On the Development of a Collaborative Knowledge Platform for

Engineering Sciences

Jonas Jepsen

, Arthur Zamﬁr

, Brigitte Boden

, Jacopo Zamboni

and Erwin Moerland

Institute of System Architectures in Aeronautics, German Aerospace Center, Hein-Saß-Weg 22, Hamburg, Germany

Keywords:

Knowledge-Based Engineering, Semantic Web Technologies, Web Application, Corporate Amnesia,

Knowledge Management, Collaboration, Knowledge Platform, Codex.

Abstract:

Knowledge is undoubtedly one of the most important assets for all organizations. Losing knowledge that once

was considered part of an organization always poses a problem. When knowledge is concentrated on single

individuals, as is often the case in research environments, this is particularly difﬁcult to avoid. This is also

true for software developed by individuals such as Knowledge-Based Engineering (KBE) applications. In

this paper we show how the Codex Framework, which is based on Semantic Web Technologies (SWT), can

be used for creating, maintaining and inspecting formalized knowledge in a way which also fosters its reuse

and execution. We show use cases where Codex was applied and discuss the advantages and shortcomings

compared to a conventional Object-Oriented Programming (OOP) approach. From these experiences we then

draw conclusions on why the adoption rate of Codex is below expectation and present a newly developed

collaborative knowledge platform which is designed to overcome the identiﬁed challenges.

1 INTRODUCTION

Organizational knowledge, the valuable pool of infor-

mation and insights within a company, serves as the

foundation for success in the fast-paced world of busi-

ness (Nonaka and Takeuchi, 1995). However, an of-

ten overlooked and consequential challenge that com-

panies face is corporate amnesia — the gradual ero-

sion or forgetting of this vital institutional knowledge,

experience, and lessons learned over time (Dalkir,

2005). As employee turnover rates rise and industries

undergo technological transformations, organizations

become increasingly fragile to losing their collective

memory. This loss hinders their ability to make in-

formed decisions, replicate past achievements, and

avoid repeating previous failures, ultimately impact-

ing their overall performance and long-term viability.

In this section we discuss causes and effects of cor-

porate amnesia in research environments as well as

means to improve its prevention.

https://orcid.org/0000-0002-3307-1978

https://orcid.org/0000-0001-7579-9628

https://orcid.org/0000-0002-2326-2762

https://orcid.org/0000-0003-3207-4138

https://orcid.org/0000-0003-4818-936X

1.1 Corporate Amnesia in Engineering

Research

In engineering research environments, avoiding cor-

porate amnesia poses a unique challenge. Individu-

als typically enter these environments after complet-

ing their master’s degrees and often continue their

stay to conduct PhD studies. Moreover, researchers

frequently dedicate their efforts to highly specialized

topics, resulting in the accumulation and creation of

a substantial amount of knowledge throughout their

thesis work. As a result, the risk of corporate amne-

sia increases due to the transient nature of researchers

and the signiﬁcant personal expertise they develop

within their speciﬁc knowledge domain. While a con-

densed portion of their knowledge is eventually doc-

umented in the form of a PhD thesis, it is crucial to

recognize that a considerable amount of knowledge

remains undocumented or is not reusable unless high

effort is involved.

Nowadays, PhD theses in the ﬁeld of engineer-

ing often include the development of some kind of

software. Whether it is for demonstrating a new al-

gorithm, method or methodology, performing simula-

tions or analyses, or creating a design tool whose ex-

tent can vary from using simple handbook methods to

fully ﬂedged Knowledge-Based Engineering (KBE)

208

Jepsen, J., Zamﬁr, A., Boden, B., Zamboni, J. and Moerland, E.

On the Development of a Collaborative Knowledge Platform for Engineering Sciences.

