Keywords

1 Introduction

The assembly of large-scale flight systems is an essential part of the value-added process in the aviation industry. Fuselages, wings, engines and other components are assembled by hand in cycle lines. In wing outfitting, the assembly of high-lift systems is traditionally done by mounting the individual components directly to the wing box. With a plug and fly assembly concept, the high-lift system can be pre-assembled, adjusted and tested as a stand-alone unit [1]. The ready-to-fly module is then joined to the wing in the final assembly line (FAL) with only a few joints. By outsourcing the assembly of the high-lift system, the cycle time in the wing outfitting can be shortened and the factory production rate increased. In addition, pre-assembly of the high-lift system can improve the ergonomics of assembly because the subassembly offers improved accessibility and can be moved to an ergonomically favourable position and orientation with less effort. For such a modular design of the wing, no reference concepts for the assembly of the high-lift system exist yet. Both the assembly organisation and the required operating equipment must be rethought.

The use of assembly jigs for precise and repeatable assembly of the components is widespread in the aerospace industry [2]. They are required to ensure accurate joining operations during the assembly of large dimensional aircraft components such as wings and high-lift systems. The jigs must be rigid and precise, and must therefore be matched to the product and the assembly process in question. This results in inflexibility with respect to shape and dimensional changes of the product [3]. New innovative jigs are needed to ensure high manufacturing accuracies and flexibility. They must be able to position large components easily and be flexible at the same time [3]. To combine high productivity and flexibility, assembly fixtures need a higher degree of automation. They also need to provide greater adaptability and improved interaction with workers [4]. To meet these requirements, collaborative robots seem suitable. They can relieve humans and protect them from physical overload by taking over heavy and repetitive tasks [5].

The aim of the research work presented here is the development of an adaptive jig that offers a high degree of adaptability with regard to product as well as process changes. The jig should enable the entire assembly process of the high-lift system in one clamping. In addition, it should offer workers the possibility for individual adjustments of the working position in order to improve both physical and cognitive ergonomics. The concept of such an adaptive jig is presented in [6]. The adaptive jig uses collaborative robots to position the components to be assembled. Consideration of physical and cognitive ergonomics has gained importance in the design of systems where humans and robots collaborate [5]. Several research efforts focus on methods to simulate physical and cognitive ergonomics with models and evaluate the acceptance of the human–robot-collaboration before building the real device. Beuß et al. propose an ergonomics study based on a simulation and virtual reality [7]. They describe that the analysis with digital humans is possible. Fritzsche shows the high degree of agreement between real and virtual ergonomics assessments and thus proves their usefulness [8].

This paper investigates the influence of the adaptive jig on the physical ergonomics. Using the assembly of a high-lift system as an example, the adaptive jig is compared with a rigid jig. Human modelling in a CAD system is used for a first evaluation of the ergonomic potential. The key indicator method is used to measure the risk of physical overload. The investigation will show if the adaptive jig is suitable for assembly and if it offers at least an equivalent ergonomic potential as a rigid jig. Based on the results, a decision can be made for or against building a physical demonstrator and conducting real-world tests.

2 Product and jig Design

2.1 Assembly Object

The investigated product is a high-lift system of a medium-range jet. However, only the assembly of the outboard landing flap with the associated supports is considered. Figure 1 shows a section with the main elements. The basic component is the aero flap support (AFS). It carries the moving components of the high-lift system and at the same time acts as an aerodynamic fairing [1]. The flap lever, actuator and landing flap are connected to the support by means of bolt connections. The high-lift unit is connected to the wing box at three points through the main and forward attachments. All bolt connections are designed to be fail-safe, i.e. they consist of two bolts slid into each other in opposite directions, which are fixed with lock nuts and locking plates. Since each bolt connection consists of at least six components, depending on the design, the components are summarised in the following under the term assembly kit (AK). It is assumed that the components are provided ready for assembly at the assembly line. Manufacturing operations such as drilling, milling, deburring or surface treatment are not part of the consideration. For example, it is assumed that the main bridge and other metal brackets on the AFS are already joined when it arrives at the pre-assembly line.

