MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A

PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN

AND ENGINEERING ENVIRONMENT

MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF

A PLASTIC INJECTION MOLD USING AN INTEGRATED
DESIGN AND ENGINEERING ENVIRONMENT
Reinier van Dijk1, Roberto d’Ippolito2, Gabriele Tosi3,
Gianfranco La Rocca4
1
Ph.D. Candidate, Delft University of Technology, The Netherlands
2
R&D Manager, NOESIS Solutions N.V., Belgium
3
Ph.D. Candidate, Universita’ degli studi di Ferrara, Italy
4
Assistant Professor, Delft University of Technology, The Netherlands

THEME

CAD-CAE Integration

KEYWORDS

Multidisciplinary Design and Optimization, Knowledge-Based Engineering,

Knowledge Management, Plastic Injection Mold Design

SUMMARY

In order to remain competitive with respect to low-cost overseas markets,

domestic moldmakers will have to increase the productivity of their engineers
and maintain high quality standard, while dealing with the problem of an aging
workforce. To increase the competitiveness of the European automotive
industry, the concept of and Integrated Design and Engineering Environment
(IDEE) has been developed within the 6th framework European project
Pegasus - Integrated engineering processing & materials technologies for the
European sector. The IDEE consists of a distributed set of software
applications able to accelerate the design process of plastic injection molds by
means of process automation and multidisciplinary optimization techniques.
The IDEE combines into a single user-friendly platform different technologies
ranging from Knowledge Management (KM) and Knowledge-Based
Engineering (KBE) to Process Integration and Automation (PI&A). Central to
the IDEE framework is a KBE application, which consists of a fully parametric
rule-based mold product model. On the basis of the geometry of the plastic part
to be molded and other (user) inputs, the KBE application can automatically
produce different mold configurations and variants. In this way, product
knowledge is leveraged to automate the repetitive and therefore time-
consuming CAD modelling activities typically encountered in industry.
Next to the mold geometry, the KBE application is able to generate different
reports that serve as input to in-house developed or commercially-of-the-shelf
MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

CAE applications, like Autodesk® Moldflow® Synergy. The resulting

software framework enables the simulation and analysis of the injection
molding process, from which important performance characteristics can be
retrieved. Other disciplinary applications are used to assess the cost and the
environmental impact of the molding process and tooling. All the IDEE
software applications are provided as web services on remote servers according
to the Software as a Service (SaaS) paradigm. Noesis’s Optimus software has
been used to link them into a seamless framework, control their execution and
support all data exchange in an automated fashion over the internet.
The scope of this paper is to describe the architecture and main functionalities
of the IDEE and provide the results of studies performed on representative
plastic parts for the automotive industry. It is demonstrated how the IDEE can
enable multidisciplinary optimization of the injection molding process and find
solutions that best balance manufacturing performance and product quality. By
means of the IDEE many concepts and customized variants can be quickly
generated and evaluated in terms of hours or days instead of weeks or months,
thereby supporting a more efficient development of innovative and complex
parts. The reliability of the evaluation is enhanced by the early use of
computer-based simulations in the design process, and the success rate of the
overall component increased because of the more disciplines included in the
evaluation (tooling design, material processing, cost, environmental impact).
The IDEE can help the designer to virtually prototype and analyze the product
and, eventually, support the “first-time-right” design principle.

1: Introduction

Now more than ever, European mold makers and car component designers are
struggling to compete with low-wage country manufacturers. From one side
the general level of complexity of products increases, while, at the same time, a
higher quality level must be achieved within a shorter lead time, at a price that
is competitive on the global market. Great benefits could be obtained by
streamlining the overall product development process through a better
integration of the supply chain. Nowadays the level of integration/collaboration
between OEM and SMEs is suboptimal. A lot of repetitive activities are taking
a too large portion of the engineering design process at expenses of lead time
and creativity. The brains drain phenomena is taking out of the arena the most
experienced engineers with the result that old errors are repeated and the use of
best practices is not systematically enforced. The effective application of
multidisciplinary design optimization techniques to improve product quality is
not yet reality. In view of developing new technologies to increase the
competitiveness of its automotive sector, the European Community has
sponsored a large collaboration project called Pegasus (Integrated engineering
processing & materials technologies for the European sector). One of the main
Pegasus goals was the development of an new software system, called
Integrated Design and Engineering Environment (IDEE), able to support
 MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

