Sheet metal costing from part design
Ylihärsilä, Mikko (2018)
2018
Konetekniikka
Teknisten tieteiden tiedekunta - Faculty of Engineering Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2018-05-09
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201804241537
https://urn.fi/URN:NBN:fi:tty-201804241537
Tiivistelmä
Manufacturing cost and machining time estimations are important processes during design, quotation and process planning. Customers would like to have responses instantly for their requests for quotations. CAD-designers would like see the impacts of design changes on cost. In process planning phase cost and machining time information can be used for selecting which machine is used to manufacture parts.
The purpose of this study was to develop cost estimation model for estimating manufacturing costs for given sheet metal 3D-design model. This study focused especially into estimating machining time and material usage for machines using laser cutting, punching, forming and shear cutting technologies. Objectives for the study were that estimation model would be capable of giving estimates automatically and fast, estimates should be reasonably accurate in order to use them in practice and it should be possible to train model based on existing production data.
Neural network based model was developed for estimating machining time and material usage. Also different methods were developed for analyzing and measuring machining times and material usages for individual parts from existing production data. These measurements were used for training neural network models. Automatic feature recognition was used for processing 3D-models. This allowed processing 3D-models from different CAD-systems. Cost estimation model was implemented as web-based service which can be integrated into client applications over internet.
Final result was software implementation of the cost estimation model that could be trained by existing production data. Model can give cost estimation simultaneously for different machine types quickly and automatically. Also estimates were evaluated to be accurate enough for practical usage.
The purpose of this study was to develop cost estimation model for estimating manufacturing costs for given sheet metal 3D-design model. This study focused especially into estimating machining time and material usage for machines using laser cutting, punching, forming and shear cutting technologies. Objectives for the study were that estimation model would be capable of giving estimates automatically and fast, estimates should be reasonably accurate in order to use them in practice and it should be possible to train model based on existing production data.
Neural network based model was developed for estimating machining time and material usage. Also different methods were developed for analyzing and measuring machining times and material usages for individual parts from existing production data. These measurements were used for training neural network models. Automatic feature recognition was used for processing 3D-models. This allowed processing 3D-models from different CAD-systems. Cost estimation model was implemented as web-based service which can be integrated into client applications over internet.
Final result was software implementation of the cost estimation model that could be trained by existing production data. Model can give cost estimation simultaneously for different machine types quickly and automatically. Also estimates were evaluated to be accurate enough for practical usage.