Fig. 1
figure 1

High-lift system with bolt connectors [9]

Full size image

To compensate for angular errors and reduce stresses, e.g. during thermal expansion, spherical bearings are integrated in all connection points. Until assembly is completed, all components are therefore movable in several degrees of freedom in relation to each other and must therefore be supported and held in position by a fixture.

2.2 Assembly Devices

Adaptive Jig. Figure 2 shows a raw construction of the adaptive assembly device. The positioning and orientation of the assembly parts is taken over by industrial robots, which have suitable end effectors for clamping the components. The industrial robots are arranged on a circular seventh axis. Due to its shape, this adaptive assembly device is also called assembly wheel [6]. The redundant kinematics give the robots an additional degree of freedom. This can be used to move the robots during assembly into a favourable position that interferes least with the workers’ work process. In this way, accessibility to the assembly points can be increased.

Fig. 2
figure 2

Adaptive Jig with support (left) and high-lift system (right) [6]

Full size image

The first robot carries the assembly while other robots feed the assembly components and position them for the joining process. Workers then assemble the bolt connectors manually. The robots are able to change the position of the assembly in space so that workers of different heights can comfortably work on the object. Furthermore, changing the orientation can prevent working overhead or while kneeling. To ensure safe operation, the robots must be equipped with functions for human-robot collaboration, like force sensors and robot skin. During assembly, each support is initially equipped in a separate assembly wheel. Then the assembly wheels with the supports are brought together and the landing flap is added.

Rigid jig. Since the Plug and Fly high-lift system is a completely new product for which no reference process exists so far, a rudimentary concept for a rigid jig had to be created for this study. Ergonomic requirements were considered in the design, just as they would be in an industrial design. It is assumed that the construction consists of a rigid frame of welded hollow sections. Functions for adjusting the height or orientation of the assembly are not integrated. In order to provide an approximately optimal working height for all workers, an average working height of 1100 mm was set.

The supports are inserted into a clamping device and fixed therein during assembly. The assembly is carried out in horizontal orientation, which corresponds to the flight orientation. The landing flap is assembled in the extracted condition to improve accessibility to the joining points. Figure 3 shows a CAD representation of the concept for the rigid jig. Since only the general contour and the geometric arrangement of the assembly parts in the jig are relevant for determining the influence on an ergonomic working procedure, details such as the clamping devices were not designed.

Fig. 3
figure 3

Rigid jig holds support and landing flap

Full size image

3 Process Design

To evaluate the effects on ergonomics, the fixtures must be considered in the context of a work process. For the assembly of the high-lift system, an assembly sequence was determined experimentally in workshops [10]. Based on this assembly sequence, work steps have been defined and an allocation of labor in the cycle line is determined.

3.1 Determination of the Work Steps

The key indicator method [11] is used to compare the effects on ergonomics. It is used to evaluate the work processes both when using the adaptive jig and when using the rigid jig. The risk values can then be compared in conclusion. Depending on the type of load, different forms must be used in the KIM. Each of the assembly processes considered is therefore first broken down into work steps, each of which contains only operations of a uniform load type. For the assessment of the assembly of the high-lift system, the forms for the assessment of Lifting, Holding and Carrying of loads (LHC) as well as for the assessment of Manual Handling Operations (MHO) are sufficient. Methods-time Measurement (MTM) was used to determine the working times for these work steps. Table 1 shows an overview of the worksteps defined for the two work processes.

Table 1 Work steps for the assembly process with adaptive jig
Full size table

3.2 Work Scheduling

Assuming that 63 aircraft are to be produced per month and that 17 shifts of seven hours each are available per week, this results in a maximum cycle time of approximately 3.75 h per wing (i.e. outboard high-lift system). The actual assembly time per high-lift system must be shorter than the cycle time. The assembly time comprises the basic time, the recovery time and the distribution time. The basic time is formed by the sum of the MTM values of all work steps. The recovery time corresponds to legal requirements and the distribution time is estimated based on values from the Federal Ministry of the Interior and Community. Since some work steps can only be performed by two people, at least two workers must be scheduled to perform the assembly. Detailed work planning shows that at least three people are required to complete all work steps in the required cycle time. This applies to assembly with the adaptive jig as well as to assembly with the rigid jig. For both assembly processes, a task allocation is carried out in which three people are equally occupied. The basic assembly time is then approximately 2.8 h per high-lift system. This leaves sufficient recovery and distribution time within the cycle time. The allocation of the tasks is included in the determination of the key indicators in the following chapter and is primarily represented there by the task durations and the number of repetitive movements.