collaborative and integrated development of plastic injected (automotive)

components.
The IDEE is an integrated suite of software modules deployable via web
connections according to the Software as a Service (Saas) paradigm, see figure
1. It can support and accelerate the development of plastic injection molds and
perform analysis and optimization of the injection molding process of car
components by means of computer simulations. The IDEE leverages three core
technologies, namely Knowledge Management (KM), Knowledge-Based
Engineering (KBE) and Process Integration and Automation (PI&A) to achieve
a higher level of integration in the design process and effectively support
Multidisciplinary Design and Optimization (MDO).
Workflow User IDEE framework
Authentication Analysis
Web Services
Internet

Workflow User
Injection molding
Structural sim.
Dynamic sim.
CAD tools
CAE tools
….

Workflow User

Figure 1: IDEE Framework architecture overview

Chapter 2 will introduce the three above-mentioned technologies and elaborate

on some of the main components of the IDEE architecture. In order to
demonstrate the IDEE capability, a proof of concept optimization workflow
will be described in chapter 3. Although the specific IDEE implementation
described here concerns a simple plastic component, the implementation of the
overall framework is still representative of more complex design problems,
such as the cooling system optimization process for the injection of a plastic
car fender, as actually addressed in the Pegasus project (details cannot be
disclosed because of confidentiality agreement). Chapter 3 will also elaborate
on the hybrid optimization methodology that has been implemented within the
IDEE. It will be illustrated how a given cooling system design can be
improved, while keeping computational cost low and shortening the total
design lead time.

2: Integrated Design Engineering Environment

Although the eventual IDEE concept is based on a close integration of three

technologies, viz. KM, KBE and PI&A, this work will focus on all
developments related towards the integration of the latter two aspects. The
MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

following sections explain how different web-services (section 2.1 and 2.2),
located on a remote server at the Delft University of Technology (DUT), are
integrated in a bigger software workflow architecture, which a user can
manipulate and drive on his client computer (section 2.3).
2.1 Knowledge-Based Engineering Application
Knowledge-Based Engineering (KBE) is the ”use of advanced software
techniques to capture and re-use product and process knowledge in an
integrated way” [1]. A KBE application is a software application that uses
Artificial Intelligence (AI) techniques to solve complex problems that would
normally be performed by a person with specific expertise. However, what
distinguishes KBE from a general Knowledge-Based System is the focus on
engineering [2]. As a result solving problems by reasoning must be
supplemented with analysis and computation. On top of that, the field of
engineering comes with the challenge of geometry manipulation and product
configuration. In this respect, “KBE represents an evolutionary step in
Computer-Aided Engineering (CAE) and is an engineering method that
represents a merging of Object-Oriented Programming (OOP), AI and
Computer-Aided Design (CAD) technologies” [3]. For a thorough technology
overview of KBE, see [4]. The objective of KBE is to reduce engineering time
and cost of product development, which is primarily achieved through the
automation of routine design tasks with retention of design knowledge. Besides
automation, KBE systems enable integrated modeling, where multiple
disciplinary views can be derived from a central product model.
Both automation and integrated modeling capabilities make KBE technology
an enabler of Multi-Disciplinary Optimization (MDO). In what is introduced as
a Design and Engineering Engine (DEE), several papers have demonstrated
this [5,6]. Inside the DEE, A Multi-Model Generator (MMG) is used to
describe a product family in terms of common building stones. The MMG has
a generative capability to automatically instantiate geometry and an ability to
create multiple perspectives on the product model. The MMG is connected to
specific disciplinary analyses modules, which process information from the
product model. Analysis results serve as input for the overall MDO process.
This DEE methodology has formed the inspiration for the current IDEE
concept. However, where previous research has focused on aerospace
applications, this research has adopted the KBE approach for the automotive
industry and added particular focus on the integration and automation of
software tools in the overall software workflow.
Genworks International‟s General-purpose Declarative Language (GDL), a
superset of the Common Lisp language, has been used as the KBE platform in
this research [7]. With GDL the Delft University of Technology (TUD) has
developed a KBE application, the so called Mold Multi-Model Generator
(3MG), that holds a parametric description of a generic plastic injection mold
system. Inside this application all rules and knowledge have been captured and
 MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