DOI: 10.5220/0012189600003598

In Proceedings of the 15th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2023) - Volume 2: KEOD, pages 208-215

ISBN: 978-989-758-671-2; ISSN: 2184-3228

applications (La Rocca, 2012). This software is often

written, maintained and run by a single person. Al-

though it contains a set of explicit instructions, operat-

ing the software requires a deep understanding of the

underlying assumptions which are often only present

as tacit knowledge gained while conducting the PhD.

The ease of extracting knowledge from software

or adapting it to meet new requirements varies de-

pending on how it was developed. In situations where

the original creator is no longer accessible, these tasks

can become highly challenging. As a consequence, a

signiﬁcant portion of the knowledge generated within

the software is effectively lost when an individual de-

parts, resulting in a notable case of corporate amnesia.

From our experience the quality of software writ-

ten in research institutions covers the complete spec-

trum from legacy codebase with poor architectural co-

hesion, to a professionally crafted architecture and

modular set of well written software applications.

In between these extremes are many different lev-

els of software quality, but the majority of scien-

tists would rather focus on developing their domain

knowledge than mastering the craft of software de-

velopment (Kelly and Sanders, 2008).

Knowing that there are multiple angles from

which corporate amnesia should be addressed by or-

ganizations, this paper focuses on technical solutions

for solving two of the identiﬁed challenges — a suit-

able way for knowledge representation and means

to reduce software development related tasks not di-

rectly contributing to doing science.

1.2 Preserving Organizational Memory

There are multiple approaches to knowledge manage-

ment for keeping existing knowledge inside an orga-

nization and to support the creation of new knowl-

edge. The authors of (Nonaka and Takeuchi, 2021)

group knowledge in two major categories, explicit

knowledge and tacit knowledge. While explicit

knowledge is formal and systematic, like words, au-

dio recordings and images, tacit knowledge is far

more difﬁcult to express and resides “within the heads

of knowers” (Dalkir, 2005). Depending on the con-

text, other types of knowledge are also addressed. The

authors of (Li and Gao, 2003) argue that although of-

ten used synonymously, implicit knowledge should be

distinguished from tacit knowledge since it refers to

knowledge that can be articulated but for some rea-

sons is not. It is crucial to acknowledge that all three

mentioned types of knowledge, with their respective

strengths and weaknesses, are essential for knowledge

creation and fostering innovation. Although being a

signiﬁcant asset in driving innovations within compa-

nies, methods for retaining tacit knowledge is outside

the scope of this paper. In this paper, our focus is

on preserving organizational memory by encouraging

scientists to externalize a signiﬁcant portion of their

explicit and implicit knowledge into a formalized and

reusable model. In order to achieve that, each indi-

vidual needs to be provided with an incentive to do

so. The presented research attempts to free scientists

from the burden of traditional software development.

In Section 2, we show how the Codex framework

has been used for the development of KBE applica-

tions and discuss its advantages and disadvantages.

Section 3 deals with the transition from Codex frame-

work to the collaborative knowledge platform and

presents the current state of the platform. In Section

4 we conclude with a summary and an outlook on the

next steps.

2 SEMANTIC WEB

TECHNOLOGIES TO THE AID

As arguments and dispute are integral to scientiﬁc

progress, an effective knowledge structure for knowl-

edge modelling should adopt a ﬂexible and open-

minded paradigm that allows for different opinions

about the same subject to exist. By making use of

Semantic Web Technologies (SWT) such as the Re-

source Description Framework (RDF) and the Web

Ontology Language (OWL) as the base modelling

language, it even supports explicitly interconnecting

models from different domains or disciplines, which

is one of the challenges for KBE applications identi-

ﬁed by (Verhagen et al., 2012).

Schema-free languages such as RDF and OWL

are an alternative to a central schema-based data ex-

change format such as CPACS (Alder et al., 2020).

