Les relations entre le vieillissement des salariés et les conditions de travail se situent à deux niveaux: l'un individuel qui est propre au vieillissement de chacun, l'autre collectif qui est marqué par les phénomènes de génération et d'époque. Ces relations ont été étudiées dans l'industrie automobile française en se centrant sur le travail répétitif sous cadence imposée et en associant des approches en démographie du travail et des approches ergonomiques. Les résultats permettent de définir des actions anticipatrices pour éviter les effets négatifs d'une double évolution : celle du vieillissement de la population des opérateurs et celle de l'organisation du travail.
Throughout their lifetime individuals change through a process of environment-associated decline and construction. But repetitive jobs, inspired by Taylorist principles, do not take these changes into account. This contradiction between aging and repetitive tasks has a negative effect on individuals and their work activity. Designing workstations and implementing work organization requires an understanding of these consequences: how does repetitive work act upon decline and construction processes? How do these intricate aging problems modify the way people achieve their tasks? Three concems have to be kept in mind: 1) work may initiate or accelerate decline processes, and it may also be at the heart of enrichment and experience; 2) as age increases, interindividual differences increase; 3) generation effects, as well as technical and organizational changes, play a part in individuals' aging. This leads to various methodological approaches: collective (demography) or more individualist (ergonomics). We have tried to apply this interdisciplinary approach to the automotive industry.
In France, the labour force is not yet "old". But the rising numbers of middle-aged workers — due to demographie trends, delayed school-leaving age, and early retirement — allows us to foresee a growing proportion of older workers in the next few years. In some firms, including the automotive sector, this evolution has already started, leading to various difficulties in human resources management, job allocation, and mobility. These issues appear to be particularly worrisome in the case of repetitive tasks. Tight time constraints, known to be painful for older workers, are now more common in this age category. Selection mechanisms which until recently had allowed the removal of the elderly from assembly lines, are no longer sufficient. In the firm where our study was conducted, this problem is all the more delicate because "soft" workstations (formerly reserved for older workers) are less numerous, cycle time is getting shorter, and just-in-time Systems are resulting in unexpected sequences of different car models.
In order to cope with this issue, the firm set up a "workstations evaluation grid", in which effort, positions, time-pressure, and organizational constraints are considered. The grid then enables the job designers to target the most difficult situations (especially for older workers) and to improve these situations.
Aware of the interest, but also the limits, of such an instrument, the firm asked us whether other factors than those which were used in the grid, had to be taken into account, and how to pay attention to interindividual variability. In agreement with the foremen and workers, we chose to study two different workshops, specializing in motor assembly and facia board fixing. In this article we examine the results concerning the motor assembly workshop, which operates with two 18-employee shifts which change weekly.
At each workstation one has to assemble a main part and a few small ones. The operator picks up the part with the matching screws, tightens it finger-tight, and then tightens it with screwing machines. This kind of task requires sensorimotor skills, together with speed and accuracy demands. Cycle time is around 1 minute 30 seconds, which necessitates fast choices of parts according to the car model.
After interviewing operators and spending time observing them, we filmed one workstation run at various periods by four operators of different ages (from 30 to 46 years old) in order to understand the different ways of performing the task.
Analysis of the videotapes show that as age increases, forms of rythmic steadiness are established. They also show that older workers develop strategies to save time and avoid overdemands on some parts of the body. For example, they combine "picking-up" activities in order to reduce the frequency of walks; and they try to replace sight by touch to avoid bending. Developing these strategies and adjustments requires, in turn, the development of fine sensorimotor skills, such as the handling of several small parts in the same hand, where each part has its own location and moves from palm to fingers without involving the operator's visual attention. Younger operators are also able to use these kinds of strategies, and in fact these can be observed now and again. But among older workers, the adjustments are more systematic and result from a long-term reorganization of their way of doing things. This reorganization, linked with their view of their own abilities and aimed at compensating for the effects of aging, is "costly" for the operators and leads them to protect themselves against potential disturbances. Consequently, as the interviews have shown, older workers are less involved in mutual aid activities, and are more likely to want to remain at the same workstation despite monotony. We were able, through demographie analysis, to verify that job-rotation is less frequent as age increases.
Such a study, although needing to be strengthened, sheds more light than experimental approaches would do on the richness and complexity of the links between work and age-related decline and construction processes. In addition, it enables us to question two widespread attitudes. The first consists in considering aging issues as secondary, because technical changes and individual motivation can easily solve them. The second consists in viewing aging uniquely as a process of decay, to which no implementation can efficiently respond.
Our study suggests that some improvements are possible, based on a key principle: the extension, as far as possible, of "room to manoeuver", at individual (choice of gestures, movements, rythm) as well as collective levels (workstation rotation). These concerns have also appeared in other studies on the same topic in various fields of activity such as mattress-cover stitching, aircraft assembling, and quality control in steelworks. Nevertheless, one must admit that repetitive tasks, and on-going organizations, leave little space for improving the working conditions of older workers and that the latter cannot always transfer their strategies and "knacks" to other situations.