A NEW KNOWLEDGE MANAGEMENT TOOL TO FACILITATE

PROCESS INNOVATION IN MANUFACTURING COMPANIES

Daniela Butan, Emma O’Brien, Mark Southern, Seamus Clifford

Enterprise Research Centre, University of Limerick, Plassey, Limerick, Ireland

Michael Pomeroy

Materials and Surface Science Institute, University of Limerick, Plassey, Limerick, Ireland

Keywords: Knowledge creation, Engineering knowledge, Process innovation, Variation Mode and Effect Analysis

(VMEA), Design of experiments (DOE), Finite Element Analysis (FEA).

Abstract: This paper presents a novel KM tool to allow companies to obtain sufficient knowledge about its process in

order to enhance its competitiveness and to innovate. To date there is no practical multidisciplinary model

that enables companies to switch from engineering chaos to a structured, robust process. This new approach

creates a reliable framework which promotes innovation and it offers a sustainable model for knowledge

creation in that the knowledge generated through its use can be continuously build upon to expand the body

of internal knowledge within the company. The model is based on existing engineering tools and exploiting

the knowledge generated through their use using the interchange between tacit and explicit knowledge thus

it is presented in the context of Nonka’s knowledge spiral. The model has been used successfully in a

number of case studies one which is presented in this paper.

1 INTRODUCTION

It has long been recognised the role of innovation in

increasing the competitiveness of a firm. Innovation

provides a mechanism for a firm to respond to

changes quickly and thus improve its lifecycle.

“Innovation involves the utilisation of new

knowledge or a new use or combination of existing

knowledge. New knowledge may either be generated

by the innovating firm in the course of its innovation

activities (i.e. through intramural R&D) or acquired

externally through various channels (e.g. purchase

of new technology). The use of new knowledge or the

combination of existing knowledge requires

innovative efforts that can be distinguished from

standardised routines”. (OECD, 2005)

The objective of this paper is to outline a method of

building knowledge about a process in a company in

order to facilitate process innovation. It will look at

the role of Nonka’s knowledge spiral in terms of

knowledge creation and will describe the use of a

proposed novel model (VDF) in the context of

Nonka’s knowledge spiral. It will outline a case

study illustrating the successful use of the VDF

model in building a significant amount of knowledge

in a manufacturing company which allowed the

company to make considerable improvements and to

innovate.

2 THE KNOWLEDGE CREATION

PROCESS AND THE ROLE OF

VDF

The knowledge creation process as outlined by

Nonka (2000) is a spiral, consisting of four phases

Externalisation, Socialisation, Combination,

Internalisation and Socialisation – Figure 1. It

consists of a conversion process between tacit

(knowledge in the minds of individuals) and explicit

(documented) knowledge. As the creation process

spirals through the interaction between tacit and

explicit knowledge the amount of knowledge in the

organisation expands.

342

Butan D., O’Brien E., Southern M., Clifford S. and Pomeroy M..

A NEW KNOWLEDGE MANAGEMENT TOOL TO FACILITATE PROCESS INNOVATION IN MANUFACTURING COMPANIES.

DOI: 10.5220/0003095003420347

In Proceedings of the International Conference on Knowledge Management and Information Sharing (KMIS-2010), pages 342-347

ISBN: 978-989-8425-30-0

 2010 SCITEPRESS (Science and Technology Publications, Lda.)

Figure 1: The knowledge spiral (Nonka 1998).

The knowledge spiral involves a number of key

phases:

• Socialisation (tacit to tacit) – sharing what

you have learned with other team members.

• Externalisation (tacit to explicit) –

documenting in some way the knowledge

you possess.

• Combination (explicit to explicit) –

selecting multiple sources of explicit

knowledge and combining it into some

form which the individual understands.

• Internalisation (explicit to tacit) – using

existing information automatically in your

daily work.

The knowledge spiral offers a method that provides

companies with a guide of what phases are required

to increase the amount of knowledge in the

organisation however it does not offer practical tools

to allow the company to create, build that knowledge

and to promote innovation. The VDF model offers a

suite of practical tools to allow companies to build

the knowledge they require for process innovation.

3 USING THE VDF MODEL IN

THE CONTEXT OF THE

KNOWLEDGE SPIRAL

The VDF model combines a number of existing

tools in order to complete the phases in the

knowledge spiral to maximise the effect of

increasing the body of knowledge in the

organisation. The engineering tools used in the VDF

model are:

• Variation Mode and Effect Analysis

(VMEA)

• Design of Experiment (DOE)

• Finite Element Analysis (FEA)

3.1 The VDF Model - Description

The new VDF model represents a powerful KM

practical tool which is capable of using the existing

tacit knowledge, converts it into an explicit

knowledge package and uses that in the most

efficient way to solve problems, optimize and

innovate in companies. Unlike process

improvement the current method creates a

multidisciplinary framework which promotes

innovation into the organization

The first component of the VDF model, the

VMEA uses a brain storming- like technique in

order to elicit the tacit knowledge in the minds of the

team involved in the process and transforms it into

explicit knowledge. The team of experts brainstorm

the factors which they think are causing process

problems, they rank these and assign them weights

using VMEA tables. Then using dedicated ranking

algorithms, the VMEA finds and prioritizes the

process characteristics for which the unwanted

variation is detrimental. This results in a list of

factors with different priority numbers those with the

highest priority numbers are the factors which have

the most impact on the process or product.