These languages provide a more ﬂexible approach,

without the need for building a consensus on a single

data structure. This ﬂexibility promotes interoperabil-

ity between different domains since data can be rep-

resented, extended and shared without strict schema

constraints. RDF and OWL enable the representation

of data with rich semantics and relationships. This

promotes better understanding and integration of data

across domains. Moreover, collaboration can beneﬁt

from a shared understanding of the meaning and con-

text of data, leading to more effective data integration

and knowledge discovery.

The authors of (Zamboni et al., 2020) elaborate

how a generic language such as RDF can be used

to overcome the challenges of collaboration in multi-

domain environments. The combination of a highly

abstract base language and more expressive additions

On the Development of a Collaborative Knowledge Platform for Engineering Sciences

209

such as OWL allows easier cross-domain knowledge

integration while simultaneously enabling indepen-

dent and effective capturing of the precise meaning

of each individual domain.

In the next section we give a brief introduction to

the SWT-based Codex framework. We discuss spe-

ciﬁc characteristics of the different modules and give

examples on how they are used. We then summa-

rize the experiences made and discuss the beneﬁts and

shortcomings of the framework.

2.1 The Codex Framework

The COllaborative DEsign and eXploration (Codex)

framework is a software framework for the devel-

opment of KBE applications based on SWT. At its

core, Codex makes use of SWT as a domain neutral

knowledge representation in order to naturally sup-

port the integration of knowledge from multiple do-

mains. SWT, especially RDF and OWL, are explic-

itly designed for linking knowledge and allow users to

combine data from multiple domains. Codex enables

users to model their knowledge in domain-speciﬁc

knowledge bases which, if well designed, can be eas-

ily integrated, reused and build upon.

This knowledge base is made up of several do-

mains, each of which describes a speciﬁc aspect or

component of the aircraft model, such as the mission

to be performed or its wings and tail. By using the for-

malized knowledge the tool computes aircraft masses,

geometrical and performance properties as a function

of external requirements in an automated way.

At the heart of codex is the codex-semantic mod-

ule, an implementation of semantic modelling using

RDF and OWL, as well as additional syntaxes such

as Manchester Syntax and Functional Syntax. It al-

lows users of the framework to efﬁciently model their

knowledge using Domain Speciﬁc Languages (DSL),

and to deﬁne and execute logical rules such as those

deﬁned by the OWL 2 Proﬁles speciﬁcation (World

Wide Web Consortium [W3C], 2012). Since Codex

is a framework for developing KBE applications, in

addition to the core module for semantic modelling,

Codex includes multiple other modules, each provid-

ing its own functionality needed for KBE applica-

tions. The three most important of these modules are:

codex-rules provides an interface to a production-

rule engine. Rule engines such as Drools (Red Hat,

2021) make it easy to formalize and execute rules that

systematically “produce” connected entities within a

model in a declarative way. Each rule has a Left-

Hand-Side (LHS) which is basically a graph query

similar to SPARQL SELECT. This LHS is compared

against the dataset and, if not conﬁgured otherwise,

rules are executed once on every match in the corre-

sponding model. When triggered, these rules can add

new statements to the model or execute user deﬁned

code.

codex-parametric adds the capability to deﬁne

mathematical constraints which can then be evaluated

by a solver. It serves as a means to describe, analyze,

and solve a parametric system through a DSL imple-

mented speciﬁcally for this purpose. For solving the

system of equations codex-parametric implements a

Solution Path Generator which analyses the system of

equations and groups them into Strongly Connected

Components (SCC) (B

olling, 2015). SCC have cyclic

dependencies and can only be solved simultaneously.

By looking at the amount and individual sizes of the

SCC, we can get an idea of the problem complexity.

In addition to problem complexity, this module analy-

ses whether a set of constraints is either well deﬁned,

over determined, or underdetermined. When well de-

ﬁned it can solve the system of equations and return

results for the missing parameter values.

codex-geometry provides the user with a large

number of Computer Aided Design (CAD) function-

alities which are essential for the development of

KBE applications (La Rocca, 2012). It deﬁnes its

own DSL to allow users to model their geometry efﬁ-

ciently.

