You are on page 1of 12

IncreasingTransferabilityofTacitKnowledgewithKnowledge

EngineeringMethods
ThiernoTounkara
TELECOM&ManagementSudParis,Evrycedex,France
Thierno.Tounkara@itsudparis.eu

Abstract:Knowledgetransferisarealchallengefororganizationsandparticularlyforthosewhohavebasedtheirstrategy
onknowledgecodificationusingknowledgeengineeringmethods.Theseorganizationsarefacingonemajorproblem:their
knowledgerepositoryisusedbyfewpersons.Why?Inthisarticle, weidentifybarriersfortransferandappropriationof
codifiedknowledgereferential.Weshowthatcodifiedknowledgetransfershouldbeaspecificcollaborativeprocesstaking
into account three aspects: complexity and specificity of codified knowledge, readers profiles, and exchange channels.
Then,weproposetoimproveknowledgetransferprocessbydevelopingnewspecificationsforthecodifiedknowledgeto
increase its transferability and by elaborating apertinent shared context for knowledge interpretation. It is anempirical
methodologywhichoptimizescontinuitybetweenknowledgecodificationandknowledgetransfer.

Keywords:knowledgemanagement,knowledgetransfer,knowledgecaptureandcodification,knowledgeengineering,
knowledgesharing,knowledgeappropriation,organizationallearning,organizationalmemories

1. Introduction
TheconceptofKnowledgeTransferwasfirstintroducedbyTeece(1977).

KnowledgeTransfercanbedefinedasaprocessinwhichanorganizationrecreatesandmaintainsacomplex,
causally,ambiguoussetofroutinesinanewsetting(Szulanski1996).

This process is a key part of knowledge management cycle and allows organizations to absorb and make
optimaluseofcrucialknowledge.

Research on knowledge transfer focuses on three themes (Harrison and Hu 2012, Dalkir 2011, Alavi and
Leidner2001,GuptaandGovindarajan2000,Zack1999,Simonin1999,Szulanski1996,Moweryandal1996,
ZanderandKogut1995):

Factors which affect knowledge transfer; they are dimensions for measuring the degree to which
knowledgecanbeeasilycommunicated,understoodandtransferred

Modes or processes of knowledge transfer which deal with mutual transformation between tacit
knowledgeandexplicitknowledge

Evaluation and measurement of the performance of knowledge transfer; the goal is to elaborate
indicatorstomeasureefficiencyofknowledgetransfer.

Our research deals with the two first themes. We refer to knowledge transfer models which consider
knowledge elicitation as a possible stage for sharing and transferring knowledge. Focusing on knowledge
engineering techniques for knowledge elicitation andorganizational memories elaboration, we explore their
limitsanalyzingcodificationeffectsonfactorswhichaffectknowledgetransfer.Thenweproposeanapproach
allowing an optimal continuity between knowledge capture using knowledge engineering methods and
knowledgetransferatindividualandorganizationallevels

2. Factorsinfluencingknowledgetransfer
Relyingonliteraturereview,wecangroupfactorsinfluencingknowledgetransferinto4dimensions:

Characteristicsofknowledge

Knowledgetransferchannels

Absorptivecapacityofreceivers

CulturalandOrganizationalcontexts

ISSN14794411
268
ACPIL
Referencethispaperas:Tounkara,TIncreasingTransferabilityofTacitKnowledgewithKnowledgeEngineeringMethods
TheElectronicJournalofKnowledgeManagementVolume11Issue3(pp268279)availableonlineatwww.ejkm.com

ThiernoTounkara

2.1 Characteristicsofknowledge
Withcharacteristicsofknowledgewecanmeasuredifferentaspectswhichmaybefacilitatorsorbarriersfor
knowledgetransfer.

Relying on the work of Zander and Kogut (1995) and Simonin (1999), we highlight three characteristics that
wouldaffectknowledgetransfer:tacitness,complexityandspecificity(ordegreeofcontextualization).

Tacitnessversusexplicitness

Polanyidescribedtacitknowledgeasthingsthatweknowbutcannottell(Polanyi,1967)andthuscanonly
betransferredthroughinteraction.Tacitknowledgeisnoteasilyarticulatedorformalizedandisdifficulttoput
intowords,text,drawingsorothersymbolicforms.Infact,tacitnessisapropertyoftheknower:itiseasily
articulatedbyonepersonbutmaybeverydifficulttoexternalizebyanother.