programmed in order to generalize a mold design into several basic building

blocks or High Level Primitives. In this way a user (human or other software
application) is able to provide several inputs to generate a specific mold design
from a unique product model definition
The 3MG is able to automate a significant part of the geometrically-related
routine design tasks typically encountered in mold design, starting with a raw
plastic part geometry file as an input (IGES, STEP or STL). Automated routine
design tasks include CAD modeling activities and mesh generation. Next to
these geometry-related steps, the 3MG pre-processes the product model to
generate analysis models for several disciplinary analysis tools, a normally
quite repetitive and time-consuming step [3]. The product model has been
integrated with cost analysis, environmental impact analysis and plastic melt
flow simulation software. Section 2.2 will focus on the latter coupling.
The capability of GDL to act as a web server, enabled a straightforward
integration of the 3MG into the IDEE as a web service, located at the TUD.
2.2 Computed Aided Engineering Integration
Autodesk® Moldflow® Synergy 2010 (Moldflow) has been integrated in the
IDEE as COTS software for plastic melt flow simulations [9]. Visual Basic
Scripts (VBS) have been developed to automate the creation of the analysis
model inside Moldflow and to set-up and drive the analyses. Several dedicated
PHP scripts have been used to make the Moldflow application available on a
server at the TUD and enable a modular integration of this software into the
IDEE workflow as a web service. From the Moldflow analysis report, several
performance and quality measures can be automatically extracted for the
overall optimization process. These include cooling and warpage data,
manufacturing cycle time, volumetric shrinkage, clamping force, etc. Figure 2
illustrates the resulting integration of the 3MG application and Moldflow; a
branch of the overall software architecture, as discussed in the next section.

Figure 2: CAD-CAE integration of mold product model and Autodesk Moldflow

2.3 Process Integration and Automation Workflow

Both the 3MG and MoldFlow analysis tool have been integrated into one
distributed, seamless design, analysis and optimization process, using
NOESIS‟ Optimus [10]. Optimus automates the initiation of product and
analysis models, drives simulations, retrieves, processes and stores simulation
results and offers several optimization algorithms. Optimus graphically
MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

represents these essential operations as modular building blocks, which a user

can assemble in different ways to build the workflow required for the design
case at hand. The workflow that drives the optimization is given in Figure 3
and will be discussed in detail in chapter 3.

Figure 3: Workflow developed for the cooling system optimization

A dedicated module that has been developed for the purpose of this research is
the communication capability between Optimus and the 3MG application.
Communication is based on the Hypertext Transfer Protocol (HTTP). Using
the „POST‟ request method, Optimus specifies all design aspects to 3MG in the
body of the request. The response to this request is then automatically handled
and interpreted by Optimus. HTTP requests are for example used to create a
new mold instance, or (on subsequent iterations) to update one or several mold
parameters. File transactions use the (Secure) File Transfer Protocol, (S)FTP.

3: IDEE for cooling system design and optimization

3.1 Optimization problem definition

In order to demonstrate the IDEE‟s capability to support MDO, a test case has
been defined. This involves the optimization of a conformal cooling system
design for a simple plastic part (figure 4).
 MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

Figure 4 – Plastic part with conformal cooling system

Being the most time consuming step in a typical injection cycle, the first
objective is to minimize cooling time. Both the number and diameter of the
cooling channels are main (input) variables (Table 1).
Name Min Max Definition Unit
Diameter of cooling channels 2 8 Integer with step of 2 mm
Number of cooling channels 2 8 Integer with step of 1 Adimensional
Table 1 – Input variable definition