After the VMEA, a statistical DOE analysis will

be performed to determine the effect of altering the

parameters on the process and the most suitable

combination of these parameters which will ensure

maximum efficiency of the process.

After the VMEA and DOE were performed, the FEA

analysis comes into play, underpinning the process.

Using specialised modelling packages and dedicated

engineering principles, FEA simulates and predicts

process behaviour and finds out factors that went

undetected by the VMEA and DOE methods.

The three components of the VDF model act as

feeds to one another and their complementary

approaches produce the most efficient analysis of the

process, creating a structured and sustainable

platform for robust process and innovation into the

company with minimum cost involved – Figure 2.

Figure 2: VDF model – Component elements.

A NEW KNOWLEDGE MANAGEMENT TOOL TO FACILITATE PROCESS INNOVATION IN MANUFACTURING

COMPANIES

343

3.2 VDF Model and the Knowledge

Spiral

The VDF model can be used as a tool in process

innovation to increase the body of knowledge in the

organisation in line with the steps outlined in the

knowledge spiral:

• Socialisation (Tacit to Tacit) – VMEA allows

employees from several departments to

disseminate their knowledge to the others using

brainstorming sessions

• Externalisation (Tacit to Explicit) – The

VMEA then documents this knowledge into a

form which can be used. The VMEA uses a

structured method to calculate the greatest

causes of problems in a process and this is

documented and fed into the DOE and FEA.

• Combination (Explicit to Explicit) – In the

VDF process knowledge is combined from a

variety of explicit processes. The DOE uses

the results of the VMEA to concentrate on the

factors that are the greatest cause of process

problems and to determine the effect on the

process of altering these factors at different

levels. The FEA uses the knowledge obtained

from the VMEA and DOE to fine tune the

process and to produce process behaviour

predictions. The FEA results will be compared

and evaluated against the results of the

experimental DOE and the predictions will be

validated. Process factors that were undetected

by the DOE will be found through the FEA

analysis, a complete body of knowledge of the

process will be produced.

• Internalisation (explicit to tacit) – The results

from the DOE and FEA are disseminated to the

original brainstorming group in a final VMEA

using the findings of the experiments and

analysis and through discussion. This

internalises the knowledge within the minds of

the individuals so they can use it in their work

3.3 The VDF Model in Operation –

Case Study

The engineering company in this case study is a

medical company which presented itself with a

product failure due to the unknown causes during

the fabrication process.

Due to confidentiality issues, the company cannot be

named, as well as their product and fabrication

process. The names will be kept confidential but the

procedure will be explained in detail. To investigate

the process and the root cause of the product’s

failure the proposed approach was the VDF model.

The investigation started off with a VMEA

brainstorming session which allowed employees

from several departments (technicians, design

engineers, managerial team, quality department etc)

to disseminate their knowledge to one another,

approach that encompasses the tacit- to- tacit aspect

of the ‘Knowledge Spiral’ model.

Then the knowledge in the minds of the team

involved in the process was transformed into explicit

knowledge through the VMEA document which

outlines the tacit to explicit feature of the

Knowledge Spiral model.

Using the VMEA structured method and the

ranking algorithms proposed by Johansson et al.

(2006), the greatest causes of variation in the process

that affected the failure of the product were

identified and documented as shown in Table 1

below. A Variation Risk Priority Number (VRPN)

was calculated which computed the effect of each

process factor on the failure of the product and

identified the process factor that needed to be

investigated further. The highest the total VRPN

number - the greater the influence of that factor on

the product failure.

Initially it was thought that Factor 1 was the

greatest cause of variation on the product but from

Table 1 it can be seen that the calculated highest

VRPN total number (1730396) corresponded to the

Sub-KPC Factor 7. It was concluded that the Factor

7 process characteristic, by its variation, had the

greatest influence on the product failure.

However the method above only provides an

indication of the factors with the greatest effect on

process variation that could ultimately affect the

product failure but it cannot show how these factors

actually impact on the process itself. Therefore

more in depth explicit analyses are needed.

The VDF model then adds a combination of two

engineering methods - DOE and FEA, which

enhances the company’s knowledge and through its

explicit to explicit approach outlined in the

knowledge spiral in Section 2 of this paper the

model creates efficient practical paths to innovation.

The engineering knowledge captured through the

VMEA brain storming session and the data provided

by the VMEA table above acted as feeds for the

remaining two explicit elements of the VDF model:

the DOE and the FEA.

