Development and evaluation of technical solution to avoid collision accidents between persons and forklift trucks

Status:

completed 07/2006

Aims:

Provide concepts to avoid accidents by:
- constructive arrangements
- optimise workflow (organisational arrangements)
- evaluation of a warning system

Activities/Methods:

- analysing the fatal accidents between 1993 and 2003 and categorisation in different accident types
- emphasise the reasons of the accident types and develop concepts
- animate the different accident types to illustrate constructive concepts
- work out a table of different sensor types and analyse their strengths and weakness
- build up sensor system to research possible combination as basis of a warning system
- work out different warning concepts

Results:

The analysis of the data yielded that there are six deadly accidents in Germany per annum. Each year one or two persons are killed because of insufficiently secured vehicles. This type of accident can be avoided by simple constructive and sensory systems.

Different types of accidents could be categorized after the evaluation and after detailed analysis of the courses of accidents. The surprising result was that most accidents occur during the forward motion of the trucks and not, as assumed, whilst moving backwards. Further, it was discovered that vehicles do not drive fully loaded. This could be an indication that the sight of the driver of a fork lift truck is limited severely by the constructive build up of the vehicle. The driver needs constructive, organisational and sensory aid in order to detect his environment and thus to avoid accidents.

The reports showed that certain danger areas around a vehicle can be defined. These can be separately treated for each direction of the vehicle. A danger area was defined to measure about 10m in front and behind the vehicle. The turn radius of a fork lift is chosen as the danger area for the loading und unloading of a truck, for example.

A sensory system must be able to monitor the defined danger area, adjust to the velocity of the vehicle, has to be fast enough to react und it must differentiate between humans and objects in order to avoid constant false alarms. A possible solution can only be realized by a combination of sensors. The requirements were listed in requirement specification. Sensors were chosen on basis of these criteria and been displayed in a matrix.

Radar-sensors in combination with passive infrared-sensors seem to be most appropriate. The combination fulfils all criteria, is inexpensive, covers all areas, is flexibly adjustable and reacts fast enough.

Last Update: