work scene mainly include various parts of the aircraft body, aircraft maintenance tools and maintenance aids. These aircraft accessories and specialized maintenance tools need to be created using an external 3D model, which is then imported into the scene using the entity import function of the JACK software. To ensure the accuracy and objectivity of the simulation, after the entities are imported into the JACK software system, it is sometimes necessary to modify and adjust the size, relative position relation, related mobile relation, and coordinate properties of all the actual datasets. Thus, a virtual work scene is set up to meet these requirements. In this paper, a preliminary 3D model of the related entities was created with the help of the Rhinoceros 5.0 system. The parts of the 3D model are shown in Fig. 3 below.

Fig. 3.
figure 3

A preliminary 3D model. a. Cooling fan for brake, b. cover for wheel fan, and c. annular protective cover

Full size image

This paper adopts the functions of standard digital human activities in JACK and builds a model of a human civil aviation maintenance personnel’s body based on Chinese anthropometric data (GB 10000-88) [3]. The digital human model is built by selecting different height, weight, and gender ratios.

All the equipment and body structures involved in the static scene are not required to be of uniform precision. To solve the difficulties associated with simulating the maintenance, we can only simulate the key attributes of the key object. Therefore, it is necessary to distinguish between the secondary features of the product model and simplify the process when the environment of the maintenance work is modeled. For example, some maintenance work associated with hangars and non-aircraft structures can be simplified, and the specific parts involved in maintenance projects can be detailed and completely simulated. Figure 4 shows a simplified version of the main wheel removal task for an Airbus A320, which was set up through the JACK static simulation module.

Fig. 4.
figure 4

Airbus A320 wheel remove operation scene

Full size image

This paper gathered the maintenance actions according to the theory of kinetic element analysis [17]. According to the standard operation process document and video of the Airbus A320 wheel removal, we performed level decomposition and analyzed the four parts of the standard demolition operation process. The four parts included removing the fan cover, removing the brake cooling fan, removing the shaft nut, and removing the main wheel. This gives the A320 main tire removal disassembly task shown in Table 1 below.

Table 1. Partial task action breakdown table of the main wheel of the A320.
Full size table

After obtaining the basic moving sequence, the TSB function in JACK8.4 was used to create a dynamic simulation of those four processes. Figure 5 shows screenshots based on the dynamic simulation video.

Fig. 5.
figure 5

Dynamic simulation video screenshot

Full size image

3.2 Task-Load Evaluation Parameter Acquisition

JACK can calculate and analyze the following data according to the relevant motion data of the dynamic simulation:

Figure 6 shows that the simulation working time \( {T_s} \) is 72 s.

Fig. 6.
figure 6

TSB simulation time analysis of CSV format output

Full size image

According to the investigation, the time \( {T_a}\; \) of maintenance management for the above maintenance tasks is 3 min, i.e., 180 s.

3.3 Calculation and Analysis of Task-Load Evaluation Results

According to the parameters of the TSB and TAT analysis tools, the task-load of the gear wheel removal task in the Airbus A320 was obtained. Using the task-load evaluation model, the detailed calculation process was as follows:

$$ TLI = \frac{T_s}{T_a} = \frac{72}{180} = 0.4 $$
(2)

The calculation result shows that the task-load of the gear wheel of the Airbus A320 is acceptable and can even be further optimized in terms of time utilization. For example, when making the maintenance work plan, the planned working time can be shortened and changed from 3 min to 2.5 min. This effectively improves time utilization while ensuring that the task-load is within acceptable limits.

4 Discussion and Conclusion

This paper combines the practical work of civil aviation personnel with time utilization as a basic idea to measure task-load. To evaluate the technical implementation, the paper creatively establishes a dynamic simulation method for actual maintenance work using TSB in JACK. Thus, by obtaining a task-load evaluation for each parameter, this method effectively solves the disadvantages from previous task-load evaluation methods, such as strong subjectivity, poor portability, and a requirement of strict laboratory conditions. In the evaluation of maintenance workloads, this paper applied a method of task-load measurement based on a time occupancy rate similar to DORATASK.

However, in practical applications, other factors that may affect the actual work should also be taken into account, such as the energy consumption rate, strength load, attitude load, and various natural environmental factors. Future research on the maintenance workload evaluation should focus on these factors.