Tacit knowledge is typically considered to be more valuable than explicit knowledge and requires more
cognitiveeffortsofasenderandreceivertobetransferred(Dalkir2011,HarrisonandHu2012).

Explicit knowledge is associated with declarative knowledge and know why. Declarative knowledge and
know why consist of descriptive elements (Garud 1997). Explicit knowledge represents content that has
beencapturedinsometangibleformsuchaswords,audiorecordingsorimages.

Complexity

Knowledgecomplexitycanbedefinedasthenumberoftoolsandroutinesusedintheprocessofknowledge
transfer (Reed and Defillippi 1990). Routines are actions based on unstated conventions that were derived
from previous experiences and can embody the application of knowledge within an organization (Szulanski
1996).

Consequently, more routines are needed to interpret and appropriate knowledge more its transfer can be
difficult(ArgoteandIngram2000).

SpecificityorDegreeofcontextualization

Specificity describes the degree to which knowledge and routines in which it is embedded can satisfy the
knowledgereceiver.Inotherterms,specificitycapturesthedegreetowhichknowledgeisdependentornot
onmanydifferentcontextsofuse(ZanderandKogut1995).

More the knowledge can be adapted to the context of the receiver, be absorbed and understood by the
receiverthemoreitisvaluable.

Forexample,knowledgetightlyconnectedwithlocalexperiencesandculture,canbeabarriertotransferand
bedifficulttotransplanttootherenvironment.

2.2 Knowledgetransferchannels
Communication processes and information flows drive knowledge transfer in organizations. Existence and
richnessoftransmissionchannelsaresuccessfactorsforknowledgetransfer(GuptaandGovindarajan2000).

Knowledgetransferchannelscanbeinformalorformal,personalorimpersonal(HolthamandCourtney1998).

Informalmechanisms(suchasinformalseminarsorcoffeebreakconversations)refertosocializationandare
moreeffectiveinsmallorganizations(FaheyandPrusak1998).
However,suchmechanismsmayinvolvecertainamountsofknowledgelossduetothelackofaformalcoding
oftheknowledge(AlaviandLeidner2001).

Formaltransfermechanisms(suchastrainingsessions)mayensuregreaterdistributionofknowledgebutmay
inhibitcreativity.

www.ejkm.com

269

ISSN14794411

ElectronicJournalofKnowledgeManagementVolume11Issue32013

Personal channels (such as apprenticeships) may be more effective for distributing highly contextual
knowledge whereas impersonal channels (such as knowledge repositories), may be most effective for
knowledgethatcanbereadilycodifiedandgeneralizedtoothercontexts.

InformationTechnologiescansupportallfourformsofknowledgetransferchannels.

2.3 Absorptivecapacityofknowledgereceivers
Gupta and Govindarajan (2000) identified absorptive capacity as a key element for knowledge transfer
process.

Absorptive capacity can be defined of as the ability of a firm to recognize the value of new, external
information,assimilateit,andapplyit.(CohenandLevinthal1990)

It seems very difficult to control absorptive capacity because knowledge must go through a recombination
process in the mind of the knowledge receiver. This recombination depends on the recipient's cognitive
capacitytoprocesstheincomingstimuli(VanceandEynon1998).

2.4 Culturalandorganizationalcontexts
Interorganizational knowledge transfer (across organizational boundaries) seems to be more complex
comparedtoknowledgetransferwithintheorganization.Therearemanyreasons:

Cultural distance can raise barriers for understanding partners and transferability of knowledge based
assets

Organizational distance (centralized vs. decentralized, innovators vs. followers, entrepreneurial vs.
bureaucratic) can accentuate the difficulty of transferring knowledge through interorganizational
relationships(Simonin1999)

InourstudywelimitthescopetoacontextofknowledgetransferwithintheOrganization.

3. Modesofknowledgetransfer
For better understanding of Knowledge transfer it is important to explore first two complementary
approaches:socialexchangeandcodification.