Although each module implements its own DSL

to provide a simple and intuitive way for modelling

the corresponding knowledge, in the current version,

it is still necessary to have a good understanding of

programming.

2.2 Experiences with KBE Applications

During the past two years the Codex framework has

been applied in different use cases. In this section

we brieﬂy present three of them and discuss the bene-

ﬁts of using the Codex framework over more conven-

tional implementations.

Fighter Design

The authors of (Zamboni et al., 2022) give a great

example on how a complex knowledge base that de-

scribes several aircraft systems, components and their

relationships is effectively combined to provide a tool

for the conceptual design of combat aircraft (Fig-

ure 1).

Compared to a previous Python-based imperative

implementation the Codex framework offers many

advantages when it comes to reusing the knowledge

base. The higher degree of granularity provided

by the splitting of component’s model and rule-sets

KEOD 2023 - 15th International Conference on Knowledge Engineering and Ontology Development

210

allows users to precisely control which knowledge

should be used.

Fuel System Veriﬁcation

The authors of (Boden et al., 2022) present how the

Codex framework and particularly codex-geometry

is used in combination with codex-rules to generate

CAD geometry for aircraft fuel systems (Figure 2).

This geometry is then used for validating geometrical

requirements using codex-rules.

The use of the Codex framework in this approach

provides the user a great deal of ﬂexibility when cre-

ating a design and makes it easy to extend the system

with custom requirements and rules.

Liquid Hydrogen Fuel Tank

Another example for the use of the Codex framework

is a KBE tool for the development of liquid hydrogen

fuel tanks for aircraft. It was used for research on

liquid hydrogen storage design trades for short-range

aircraft (Burschyk et al., 2021).

The declarative approach of Codex framework al-

lows the user to decide freely whether the design

should be performed from the inside out, from out-

side in or even a combination of both.

2.3 Towards Wider Adoption

Our experiences from using the Codex framework

indicate a number of advantages for using a more

declarative approach based on SWT for the develop-

ment of KBE applications. In addition to advanced

modelling capabilities, such as support for parame-

ter metadata covering dimensions, units, numerical

boundaries and computational tolerances, the Codex

framework presented also simpliﬁes extensibility. By

separating components and rules it improves reusabil-

ity of existing knowledge. For research software,

where use cases might change quickly, the declara-

tive approach is particularly useful as it makes the tool

signiﬁcantly more versatile. The ﬂexibility to easily

Figure 1: Combat Aircraft Design.

switch inputs and outputs allows KBE tools to be used

in a variety of ways. As long as a user provides values

for a sensible set of parameters, these parameters can

be freely chosen.

Despite the listed signiﬁcant advantages, we have

observed some barriers in the adoption of the ap-

proach. In its current implementation, users still need

to be familiar with the Kotlin programming language

to use the Codex framework efﬁciently. Users still

need to spend time on software related tasks, which

however also holds true for the approach nowadays

mainly taken by the engineering community which

bases on the application of conventional OOP meth-

ods to create KBE tools. Users still need to invest

time in software-related tasks, a requirement that also

applies to the prevalent approach in the engineer-

ing community today, where conventional Object-

Oriented Programming (OOP) methods are used to

develop Knowledge-Based Engineering (KBE) tools.

Finally, debugging a set of declarative rules can be

more difﬁcult than debugging imperative code. When

the execution order of rules is determined by a rather

complex algorithm there is no clear execution path to

follow when debugging.

With this indication of the usefulness of SWT for

the development of KBE applications and the main

barrier of requiring to spend time on software de-

velopment tasks, it became apparent that Codex has

to be taken one step further. In the next section we

present the concept and implementation of a collabo-

rative knowledge platform.

3 A COLLABORATIVE

KNOWLEDGE PLATFORM

Despite the demonstration of the value of Codex

framework for the development of KBE applications,

as indicated by the examples given in Section 2.1,

the number of people adopting their developments to

the framework was very limited. All adopters were

Figure 2: Fuel System Design.

