In this paper technical solutions and methods to efficiently share locator data within a product family are proposed.

Locator and datum creators and users in a typical product development and production environment are described. A global car company is used as a basis for discussing the complexity of sharing and reusing locator data in the PLM system.

This paper describes a new module-based way to work with variation simulation models for product family analysis. It also describes methods to optimize tolerances and locator concepts for a part or sub-assembly for all environments where it is instanced simultaneously. Thereby the effects of geometrical variations on the assembly and on the final product can be minimized for all variants in the product family.

A separate section is included to define what attributes can be defined as geometry assurance data.
All methods are evaluated and validated with industrial cases.

The definition of the content of a product platform highly affects the possibility to engineer and produce unique and optimized products to suite the customer needs.

In this paper, generic configurable autonomous sub-systems, Configurable Components (CCs), are used to define platforms that could be used to configure and instantiate product as well as process structures and to optimize instantiated part solutions.

To simulate a realistic industrial environment a configuration demonstrator has been developed and used to perform case studies in order to test the CC concept. The test cases are focused on geometrical interfaces between components. Communication functionality between the demonstrator and a CAT (Computer Aided Tolerancing) tool has been developed to enable automatic optimization of interface concepts during configuration.

In summary – the paper shows that, given variant specific input data, a knowledge based platform definition with high design band-width can be used to configure, engineer and manufacture an instanced product variant.

In this paper, a platform geometry goodness value for a part has been defined. Calculation and simulation methods have been defined and tested to be used in industrial “real-life” environments. Present calculation and simulation methods for assembly analysis in a single product development have been used as a basis. These methods have been further developed and adapted to suit product family development, or platforms. The platform geometry robustness value could be used for optimization of a part or an assembly, by means of geometric variation, not only for one product environment but also for a complete product family simultaneously. This decreases the risk of sub-optimization of part location and assembly concepts. All equations and mathematical connections are described in detail in the paper but, due to the mathematical complexity of 3D modelling, the calculations have been performed in a geometry simulation tool. The proposed theories are based on previous work by Söderberg [1] presented mainly in section 2. The new contribution from this paper is mainly presented in sections 4.2 and 5.

Geometry interactions between geometry interfaces (locating schemes and mating geometries) of parts in a platform environment composed by Conifgurable Components, CCs, are defined and tested in this paper. The interactions are defined as sub-objects within the already defined CC-object. A case study is performed where these CC-objects, with their geometry interfaces and geometry interactions, are defined in a PDM and CAD environment where the functionality has been defined using separate objects exposed in the PDM-structure.

A setup of software consisting of a configurator system, a PDM system and a CAD system, is created to model product platforms, using the CC concept, and to configure and instantiate variants from this platform. A simplified platform is modelled and configuration is done with focus on geometry interfaces and geometry interactions. The work shows, in a conceptual way, that this setup is possible to use within the context of CC modelling and that geometry interactions could efficiently be used to create geometry compatibility between geometry interfaces in the CC objects.

A robust tolerance design concept based on locating schemes (for example, used in the car industry for sheet metal parts) and the methods and tools connected to that concept are applied on a new robot development. A case study of a robot assembly is performed in a new robot development project. The robot is mainly designed using machined parts, and traditionally these parts are toleranced using feature-to-feature based tolerances and linear dimensioning.

The goal of this case study is to see if there are any advantages to using a new approach when designing robots to enable high geometric quality at a low cost.

With limitations, the study shows that it seems possible to effectively use the tolerance design concept in robot development with assemblies consisting of machined parts.

In this paper, a platform geometrical sensitivity value for a part has been defined. Calculation and simulation methods have been defined and tested to be used in industrial “real-life” environments. Present calculation and simulation methods for assembly analysis in a single product development have been used as a basis. These methods have been further developed and adapted to suit product family development, or platforms. The assembly geometrical sensitivity value can be used to predict the effect of tolerance stacking without having data of tolerance sizes available. Using sensitivity calculation in each assembly step gives an indication of the risk of functional failure and non-fulfilled specifications due to tolerance stacking. The platform geometrical sensitivity value could be used for optimization of a part or an assembly, by means of geometric variation, not only for one product environment but also for a complete product family simultaneously. This decreases the risk of sub-optimization of part location and assembly concepts. Using the platform geometrical sensitivity value, the effect of tolerance stacking could be predicted for all assemblies conceptually and the result can be used to dimension specific part tolerances. All equations and mathematical connections are described in detail in the paper but, due to the mathematical complexity of 3D modeling, the calculations have been performed in a geometry simulation tool. Further research needs to be done to establish a proper working procedure using platform geometrical sensitivity value.

Geometrical part robustness is used today as an engineering criterion in many manufacturing companies. The goal is to minimize the effect of geometrical variation by optimizing the locating schemes for the parts. Several methods and tools now exist to support geometrical robustness optimization for parts, but also for assemblies. In this paper the focus is on geometrical decoupling, which is one parameter of geometrical robustness of the different locating strategies in a complete assembly line. A goodness value is proposed that describes the level of geometrical couplings in a complete assembly line together with the part robustness value. By calculating this goodness value it is possible to predict the geometrical sensitivity of a complete assembly line as well as predicting the risk of geometrical variation in the final product. To illustrate the definition of this goodness value, and also the purpose of calculating it, a case study is used where a part of a sheet metal assembly line is described. Several different scenarios (assembly concepts) are applied to clarify the meaning and to validate this definition of the goodness value. The case study shows that the goodness value gives a good indication of the level of geometrical couplings within the assembly line and that this value can be used to evaluate different assembly concepts, with their locating concepts, against each other. The goal is to have a more robust and also geometrically decoupled assembly line which enables root-cause analysis in production, and also optimizes the geometrical quality minimizing the effect of geometrical variation of the final product from the plant.

This paper describes a new design approach that implements the three principles of Set-based Concurrent Engineering by using the concept of Configurable Component modelling. Several case studies has proven the efficiency of Configurable Component modelling as well as the Set-based philosophy, and by combining these two research areas, a computer based modelling of Configurable Component objects is used to support the Set-based philosophy. The approach is demonstrated by a case study that indicates a promising future of combining Set-based Concurrent Engineering with Configurable Component modelling for re-design problems.