KMDL FOR INNOVATION AND PRODUCTION RAMP-UP

PROCESS EVALUATION

A Case Study

Julian Bahrs and Priscilla Heinze

Chair of Business Information Systems and Electronic Government, Potsdam University

August-Bebel-Str. 89, 14482, Potsdam, Germany

Keywords: Case study, Knowledge intensive business process, Business process oriented knowledge management,

Process modelling, Knowledge process.

Abstract: The Knowledge Modeling and Description Language (KMDL) is a method for analyzing knowledge

activities in business processes. This contribution presents version 2.1, the latest version of the method

KMDL in a real life scenario. In the case study presented in this contribution we aim to review the

practicality of the KMDL procedural model and the benefit gained from its application as it allows the

identification of problems. KMDL analysis delivers the identification of causes as well as measures to

overcome these problems, which are highly accommodating for process improvements.

1 INTRODUCTION

Knowledge management has not derived to an

ultimate solution. A mixture of methods and

instruments individually combined has to match the

company specific goals, culture and requirements.

Business process oriented knowledge

management is seen as bridging link that

individually combine methods and tools from both

worlds in the context of a process with a demand

driven perspective. The business process itself

stands as a starting point and the design concept for

knowledge management (Abecker et al., 2002). In

fact, it is the area for application and learning as well

as context for sharing knowledge. Several

approaches have been developed from various

researchers (Woitsch and Karagiannis, 2005, Heisig,

2003, Kim, Lim and Mitchell, 2006, Allweyer, 1998,

Remus, 2002). However, only limited results and

experiences from projects have been published.

Exceptions are (Telesko and Karagiannis, 2002) as

well as (Fröming, Gronau and Schmid, 2006), who

present the application of a prior version of KMDL

(Gronau and Weber, 2004) in the domain of

software engineering.

One of the aims of this paper is to introduce the

expanded functionality of the former Knowledge

Modeling and Description Language (KMDL)

version 2.1 (see http://www.kmdl.de). To facilitate

this aim, we will provide some practical evidence to

answer the following research questions by means of

a case study:

 How does KMDL help to identify and to

highlight the knowledge activities or

processes?

 How is KMDL an appropriate method to

analyze and improve knowledge intensive

business processes?

 How can KMDL be improved? Which benefits

do the expanded functions in version 2.1

offer?

2 CASE STUDY

Our reason of choosing case study as the approach is

explained by (Yin, 1993), as he defined case study

as „[…] an empirical inquiry that investigates a

contemporary phenomenon within its real-life

context, especially when the boundaries between

phenomenon and context are not clearly evident“.

The underlying theory of the KMDL application was

described in the former section.

2.1 Case and Data Selection

The selected case study took place in a major

German household appliances producing company,

280

Bahrs J. and Heinze P. (2009).

KMDL FOR INNOVATION AND PRODUCTION RAMP-UP PROCESS EVALUATION - A Case Study.

In Proceedings of the International Conference on Knowledge Management and Information Sharing, pages 280-283

DOI: 10.5220/0002296402800283

 SciTePress

which has chosen to stay anonymous in this

publication. The project this case represents covers a

thorough implementation of the KMDL method in

evaluating the company’s innovation and production

ramp-up processes.

A qualitative data gathering approach as research

method was selected, due to its conformation with

the KMDL method. The result of the examination

will be analysed along with a comprehensive

literature research. However, since it is not up to the

analysts to perform the recommended measures and

changes, the direct impact of the KMDL method to

the company cannot be presented in this paper.

2.2 Implementation

In the next sub-section, we will define the actors

involved in the case study (from hereon will be

referred to as project) and the identified problems.

Subsequently, we will illustrate the application of

KMDL following the structure of the KMDL phases

(see Chapter 2), including its achieved results.

Generally categorized, there are two types of

actors involved in the project: The client and the

analysts.

The client in this case is a large and globally

active company, which comprises of multiple

business units. The one focused on has a central

product development with more than 10 globally

distributed production sites. Approximately 200

people work in the product development, including

hardware engineering and software customizing.

Fourteen people are directly involved in the project

from the client side while the analysts comprise of

two assistant researchers with expertise in the

KMDL method and experiences from former other

process evaluation and optimization projects The

project was initiated at the first place because there

was a disparity between the anticipated and actual

processes at the client side, as discovered and agreed

on in several initial meetings by means of

unstructured interviews.

Due to the global distribution, the client had

been faced with a delayed production start. The

problem lies in the handover process of the bill of

material from the development to the production

sites, despite the fact that a central product

development was instigated.

2.3 KMDL Phases Implementation

2.3.1 Project Acquisition (P0)

A thorough KMDL analysis dealing with the

structure, the processes, and the information systems

of the client was offered. The project also includes

particularly the factors related to information or

knowledge transfer and creation in the handover

process.