On the Development of a Collaborative Knowledge Platform for Engineering Sciences

211

people with an afﬁnity towards software development

and were either willing to learn, or already familiar

with the Kotlin programming language. In addition,

the developers also had at least a basic understanding

of SWT. Although there are parallels between OOP

and modelling in OWL, the learning curve can still be

quite steep when switching from one to the other. Es-

pecially understanding the implications of the open

world assumption and non-unique naming assump-

tion can be difﬁcult at ﬁrst.

Recognizing that scientists prefer to engage in sci-

entiﬁc work rather than software development (Kelly

and Sanders, 2008), a web application was initiated

to relieve users from the traditional software develop-

ment demands. In order to achieve that, we decided

to make even further use of SWT and also describe

parametric and production rules in RDF. While this

insight is signiﬁcant, it is also crucial to enable the

”power users” who favour working with code over a

GUI to do so. This led us to the implementation of

the web-based Codex platform, described in the cur-

rent section. Even though the original intent is creat-

ing a platform for developing KBE applications, we

decided to build the core of the platform in a way

that makes as few assumptions on its use as possible.

Users should have full access to the core platform via

a web API, enabling them to implement their own ap-

plications on top of the platform. The core of the col-

laborative knowledge platform serves as the backbone

for client applications, such as graphical user inter-

faces, without prescribing their speciﬁc use-cases. It

provides the community with a platform for effective

knowledge collaboration as well as the ability to ex-

tend the capabilities of that platform without the need

of maintaining their own server infrastructure.

3.1 Platform Concept

In (Zamboni et al., 2022) and (Boden et al., 2022)

we established that SWT are an excellent medium for

precise knowledge capture and cross-domain knowl-

edge exchange. Despite these advantages, however,

it is not widely used outside a few select domains

that rely mostly on logical inference capabilities. By

considering the analogy of RDF as a ”protocol for

knowledge exchange,” to TCP/IP as a protocol for in-

formation exchange (Internet Engineering Task Force

[IETF], 1981), we can deduce a hypothesis that might

help explain its limited adoption despite its clear ben-

eﬁts. Protocols such as TCP/IP are extremely use-

ful and at the core of what made the Internet suc-

cessful and widely adopted. However, most peo-

ple who actually use the Internet on a daily basis

are completely unaware of its inner workings, and

in particular have no idea how TCP/IP works or that

it even exists, nor should they. Such infrastructure

technology is essential for a system to work but it

fades into the background as more abstraction layers

are built on top of it (International Organization for

Standardization [ISO], 1994) and what remains vis-

ible are domain-speciﬁc and easy to use interfaces

such as web, desktop, or mobile applications where

the individuals can focus on their task at hand in-

stead of thinking about TCP/IP packets. Such lay-

ers of abstraction are mostly absent from the RDF

“knowledge-protocol”. Of course, there are exist-

ing applications with graphical user interfaces, such

as Prot

e (Musen, 2015) but in order to effectively

use these applications you still need to be familiar

with the “knowledge-protocol” layer and understand

RDF and OWL. Therefore, our hypothesis for the lack

of wider adoption of SWT in engineering design is

the result of missing high-level abstraction layers that

will make the low-level protocol layer fade into the

background and make effective knowledge collabora-

tion based on SWT more accessible to non-experts.

With this hypothesis in mind, the aim was to cre-

ate a collaborative knowledge platform utilizing open

SWT standards. The expectation is that the platform

will improve cross-disciplinary knowledge exchange

upon its adoption by a dynamic community, espe-

cially in engineering sciences, such as the complex

ﬁeld of aerospace engineering. In order to achieve

these goals and overcome the challenges described

above, the platform must meet several essential re-

quirements. The following is a condensed selection

of the top-level requirements that comprise the key

aspects of a collaborative knowledge platform.

