A 3D CAD knowledge-based assisted injection mould design system Abstract This paper presents the basic structure of an interactive knowledge-based injection mould design system (IKB-MOULD)。 The basis of this system arises from an analysis of the injection mould design process76178

for mould design companies。 This injection mould design system covers both the mould design process and mould knowledge management。 IKB-MOULD integrates the intelligent design process and knowledge management with many developed interactive tools in a commercial solid modelling software environment。 

Keywords Injection mould design · Knowledge base · Injection mould object representation 

1 Introduction

In recent years the plastic product manufacturing industry has been growing rapidly。 A very popular moulding process for making plastic parts is injection moulding。 The injection mould design is critically important to product quality and efficient product processing。 Mould-making companies, who wish to maintain the competitive edge, desire to shorten both design and manufacturing leading times by automating the design process。 Thus, the development of a computeraided injection mould design system (CAIMDS) is becoming a focus of research in both industry and academia。 Recently published papers show that research in automatic mould design focuses on inpidual components of the mould process。 For example Ong et al。 [1] and Ravi [2] focused their research on the feeding system。 Wang et al。 [3] focused their research on the ejection system。 Others focus their research on the general design。 Most research done on the general injection mould system can be classified into two areas: (a) functional, conceptual and initial mould designs; and (b) algorithms to automate mould generation。

Functional, conceptual and initial designs of the injection mould are applied mainly to the pre-mould design。 Such design involves selecting a suitable mould base, arranging the cavity layout, designing the runner and designing the gate。 The objective is to come up with a large number of very different product ideas for a certain requirement。 Britton et al。 [4] addressed injection mould design from a functional perspective by presenting the Function-Environment-Behaviour-Structure (FEBS) model。 The study fostered a wide range of design alternatives。 Costa and Young [5] proposed a product range model (PRM) to support the reuse of design information in variant design cases。 The general structure of a PRM is defined in terms of design functions linked with sets of design solutions, interactions between potential solutions and knowledge links。 Ye et al。 [6] presented an approach to automatic initial design with algorithms that calculate the cavity number and automatically lay out the cavity。 The initial injection mould design involves extensive empirical knowledge of the structure and functions of the mould components。 Thus, a lot of researchers adopt a knowledge-based approach。 Several knowledge-based systems (KBSs) were developed to advise plastic material selection, capture injection mould part design features, analyse mouldability, automate the mould design process and develop mould design for manufacture。 Examples of such systems are GERES (Nielsen [7]), PLASSEX (Agrawal and Vasudevan [8]), EIMPPLAN-1 (Chin and Wong [9]), CADFEED (Ong et al。 [10]), ICAD (Cinquegrana [11]), IKMOULD (Mok et al。 [12]) and KBS of Drexel University (Tseng et al。 [13])。 However, these KBSs consider only certain aspects of the total design。

As for the automatic generation of an injection mould, a number of theoretical research works were conducted to automatically determine the parting direction, to determine the parting line, to generate the parting surface, to recognise undercut features and to generate the core/cavity。 Ravi and Srinivasan [14] presented nine rules that can be used by the mould design engineer to develop a suitable parting line in the product。