The subject of analysis is the product

development process including the handover process

as part of the process.

2.3.2 Setting Focus (P1)

During the initial meetings possible areas of

examination within the production development

processes were also discussed. In order to examine

the process intensively, the analysts decided to

observe two process instances from the recent past,

which represent a real-life example and serve as

basis of the interviews. As a matter of fact, one of

these instances has an exceptionally long delay of

production start.

The objectives for every KMDL project were

derived in mutual discussions. In this project, they

include a complete, unambiguous and redundant-

free documentation, a punctual handover and

development process cycle and the employment of

minimal efforts for the hand-over.

2.3.3 Capturing the Process-based View (P2)

A kick-off workshop including a KMDL basic

training was performed in conjunction with Phase 2

and Phase 3. During this workshop the prepared

objectives were approved by the project participants.

Due to the limited time frame, Phase 2 was

performed backwards. Instead of performing

interviews to gather data, the analysts primarily

generated a draft model of the process view based on

the information they obtained in the initial meetings

and presented them afterwards in the kick-off

workshop. Suggestions for improvements and

possible focus points for the activity view were then

discussed among the project participants.

By discussing the (pre-modelled) process view

with the project participants, the analysts gain the

validation needed for them to proceed to the next

phase. Through the workshop, the client gains a

prompt understanding of the process as well as a

new perspective of the examined process.

A validated process view visualizing the

important steps of the development project including

initialization, conception, pre-series conduction up

to the hand-over to production sites, which marks

the start of the production was produced.

This phase aims to identify performed the

knowledge intensive tasks. As explained before,

since a pre-modelled process view was developed

and presented in the workshop, the identification of

the tasks took place simultaneously.

KMDL FOR INNOVATION AND PRODUCTION RAMP-UP PROCESS EVALUATION - A Case Study

281

Identifying knowledge intensive tasks provides

an outline. This outline is essential for the

development of the activity view, which provides

process visualization in a higher detail. After the

tasks were identified, relevant interview partners

were identified.

2.3.4 Capturing the Activity-based View

(P4)

In capturing the activity-based view, the analysts

used the interview method as recommended by the

KMDL procedural method. This phase aims to

gather as much information as possible about the

activities taking place within each task in the

process.

A total of twelve semi-structured interview

sessions were performed with the client. They

included employees from the engineering, the

management and ramp-up agents from production

sites. The process model as well as the two selected

process instances were used as a guideline for the

interviews.

Each interview was documented in a protocol,

which was used to developthe activity models.

Subsequently, each protocol was mailed to the

interview partner for approval and clarification of

any remaining ambiguities. A total of 10 activity

models were generated.

For the analysts, this phase serves as the basis

and data resource for their further examination.

2.3.5 Analysis (P5)

The identified problem lies in the handover of the

bill of materials. Two different process instances

from the past reflecting distinctive bill of materials

compilation processes were selected.

The first instance shows that only segments of

the bill of materials are compiled at the development

site, which results in an inconsistency as the

compilation was completed at the production sites.

The second instance shows a complete compilation

of the bill of material.

In both cases, adoptions to local markets take

place at the production site. Most development

works are split up into technical segments

anddocumented within these segments. The

documentation involves a central database with

change notes as well as material follow-up chart per

segment

Using the KMDL method, it was found that the

first instance relies predominantly on informal

communication mainly dealing with identifying

which version of a segment should be used for a

complete product. This finding interestingly

contrasts the documentation efforts made during the

actual development.

Furthermore, required documents are often no

longer up to date and must be validated. This might

be caused by the fact that the ramp-up personnel

from the production site only has limited access to

these documentation.

In the second instance, a significant domination

of internalization and combination was found. A

large number of documents, often having no

relevance to the production site, need to be

interpreted. A crucial document called the material

follow up chart, despite its employment across all

segments, within each segment, and sometimes even

each project, uses individual formats. Having to

adjust each time to a new document format

containing the same content is a waste of time and

increase risk of error. Furthermore, a large number

of development projects were found, although it

often entails only minor changes.

Based on the above findings, the analysts

suggested a standardization of processes, especially

by defining responsibilities for pre-series purchasing

as well as by reverse scheduling, i.e. by assigning

roles for list maintenance as well as processes

review. The use of the list should be encouraged or

even made mandatory. Subsequently, this list could

be linked with the existing systems, e.g. to access

drawings or change orders. The centrally created

master bill of material of the one instance should be

made part of the standardized interface towards

production sites, since it would provide an immense

time saving at the production site. Their creation at

the development site would allow informal

knowledge acquisition and validation of documents.

Lastly, the total number of products and variants

developed should be examined, whether the amount

is truly necessary.

