Search

The development of more comfortable seats is an important issue in the automotive industry. However, the development of new car seats is very time consuming and costly since it is typically based on experimental evaluation using prototypes. Computer models of the human-seat interaction could accelerate this process. The objective of this paper is to establish a protocol for the development of seat models using numerically efficient simulation techniques. The methodology is based on multi-body techniques: arbitrary surfaces, providing an accurate surface description, are attached to rigid bodies. The bodies are connected by kinematic joints, representing the seat back recliner and head restraint joint. Properties of the seat foam and frame have been lumped together. Further, experiments have been defined to characterise the mechanical properties required for the seat model for comfort applications. The protocol has been exemplified using a standard car seat. The seat model has been validated based on experiments with rigid loading devices with human-like shapes in terms of force-deflection characteristics. The response of the seat model agrees well with the experimental results. Therefore the presented method can be a useful tool in the seat development process, especially in early stages of the design process.

Seating comfort is becoming increasingly important for the automotive industry. Car manufacturers use seating comfort to distinguish their products from those of competitors. However, the development and design of a new, more comfortable seat is time consuming and costly. The introduction of computer models of human and seat will accelerate this process. The contact interaction between human and seat is an important factor in the comfort sensation of subjects. This paper presents a finite element (FE) model of the human buttocks, able to predict the pressure distribution between human and seating surface by its detailed and realistic geometric description. A validation study based on volunteer experiments shows reasonable correlation in pressure distributions between the buttocks model and the volunteers. Both for simulations on a rigid and a soft cushion, the model predicts realistic seat pressure distributions. A parameter study shows that a pressure distribution at the interface between human and seat strongly depends on variations in human flesh and seat cushion properties.