The DOE was performed on the process stage

named Factor 7 to determine the process optimum

running parameters and the most suitable

KMIS 2010 - International Conference on Knowledge Management and Information Sharing

344

Table 1: VMEA Table – tacit to explicit document.

combination of these parameters which will ensure

the product meeting the life outlined in the

specifications. As there were too many factors in this

process stage which would make the DOE

experiment very expensive, the most significant

measurable factors had to be taken into

consideration. Out of the 22 factors having an

influence on the variation of the Factor 7 process

stage, only 7 factors were identified as being

significant and measurable and they formed the main

elements of the DOE design – Table 2. The process

was run 14 times with these factors at different

combinations of high and low settings, the effect of

these settings on factor 7 for each run was recorded.

Using the statistical package Wisdom the R

values

were computed. The R

value indicates how much of

the variation was attributed to that factor.

Good R

values, over 80%, were obtained for all of

the factors from the DOE experiment. Therefore,

almost all experiments found at least 80% of the

causes of variation.

An optimum process set up was found by the

Table 2: Extract DOE design table, explicit to explicit

approach.

The DOE analysis also showed that the product

exhibited a non-uniform microstructure after

fabrication and that was considered a possible cause

for the product failure.

Still more research had to be done to capture all

of the process factors that have an impact on the

A NEW KNOWLEDGE MANAGEMENT TOOL TO FACILITATE PROCESS INNOVATION IN MANUFACTURING

COMPANIES

345

product behaviour.

In the meantime, a FEA analysis was performed

to simulate the product behaviour in order to get an

understanding of the product parameters that were

most likely to be influenced by the variation into the

fabrication process and which could contribute to the

failure.

The FEA Ansys multiphysics package was used to

simulate the product behaviour – a thin metallic

plate vibrating at a very high frequency. Different

vibration mode shapes were found for different

values of the Plate Natural Frequency - NF

The results (Figure 3 – middle), were compared

with the literature models (Figure 3 - top) and

experimental readings (Figure 3 - bottom). Good

correlation was found, therefore the FEA model was

declared valid. A series of important predictions of

product behaviour related to the material properties

and geometrical characteristics were made, results

that could not be identified by the VMEA and the

DOE analysis described earlier.

Figure 3: FEA simulation, explicit to explicit approach.

Figure 4: FEA simulation, process behaviour, explicit to

explicit approach.

After the product behaviour was modelled, the same

FEA package simulated the process behaviour using

the Ansys multiphysics Fluid option and the

optimum setting parameters found through the DOE

analysis.

A lot of variation during the fabrication process

was predicted due to the flow behaviour - Figure 4,

factor that was not possible to be identified by the

VMEA and the DOE analysis and that could

contribute to the premature failure of the product

and to low yield.

Based on the FEA results above, a new feature of the

fabrication process was designed to ensure a more

uniform flow distribution – Figure 5. A more

consistent product’s microstructure and higher yield

were expected.

Figure 5: FEA simulation, new process design feature.

The new process design feature along with the

knowledge captured by modelling the product

behaviour and the DOE analysis were implemented

into the process.

A better product’s microstructure uniformity was

achieved, the product met the life expectancy

outlined in the Specifications, the process became

fully controllable and an increase in yield by 80%

was recorded.

In the final stage of the VDF approach, the

results from the DOE and FEA were disseminated to

the original brainstorming group in a final VMEA

using the findings of the experiments and analysis

and through discussion, allowing the knowledge to

come back full circle to the employees in a similar

explicit to tacit manner as in the knowledge spiral

model.

This final step internalises the knowledge within the

minds of the individuals so they can use it in their

work.

4 CONCLUSIONS

The VDF multidisciplinary approach proved its

efficiency and validity through the successful case

study results described in the Section 4 above. The

VDF model can be used as a tool in process

innovation to increase the body of knowledge in the

KMIS 2010 - International Conference on Knowledge Management and Information Sharing

346

organisation in line with the steps outlined in the

knowledge spiral. Unlike process improvement the

current method creates a multidisciplinary

framework which promotes innovation into the

organization

The model can be iteratively used to expand the

engineering knowledge in the organisation. The

knowledge developed in the model and recorded in

the FEA can be used to determine the impact of

other alterations on the process if they are required

as a result of market changes (such as a change in

technology, raw material) or customer requirements.

These can be used as a basis to expand the

knowledge about the process by conducting a

VMEA on the factors which may cause problems in

the new process and conducting DOEs on these

factors to obtain in depth information. Thus the VDF

model offers a sustainable process for the creation of

engineering knowledge which can be continuously

built upon and enhance the competitiveness of the

firm.

REFERENCES

Johansson, P., Chakhunashvili, A., et al. 2006. Variation

Mode and Effect Analysis: a practical tool for quality

improvement. Quality and Reliability Engineering

international (in press).

Chandler, D., A., Hagstrom, P., Solvell, O., 1998. The

Dynamic Firm: The Role of Technology, Strategy,

Organisation and Regions: Oxford University Press.

Nonaka, K., 2000, The concept of Building a foundation

for knowledge creation. Handbook of knowledge

management, 7

OECD, 2005, Guidelines for Collecting and Interpreting

Innovation Data, (3

ed.)

A NEW KNOWLEDGE MANAGEMENT TOOL TO FACILITATE PROCESS INNOVATION IN MANUFACTURING

COMPANIES

347