They also suggested a standardization of the

interfaces towards production sites to make contacts

known at both ends. The material follow-up chart

should use a common template in order to improve

its comprehensibility.

For the client, the recommended measures

should serve as a basis for their process

optimization. Being shown the actual process as

observed from the knowledge management angle,

the client gets a direct overview of their process’

vulnerability. However, the analysts cannot directly

influence the positive result of the improvement of

the process since the implementation of the

measures remains the task of the client.

2.3.6 Development of a Qualified Concept (P6)

The preliminary results and measures were discu-

KMIS 2009 - International Conference on Knowledge Management and Information Sharing

282

ssed with the client. Their feedback serves as a basis

to validate the above findings and to check the

possibilities for the measures to be suggested. The

potentials and problems as well as excerpts from

KMDL models were presented during the result

workshop. Subsequently, the client discussed the

measures described above and jointly prioritized

them by considering their estimated benefit and

costs.

3 CONCLUSIONS

AND OUTLOOK

Objective 1: How does KMDL help to identify and

to highlight knowledge activities or processes?

KMDL enables process analysts to gain an in-

depth understanding of a process and the underlying,

often previously unidentified knowledge activities.

By means of interviews as well as feedbacks and

approvals on various levels, the actual situation can

be reflected and intensely examined.

The interviewed employees can bring up the

problems concerning the knowledge transfer,

generation, sharing and the like according to their

field of expertise. The client is able to look into their

business processes beyond the typical flow oriented

point of view. The process and the activities are

reviewed by multiple actors from different

perspectives, which encourages unbiased views. As

a result knowledge activities are documented and

made visible.

Objective 2: How is KMDL an appropriate

method to analyze and improve knowledge intensive

business processes?

Prior to the application of the method the client’s

understanding of the problem was only vague.

During the project, this understanding was

broadened and concrete causes of the problem were

identified by obtaining transparently reflected

processes and activities within the processes. They

also benefit from the analysis and the recommended

measures.

Apart from that, clients of KMDL projects do not

need a comprehensive method knowledge Any

inquiries about the method or the analysis could

easily be transferred during the two workshops.

By focusing on concrete instances of the process

a better caption of the actual activities was obtained.

However, two instances are hardly enough evidence

for empirical analysis. It remains an issue of

interpretation to evaluate the models, which is why

the analysts are required to have sufficient

experiences in the field.

Objective 3: How can KMDL be improved?

Which benefits do the expanded functions in version

2.1 offer?

As described, in this case the procedural model

has been simplified, due to very limited resources.

However, the separation of the process and activities

views has proven to be a significant improvement.

The process model acts as source for an overview

and reference for a detailed analysis of knowledge

activities. Transfers and applications can also be

traced within various activities.

In capturing the general characteristic of the

communication, the evaluation of small numbers of

process instances produces only exemplary results.

Formal and quantitative methods are more

appropriate to capture communication. These

communication acts can indicate additional

knowledge exchange.

Only recently KMDL was extended with an

additional view for communication (Müller, 2008),

which captures actors, communication acts as well

as communication instruments. The view

differentiates types of communication by anytime /

anyplace matrix. New angles such as a planned

(e.g. scheduled during a meeting) and random (e.g.

met in the hall) communication act are used to

validate the activity view and to incorporate random

acts of (informal) knowledge exchange into the

analysis.

REFERENCES

Abecker, A., Hinkelmann, K., Maus, H. & Müller, H.-J.,

2002. Integrationspotenziale für Geschäftsprozesse

und Wissensmanagement. In: Geschäftsprozessorientiertes

Wissensmanagement, pp. 1-22. Springer. Berlin.

Allweyer, T., 1998. Knowledge Process Redesign -

Modellierung und Gestaltung der Wissensverarbeitung

im Unternehmen. IDS Prof. Scheer GmbH.

Fröming, J., Gronau, N. & Schmid, S., 2006. International

Journal of Knowlegde Management (IJKM),vol2, pp.

32-51.

Gronau, N. & Weber, E., 2004. Defining an Infrastructure

for knowledge intensive Business Processes. In:

Proceedings of IKNOW 2004, pp. 424-431. Graz,

Austria,

Heisig, P., 2003. Business Process Oriented Knowledge

Management. In: Knowledge Management - Concepts

and Best Practices, pp. 15-44. Springer. Berlin.

Kim, S.-K., Lim, S.-H. & Mitchell, R. B., 2006.

International Journal of Knowlegde Management

(IJKM), vol 2, pp. 17-31.

Müller, C., 2008. Graphentheoretische Analyse der

Evolution von Wiki-basierten Netzwerken für

selbstorganisiertes Wissensmanagement, Gito. Berlin.

KMDL FOR INNOVATION AND PRODUCTION RAMP-UP PROCESS EVALUATION - A Case Study

283