However, a fast and non-uniform cooling process can negatively affect the
quality of the final product (e.g. introduce warpage). Therefore the final
objective is to find the optimal balance between performance and quality,
hence by minimizing both cooling time and warpage. Table 2 shows all the
(output) design variables taken into account during the optimization process.
Name Unit Name Unit
Cooling time s Pumping power W
Filling time s Bulk temperature K
Packing time s Part surface temperature K
Minimum X displacement mm Cavity surface temperature K
Maximum X displacement mm Shear rate 1/s
Minimum Y displacement mm Pressure drop over cooling circuit Pa
Maximum Y displacement mm Peak pressure Pa
Minimum Z displacement mm Maximum injection pressure Pa
Maximum Z displacement mm Shrinkage criterion Adimensional
Clamp force N
Table 2 – Output variable definition

The minimum and maximum displacements along the primary axes, are the
primary results used to assess the geometrical deformation of the part during
the cooling process due to shrinkage and distortion. They express the
difference between the dimensions of the mold cavity and the molded part,
once extracted. The other results shown in Table 2 are needed to control
different process parameters or are considered useful to increment the
knowledge company database.
3.2 Software Workflow
The resulting workflow used for this optimization is schematized in figure 3.
As first step all inputs are defined, consisting of “Configuration” and
“DesignParameters”. The “Configuration” box contains information like
project name, the addresses and ports of the different web and proxy services
and the folder to store the project file. The workflow can be easily set-up to
work on different user/client computers by simply editing these configuration
parameters. The “DesignParameters” box holds the definition of the design
variables. These variables parameterize different aspects of the analysis process
and allow the designer to investigate different configurations. The configured
MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

workflow also allows to study the sensitivity of the final configuration towards
parameter variations such as geometric tolerances and temperature variations.
After the input definition, a block of HTTP „Post‟ requests follows instructing
the 3MG application to create a mold instance based on the variables values
defined in the input block, to update the mold instance after a change of the
variables values, to export part and mold geometry and finally to write a subset
of VBS scripts necessary for the Moldflow analyses. The use of individual
requests over one global request give several advantages, including:
 simplicity in writing single, short requests and reduced likelihood of
mistakes due to a compilation of a long and complicated global request;
 modularity/scalability, resulting in the situation where individual request
can be added or removed for different, customized workflows.
Optimus automates all data exchange between web services or between service
and client computer (for example geometry and scripting files) using (S)FTP
connections. Intermediate results are locally copied on the user computer and
form a precious knowledge base of results for the organization. For the purpose
of user-friendliness, the transfer and copy of data have been captured in two
elementary modules, the „copy‟ and „move‟ classes, that can be used as generic
descriptors of specific FTP operations and are offered as a plugin to the
Optimus product. The „move‟ class of commands is used to store files in the
user computer by deleting it from the remote one, thus preventing undesired
access to this data by web service users, thus securing data.
The workflow in figure 3 uses five FTP operations, two to get results and
scripts from the 3MG and three to send these files to the right Moldflow
directories. After this, two HTTP requests take care of the simulation process.
The first request creates the Moldflow model, the second starts the simulation.
At the end of the simulation, all results are stored on the user computer and
values of interest (table 2) are automatically extracted using elementary data
management operations.
3.3 Optimization methodology
The plastic product used for this demonstration case is quite simple and
therefore requires little computer resources. Moreover, as Table 1 points out,
there are only 28 possible product configurations. For this reason, it is possible
to analyze them all and select the optimal one. However, anticipating more
complex industry relevant design problems, the large number of configurations
and computation time would invalidate such an approach. In this case a second
approach is used, based on hybrid optimization techniques to reduce
computational time. The former “brute-force” approach is still used to validate
the latter.
The hybrid optimization methodology used in this work consists of the
combination of Design of Experiments (DOE), Response Surface Modeling
(RSM) and a global optimization algorithm, discussed next.
 MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