3.1 Socialexchangeversuscodification
Wecanshareandtransferknowledgethroughsocialexchangewhichisaprocessofpersonalcommunication
andinteraction.Itisasocialisationprocess(focusingontacitknowledge)asdescribedbyNonakaandTakeuchi
(1995)intheirSECIknowledgemanagementmodel.

Knowledge codification is the process for transforming knowledge into a tangible, explicit form such as
document,thatknowledgecanthenbecommunicatedmuchmorewidelyandwithlesscost.

In our article, we analyse knowledge transfer strategy based on knowledge codification using knowledge
engineeringmethods.

3.2 Knowledgetransfermodels
We present here two theoretical models with distinct perspectives. These models bring a conceptual
frameworkformanyknowledgetransferprocesses.Theyhavebeenreviewedanddiscussedbyacademicsand
practitioners(Dalkir2011,HarrisonandHu2012).

These two models give us a better understanding of knowledge codification role in knowledge transfer
process.

SECImodel

TheSECImodelofNonakaandTakeuchihasproventobeoneofthemorerobustinthefieldofKM.thismodel
focuses on the knowledge conversion between tacit and explicit knowledge. It describes how knowledge is

www.ejkm.com

270

ACPIL

ThiernoTounkara

accumulatedandtransferredinorganizationsfollowingfourmodes:socialization,Externalization,Combination
andInternalization.

Socializationisthesharingoftacitknowledgethroughsocialinteractionssuchasfacetoface.

Externalizationistheprocessofconvertingtacitknowledgeinavisibleform:explicitknowledge.Itisaway,for
organizations,tomakeknowledgetangibleandstoreitinmanuals,databasesinordertobeeasilyshared.In
thismode,knowledgeengineeringmethodsareuseful.

Combination is the process through which discrete pieces of explicit knowledge are recombined into a new
form.

Internalizationisthelastconversionprocess(fromexplicitknowledgetotacitknowledge)whereknowledgeis
convertedintopersonalmentalmodelsandthencanbeusedinanoptimalwaytoachievetasks.

BOISOTISpaceKMModel

BOISOTKMmodelisaconceptualframeworkincorporatingatheoreticalfoundationofsociallearning.Boisot
(1998) suggested that knowledge is structured, understood and transferred through three dimensions:
codification,abstractionanddiffusion.

Codificationreferstothedegreetowhichknowledgecanbeencoded(evenifthereceiverdoesnothavethe
facility to understand it) while abstraction refers to a low level of knowledge contextualization (easy to be
generalizedtoothercontexts).

The assumption is that well codified and abstract knowledge is much easier to understand than highly
contextualknowledge.

Consequently,fortacitknowledgewithhighcontextuallevel(highdegreeofspecificity),thereisariskofloss
ofcontextduetocodificationwhichisabarrierforknowledgetransfer.Thatisoneofthelimitsofknowledge
transferprocessrelyingonorganizationalmemoriesbuiltwithknowledgecodifiedprincipallyusingknowledge
engineeringtechniques.

Highly contextual knowledge need a shared context for its interpretation and that implies facetoface
interactionandinageneralwayasocializationapproachasintheSECImodelofNonakaandTakeuchi(1995).

Inthismodel,codificationandabstractionworktogetherandfacilitatetheknowledgediffusionandtransfer.

4. Codifyingwithknowledgeengineeringmethods:Barriersforknowledgetransfer
Understandabilityanddiffusibilityofcodifiedknowledgewithknowledgeengineeringtechniquesdependon
manyfactors:

accessibilityandreadabilityofusedformalismsfortheknowledgereceivers(Dalkir2011)

knowledgereceiversprofiles(background,contextofknowledgeuse,preferencesforlogicalstructuring
andunderstandingprofiles)(Tounkaraandal2002)

levelofdescriptionofcomplexandspecificknowledge

exchangechannelsbetweenKnowledgesources(expertsorspecialists)andpotentialfutureusers

4.1 Multiplicityofformalisms
Knowledge engineering methods lead to a set of models and each of them correspond to a specific type of
knowledge.ForexampletheCommonKADSmethodologyproposesfivetypesofmodels(Diengandal2000):

