Overhaul processes of long-living, cost-intensive machines
and facilities are time-consuming tasks. They aim either at full
recovery of the original product condition through repair or
spare part exchange or at modernization for performance
enhancement. However, decision for an overhaul should be
carefully considered, because realization may be difficult und
time-critical. Every overhaul process is unique and based on a
thoroughly diagnosis of product condition. This makes it a risky
and hard to plan project. In this context speed of overhaul
operation is essential for avoiding costs due to machine down
times. Obsolescence of components and modernization goals
demand for an efficient reengineering to design geometric models
for production. Modern 3D scanning technologies deliver 3D
models of actual product geometry and allow deviation and
tolerance analyses in case of available reference models.
However, optical limitations and difficult part disassembly make
3D digitization still a laborious task and is additionally followed
by a high effort in data post-processing. This paper depicts a new
approach to facilitate reengineering processes through advanced
methods in 3D scan data analysis of non-disassembled products.
Implementations allow parts identification in 3D assembly scans
through shape recognition and database search for provision of
needed CAD1 data.