3.3.1 Design of Experiments

The first methodology step is the DOE construction. The DOE is a collection
of statistical techniques for planning, analyzing and interpreting sets of
experiments, in order to investigate the correlation between inputs and outputs.
In this work one of the most commonly-used random DOE techniques, called
Latin Hypercube Designs (LHDs), has been used. Particularly a LHDs of 15
experiments (instead of 28) has been built, which needed one hour of
computation time (4 minutes for each simulation). From a careful analysis of
DOE warpage results it was observed that the variability of the minimum
displacement along X direction and the maximum displacement along Z
direction are of one order lower than the other displacements. For this reason
they have been neglected in subsequent steps.
One of the main results of a DOE is a matrix with correlation coefficients
between input and output quantities. A correlation coefficient of 1 indicates a
perfect direct correlation; on the other hand, a correlation coefficient of -1
indicates a perfect inverse correlation; 0 indicates the absence of correlation
(i.e. independence). It is important to note that the correlation coefficients are
average linear correlation coefficients, meaning that they are computed as an
average over the entire set of experiments used and thus express global linear
trends. As a consequence, they cannot provide local information about a
specific region of the design space, nor can they give information about
quadratic or higher order correlations. However, this is not a limitation as in
most engineering cases higher order correlations or local correlations are either
very difficult to model or not useful in optimization strategies, respectively.
Figure 5a shows the correlation between different displacements. It is evident
that all displacements have a strong correlation (in this case the minus sign
between minimum and maximum displacement indicates a direct correlation,
because the max. displacements are positive numbers while the min.
displacements are negative numbers). Hence, a minimization or maximization
of one displacement, also induces a minimization or maximization of the
others. For this reason, and considering the fact that the variability of the
maximum displacement along the X axis (MaxDispX) is the highest, it has been
decided that the minimum warpage objective can be reduced to the
minimization of MaxDispX.
MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

(a) (b)
Figure 5 – Correlation between the displacements taken into account in the
analysis (a) and correlation matrix between input and output (b)

As mentioned above, the other objective is to minimize cooling time

(CoolingTime). Figure 5b indicates correlations between the input and the
output of interest. It can be noted that the diameter has a very low influence on
the output values, while the number of cooling channels (Cooling-Nr) has a
strong inverse correlation with both the two outputs. This means that the
CoolingTime and the MaxDispX are cooperative objectives and that the optimal
solution is the one with the highest number of cooling channels.
3.3.2 Response Surface Modelling
DOE have formed the basis for a first investigation of the general problem. But
for a more exhaustive study, RSM is used. In general, RSM is a method to
construct global approximations or interpolations of the whole system behavior
just based on the results calculated for a limited number of points in the design
space. For the present case, the RSM technique is based on the 15 results from
the Latin Hypercube DOE. The current study uses Linear Radial Basis
Functions (Linear RBF) to interpolate the RSM . This method is one of the best
interpolation techniques when there are only few experiments and when inputs
are discrete, as in the studied case.
Figure 6a shows the 3D response surface for MaxDispX, related to the diameter
and number of cooling channels. It is evident that the higher the number of
cooling channels, the lower the warpage along the X axis. This is in agreement
with the DOE correlation results. However the response function serves two
other important aspects. Firstly it does not average out correlations, hence the
designer is able to evaluate the influence of varying diameters on different
cooling system configurations. Secondly, the surface highlights the presence of
a local minimum near the configuration composed of four channels with a
diameter of 6mm. This last result, hardly predictable with other methods, could
be very helpful for further company evaluation in which also cost and the
environmental impacts are considered.
 MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

Figure 6b shows the cooling time response function. The surface confirms a
strong influence of number of cooling channels on cooling time, while the
diameter has a very low influence; moreover this low effect of the diameter
decreases when the number of channels grows. Finally, a doubling of cooling
channels from 4 to 8 has less influence than a similar doubling from 2 to 4.
This behavior could be of a great interest in an analysis that also considers the
manufacturing complexity and the cost of the cooling system.