4 Determination of Key Indicators

The key indicator method (KIM) has become a standard across companies to evaluate the ergonomics of a working process [12]. In the automotive industry for example, it is used to evaluate assembly activities. The different key indicator methods are designed as a basic methodological for the risk assessment. They describe the most important stress factors (key indicators) in ordinal scales and determine the degree of likelihood of physical overload [11]. The methods are well evaluated and digital forms are provided for easy execution [13].

To carry out the KIM, the postures of the workers during the assembly processes have to be observed and the assembly times have to be determined. Since the study is conducted before the concept is finalised and the jig is actually built, the study takes place with virtual objects. Both the adaptive and the rigid jig exist as CAD models. The programme Siemens NX 11 is used to integrate human models into the CAD models. Four different human models, representing the 5th and 95th percentile of the male and female German population respectively, are used for the investigation. Each of the previously defined work steps is reproduced in a CAD scenario. In each case, a posture that is characteristic for the assembly step is simulated with the human models. Based on these scenarios (e.g. Figure 4), the posture can then be evaluated within the scope of the KIM. In addition, the human models are used to check whether there is sufficient visibility of the assembly spot and the workers’ own hands. The key indicators that cannot be clearly identified in the virtual model (such as work organisation) are always assumed to be best possible.

Fig. 4
figure 4

Example of posture and view analysis with human models in Siemens NX 11 (Visibility of jig is deactivated.)

Full size image

Table 2 shows and 3 show the results of the KIM for both assembly jigs. The risk scores are classified in four categories. Risks below 20 correspond to a low load of intensity and risks between 20 und 50 indicate a slightly increased load. Both are acceptable. Risk values between 50 and 100 belong to substantially increased load intensities and afford a redesign of the workplace since physical overload is possible for normally resilient persons. Work steps with a risk value above 100 have a high load intensity and will likely cause physical overload to all persons. [14].

Table 2 Work steps and corresponding risk values for the assembly process with rigid jig
Full size table
Table 3 Work steps and corresponding risk values for the assembly process with adaptive jig
Full size table

It can be seen that most works steps cause a low or slightly increased physical load. However, the holding processes may cause an increased load especially for female workers. The rigid jig requires the manual positioning of the fairing within the jig. This work step causes loads that are too high for women. So this work step needs to be redesigned if the rigid jig should be use. While using the adaptive jig the highest risk occurs during the providing of the actuator to the robot. This step can easily be eliminated by providing the actuator with a carrier. While the values in Table 2 and 3 only represent the risk of single work steps, Table 4 and 5 show the cumulated values for a whole working day, considering the working process described in chapter 3.2. Since workers perform different tasks during their shift, the risk is evenly distributed. Only women have a high risk of physical overload during lifting and holding since they all have to position the fairing together when using the rigid jig.

Table 4 Cumulated risk values per working day and person with rigid jig
Full size table
Table 5 Cumulated risk values per working day and person with adaptive jig
Full size table

5 Conclusion and Summary

It has been shown that the adaptive jig offers the same or even better ergonomic performance compared to a rigid jig. In particular, it provides very good support when positioning heavy objects. The adaptive jig therefore improves the work process. It can be assumed that work processes such as adjustment, electrical equipment or painting also benefit from the adaptive jig. In particular, when several people with different physical constitutions work together and when integrating people with physical disabilities, the adaptive device offers further potential for improving ergonomics. For example, the assembly object can be brought into an orientation where the assembly points are presented to the employees at different heights.

The adaptive design makes it possible to carry out the entire production process of a high-lift system in a single setup. There is no need for relocation and remeasurement in another fixture. Further potential arises when considering the adaptive jig over its life cycle. The jig can be adapted to changes in the production process or to product changes with little effort. This results in a very long service life. Because not all aspects of work ergonomics could be investigated with the chosen method (e.g. psychological factors), the use of a physical demonstrator is necessary for a complete evaluation.