Managing & Publishing RDF models Users

MUST be able to create, read, update, delete and

publish RDF models in a persistent data storage.

Access Control Capabilities User access to mod-

els MUST be controlled and manageable by users

with corresponding permissions.

User Management There MUST be capabilities to

manage (identify, add, update, and remove) users.

Authentication There MUST be capabilities to au-

thenticate users.

Document Resource Storage There MUST be a

document storage system for storing and sharing

arbitrary data ﬁles (as resources, including pub-

lished models).

Real-Time Synchronization Updates from users

working on the same model SHALL be synchro-

nized. This implies users inspecting a model are

informed of updates from other users in real-time.

KEOD 2023 - 15th International Conference on Knowledge Engineering and Ontology Development

212

Figure 3: Collaborative Knowledge Platform.

Figure 3 illustrates that these six pillars, based on

a persistent storage, are the foundation for the col-

laborative knowledge platform and can be used for a

large variety of client applications. In Section 3.3,

we provide an example of how these basic features

can be utilized to implement a client for collabora-

tive knowledge modelling, acting as the aforemen-

tioned higher-level abstraction layers on top of the

RDF ”knowledge-protocol”.

This is largely in contrast to the Codex framework

implementation where all the models are stored in

memory only, unless explicitly exported to ﬁle. When

deciding to directly store all the knowledge in RDF

models, the exchangeability and reusability is signiﬁ-

cantly increased.

In the next section, we give a more detailed

overview of the architecture and implementation of

the core of our extensible collaborative knowledge

platform.

3.2 Platform Implementation

Based on the listed requirements, the groundwork for

the collaborative knowledge platform has been estab-

lished. The resulting system provides a simple but

powerful foundation for all additional application fea-

tures. It improves extensibility by focusing on a lim-

ited set of basic features and then implementing ad-

ditional features as client applications of this basic

service. This includes the user interface as well as

the features connected to KBE as described in the use

cases from Section 2.2. Every service that is build on

top of the platform core could be replaced by a custom

implementation, thus providing the community with

an environment to build their own tools. Thanks to

the open approach, the community can construct tools

whilst making use of already existing tools, much like

software libraries utilizing other software libraries.

This implies that once domain knowledge is imple-

mented and shared, others can easily leverage it to

develop new domain-speciﬁc tools.

Figure 4 illustrates the current architecture of the

collaborative knowledge platform. Looking at the re-

quirements from the previous section one can map the

responsibilities as follows.

The platform consists of a Gateway that is used to

deliver the front-end application to the user’s browser.

It also delegates requests to the Real-Time Service

(RTS), Core Backend Service (CBS) and Document

Resource Service (DRS) as needed.

The Core Backend Service (CBS) is at the cen-

ter of the Codex platform. It implements user man-

agement, access control, and RDF model registration.

Access control is implemented on a very fundamen-

tal level. There are four types of access which can

be granted for each model, which are read, write,

delete, and manage. More complex types of authen-

tication can always be implemented by client appli-

cations which then act as a delegator for these fun-

damental permissions. Published models can either

be stored in a separate dataset on the server or in the

DRS. While storing published models in a separate

dataset has the advantage of allowing SPARQL re-

quests on those models directly, saving them in a con-

tent addressed DRS can provide several useful char-

acteristics, such as verifying data integrity, guaran-

teeing immutability, and providing resilience to data

loss. The latter aspect is especially useful since one

of the challenges of SWT is to assure that a resource

can be long-term locatable (dereferenceable).

The Real-Time Service (RTS) provides the abil-

ity to synchronize data across multiple users in (near)

real time. It uses a modern approach to conﬂict reso-

lution (Nicolaescu et al., 2016) that works in both cen-

tralized and decentralized (peer-to-peer) collaborative

environments. Our implementation also supports the

continuation of work when the network connection is

lost, and the changes can be synchronized later in a

conﬂict-free manner.