(a) (b)
Figure 6 – Response surfaces for the MaxDispX (a) and cooling time (b)

3.3.3 Global Optimization

The last step researches the true optimum, i.e., the combination of design
variables that yields the minimum cooling time and warpage. For this purpose
two single objective optimizations have been performed using a Differential
Evolution (DE) algorithm applied to the response function. The DE is a
heuristic method for global optimization and needs more experiments (and so
more computational time) with respect to the local optimization algorithm.
However, because the DE algorithm is applied to the response surfaces, which
are mathematically defined, it only requires a few seconds to evaluate.
Input-Output Optimum based Verification Optimum based on
on RSM experiment of RSM “brute-force”
Optimum method
Diameter [mm] 4 4 2
Cooling number [-] 8 8 8
Max X warpage [mm] 0.55063 0.55063 0.54532
Cooling time [s] 15.41 15.41 15.33
Table 3 – Optimization results

The two single objective optimizations have reached the same configuration
that minimizes both the cooling time and the warpage. This configuration is
presented on table 3 (second column). A final simulation on the optimal
configuration (8 channels of 4 mm) confirms the accuracy of the response
surface method (third column). Finally this optimization methodology is
compared to an optimization based on all possible configurations (at a
MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

computational cost of almost two hours, instead of one). Column 4 shows that
the optimal configurations slightly differs, however with negligible differences.
It is therefore safe to assume that the hybrid methodology is reliable.

4: Conclusions and future work

The IDEE workflow developed in this work automates modeling and

simulation of an injection molding process. Software applications are
connected to the IDEE using the „Software as a Service‟ paradigm.
Communication is performed using HTTP requests and data transfer via SFTP.
In this way the workflow execution is modular, fast and secure.
The IDEE is an important platform supporting the designer during product
development; it enables a thorough exploration of the design space, while
reducing design time. With the developed workflow and the use of a hybrid
optimization methodology, it has been possible to reliably obtain an optimal
cooling system configuration that minimizes cooling time and warpage, at half
the computational cost of the more basic “brute force” optimization method.
Moreover some important results regarding the interaction between the input
and output variables have been obtained, that prove useful for future work and
decisions.
While this paper has illustrated the ability of the IDEE just for a simple case,
the flexibility and scalability of the system supported by the use of smart
optimization techniques allow designers to tackle the analysis of more complex
objects and problems.

5: Acknowledgements

Financial support to this research has been provided by the European

Commission under the 6th Framework Programme for the research project
“Pegasus – Integrating engineering processing and materials technologies for
the European automotive sector” (Contract Number NMP2-CT-2006-026673).
REFERENCES
1. STOKES, M. - Managing Engineering Knowledge, MOKA: Methodology
for Knowledge-Based Engineering Applications. ASME Press, 2001
2. MILTON, N.R. - Knowledge Technologies, Polimetrica, 2008
3. CHAPMAN, C B, PINFOLD, M - The application of a knowledge based
engineering approach to the rapid design and analysis of an automotive
structure, Advances in Engineering Software, Vol 32, Issue 12, pp. 903-
912, 2001
4. COOPER, D, ROCCA, G LA - Knowledge-based Techniques for
Developing Engineering Applications in the 21st Century, 7th AIAA
ATIO Conference, AIAA, Belfast, Northern Ireland, 2007
5. ROCCA, G LA, TOOREN, M J L VAN - Enabling distributed
multidisciplinary design of complex products: a Knowledge Based
 MULTIDISCIPLINARY DESIGN AND OPTIMIZATION OF A
PLASTIC INJECTION MOLD USING AN INTEGRATED DESIGN
AND ENGINEERING ENVIRONMENT

Engineering approach, Journal of Design Research, Vol 5, No. 3, pp. 333-

352, 2007
6. LAAN, A H VAN DER, TOOREN M J L VAN - Parametric modeling of
movables for structural analysis, Journal of Aircraft, Vol 42, pp. 1606-
1614, 2005
7. COOPER, D - Knowledge Base Techniques in Web Applications: a
Tutorialfor Genworks‟ GDL/GWL, http://www.genworks.com, 10-1-2010
8. KAZMER, D O – Injection Mold Design Engineering, Hanser, 2007
9. AUTODESK - Autodesk Moldflow Insight Brochure,
http://usa.autodesk.com, obtained: 10-1-2010
10. NOESIS SOLUTIONS – Optimus Details,
http://www.noesissolutions.com/Noesis/node/19, obtained: 10-1-2010