Authentication is outsourced to an Authentica-

tion Server (AS).

In the next section the ﬁrst client application

which is implemented using this platform is pre-

sented.

Figure 4: Codex Server Architecture.

On the Development of a Collaborative Knowledge Platform for Engineering Sciences

213

Figure 5: Resource view of the web application.

3.3 A User Interface for Knowledge

Modelling

As a ﬁrst client implementation a Knowledge Mod-

elling Interface for RDF and OWL models is currently

being developed. It serves as a proof of concept for

the platform and its extensibility, and also as the base

for the migration of the additional KBE features de-

scribed in Section 2. The client’s scope is reduced in

comparison to the Codex framework and so far, cov-

ers the modelling aspect and some client speciﬁc fea-

tures (user settings, user feedback and admin panel).

The web interface shown in Figure 5 allows users

to create and manage their models and the corre-

sponding access control. To store additional infor-

mation about the user’s model, the client actually re-

quests two models to be created on the platform, one

for the user’s model data and the other one containing

meta data on the model. This is one of the advan-

tages of having a simple open platform, it does not

force clients into a predeﬁned structure beyond the

few simple implications of RDF, but allows them to

deﬁne their own custom data structure in a semantic

way. If desired a client speciﬁc meta-model ontology

can be deﬁned and also stored on the platform either

privately or openly to share it with others.

One of the most important features for the client

is the capability for users to collaborate with each

other on a model in real-time. Therefore, updates

to the model are exchanged between all users cur-

rently “connected” to the model in real-time. To raise

awareness on which users are currently connected to

a model, this information is indicated by the client on

the left side of the view as shown in Figure 5. Users

connected to the same model are shown by their user

icons. OWL imports play a special role for models

since they provide a mechanism to link different mod-

els to each other. It allows for simple reuse of models

and is also important for integrating between different

models. Due to this special role the client provides

a specialized interface for managing model imports.

In the model settings, users with manage permissions

can manage access control for all users.

4 CONCLUSIONS & OUTLOOK

“Scientists are primarily interested in doing science,

not software.” (Kelly and Sanders, 2008). In engi-

neering research environments, this inherent fact can

be linked to cases of corporate amnesia, implying a

loss of organizational knowledge. In this paper, the

discussion centers around why engineering research

organizations are particularly affected by corporate

amnesia, and it is demonstrated how technical solu-

tions such as SWT can be leveraged to preserve or-

ganizational knowledge While the ﬁeld of knowledge

management is wide, this research focuses on the do-

main of KBE. As illustrated in Section 2, previous

efforts have shown the value of SWT for the develop-

ment of KBE applications. It enhances the reusability

of knowledge, simpliﬁes extensibility, and provides

KEOD 2023 - 15th International Conference on Knowledge Engineering and Ontology Development

214

the ﬂexibility to choose inputs and outputs, result-

ing in more versatile applications. In this respect,

the Codex framework has generally been proven valu-

able. However, experience has shown that the barrier

to using the Codex framework is currently too high

for widespread adoption by scientists. To broaden

the accessibility of technological solutions to a larger

user group, a decision was made to develop a web-

based application using the current framework. This

application offers a collaborative user interface that

enables non-experts in SWT to engage in effective

knowledge collaboration.

The platform concept, as discussed in Section 3.1,

and its implementation (covered in Section 3.2), serve

as a foundation for a wide range of applications.

While the usability of the platform is demonstrated

through the implementation of the web application for

knowledge modeling in Section 3.3, it is yet to be de-

termined whether the goal of achieving widespread

adoption can be realized. This depends on the effort

required for modeling domain knowledge on the plat-

form, and whether the Return on Investment (ROI)

justiﬁes this effort. The answer to this question can

only be provided once additional KBE features and

domain-speciﬁc interfaces for engineering are devel-

oped and tested. Therefore, after ﬁnalizing the publi-

cation of models, the KBE related services will be ad-

dressed and additional user interfaces are going to be

developed. Meanwhile, even in the absence of those

speciﬁc KBE features, the platform will be utilized

for knowledge capturing and integration tasks. This

way, it aims to prevent corporate amnesia and instead

foster a robust corporate memory.

REFERENCES

Alder, M., Moerland, E., Jepsen, J., and Nagel, B. (2020).

Recent Advances in Establishing a Common Lan-

guage for Aircraft Design with CPACS. In Aerospace

Europe Conference 2020.

Boden, B., Cabac, Y., Burschyk, T., and Nagel, B. (2022).

Rule-based Veriﬁcation of a Geometric Design using

the Codex Framework. In 33rd ICAS.

Burschyk, T., Cabac, Y., Silberhorn, D., Boden, B., and

Nagel, B. (2021). Liquid hydrogen storage design

trades for a short-range aircraft concept. In DLRK

2021.

olling, M. (2015). L

osungspfadbasierte Analysen im En-

twurf komplexer Systeme. PhD thesis, University of

Stuttgart.

Dalkir, K. (2005). Knowledge Management in Theory and

Practice. Elsevier.

International Organization for Standardization [ISO]

(1994). Information technology — Open Systems

Interconnection — Basic Reference Model: The

Basic Model. ISO/IEC 7498-1, ISO.

Internet Engineering Task Force [IETF] (1981). Transmis-

sion Control Protocol. Technical Report RFC 793, In-

ternet Engineering Task Force.

Kelly, D. and Sanders, R. (2008). Assessing the Quality of

Scientiﬁc Software. First International Workshop on

Software Engineering for Computational Science and

Engineering, Leipzig Germany, May 2008.

La Rocca, G. (2012). Knowledge based engineering: Be-

tween AI and CAD. Review of a language based

technology to support engineering design. Advanced

Engineering Informatics: the science of supporting

knowledge-intensive activities, 2012(26):159–179.

Li, M. and Gao, F. (2003). Why Nonaka highlights tacit

knowledge: a critical review. Journal of Knowledge

Management, 7(4):6–14.

Musen, M. A. (2015). The prot

e project: a look back and

a look forward. AI Matters, 1(4):4–12.

Nicolaescu, P., Jahns, K., Derntl, M., and Klamma, R.

(2016). Near Real-Time Peer-to-Peer Shared Editing

on Extensible Data Types. In Proceedings of the 2016

ACM International Conference on Supporting Group

Work, page 39–49, New York, NY, USA.

Nonaka, I. and Takeuchi, H. (1995). The knowledge-

creating company: How Japanese companies create

the dynamics of innovation. Oxford University Press,

New York.

Nonaka, I. and Takeuchi, H. (2021). Humanizing strategy.

Long Range Planning, 54(4):102070.

Red Hat (2021). Drools Documentation 7.49.0.Final. Avail-

able at: https://docs.drools.org/7.49.0.Final/drools-

docs/html single/index.html. Accessed: 20 June

2023.

Verhagen, W. J., Bermell-Garcia, P., van Dijk, R. E.,

and Curran, R. (2012). A critical review of

Knowledge-Based Engineering: An identiﬁcation of

research challenges. Advanced Engineering Informat-

ics, 26(1):5–15.

World Wide Web Consortium [W3C] (2012). OWL 2 Web

Ontology Language Proﬁles (Second Edition). W3C

Recommendation, World Wide Web Consortium.

Zamboni, J., Zamﬁr, A., and Moerland, E. (2020). Semantic

Knowledge-Based-Engineering: The Codex Frame-

work. In 12th IC3K, volume 2, pages 242–249.

Zamboni, J., Zamﬁr, A., Moerland, E., and Nagel, B.

(2022). A semantic knowledge based engineering

framework for the rapid generation of novel air ve-

hicle conﬁgurations. In 33rd ICAS.

On the Development of a Collaborative Knowledge Platform for Engineering Sciences

215