Taskmodelofthebusinessprocessoftheorganization

Agentmodeloftheuseofknowledgebyexecutorstocarryoutthevarioustasksintheorganization

KnowledgeModelthatexplainsindetailtheknowledgestructuresandtypesrequiresforperformingtasks

Communicationmodelthatrepresentsthecommunicativetransactionsbetweenagents

www.ejkm.com

271

ISSN14794411

ElectronicJournalofKnowledgeManagementVolume11Issue32013

Designmodelthatspecifiesthearchitecturesandtechnicalrequirementsneededtoimplementasystem
includingfunctionsdetailedbytheknowledgeandcommunicationmodels

So,expertiseiscodifiedthroughformalisms(whichareoftendiagrams)dependingonthetypeofknowledge.

We can point out many difficulties associated with the multiplicity of models: Accessibility, readability and
understandability/intelligibility. The profile of knowledge receivers can accentuate those barriers: are they
familiartotheuseofmodels?Whatabouttheircognitivepreferencesofapprenticeship:aretheymoretextual
thanvisual?

Knowledgeengineeringmethodsonlyfocusonthecodificationofthetacitknowledgeofknowledgeablestaff
(expertsorspecialists)buttheydonottakeintoaccountappropriationandorganizationallearningcapabilities
ofreaders(potentialfutureusers).

4.2 Heterogeneityofreadersprofiles
InanOrganization,readersdonothavethesamelevelofexpertiseandtheirprofilescanbeheterogeneous
(background,contextsofknowledgeuse,preferencesforlogicalstructuring,understandingprofile,familiarity
withmodels,etc.)

However, the logical structuring and the presentation of the tacit knowledge codified are not guided by
learninglevelsoffuturereadersbutonlybytheconceptstackledwheninterviewingexperts/specialistsandby
themodelsstructure.

4.3 Background
Aknowledgereceiverwithimportantpriorknowledge(relatedtotheknowledgedomain)andfamiliartothe
use of models may have a greater absorptive capacity. It may be easier for such receiver to decode and
assimilateknowledgewithhighlevelofcomplexity.

4.4 Contextsofuse
More the distance between the receivers context of use and the described one is important, more the
knowledge receiver will make important cognitive efforts to adapt knowledge. This case happens when the
codifiedknowledgeisveryspecifictotheknowledgesourcescontext.

4.5 Preferencesforlogicalstructuringandunderstandingprofile
Preferences for logical structuring depend on the learning level of knowledge receiver. For a novice,
understanding concepts before procedural tasks could be more logic. On the other hand, an expert would
perhapspreferastructuringguidedbyproblemssolving.

Understanding profile can be assimilated to the cognitive preferences of the reader when learning: textual
and/orvisualpreferences.

When knowledge domain is codified taken into account logical structuring and cognitive preferences of the
reader,knowledgetransfercanbeacceleratedbecausetheknowledgereceivermakeslesscognitiveeffort.

4.6 Levelofdescriptionofcomplexknowledge
Moretheknowledgeiscomplexmoreitstransfercanbedifficult.
Toreducecomplexity,weproposecomplementaryactivitiestoenrichcodifiedknowledgereferential:

Identifyingsetsofcomplexknowledgealreadycodified

Describingandillustratingroutinesinwhichidentifiedcomplexknowledgeisembedded

Organizing exchange (with adequate knowledge transfer channels: informal or formal) between experts
anduserstohelpthembuildasharedcontextforinterpretation.

www.ejkm.com

272

ACPIL

ThiernoTounkara

4.7 Levelofdescriptionofspecificknowledge
It may be difficult for experts to explicit some sets of knowledge without strong link to situations they
experienced.Forthosesetsofknowledgewithhighdegreeofspecificity,knowledgereceiverhastomakean
important cognitive effort to generalize (abstract) the knowledge and to recontextualize it for his personal
use.

Weproposethreeactivitiestofacilitatethisabstractionstep:

Identifyingsetsofspecificknowledgealreadycodified

Elicitingwithexpertsgeneralprincipleswhichguidetheuseofidentifiedspecificknowledge

Identifyingandillustratingwithexpertsotherpossiblecontextsofuse

4.8 Exchangechannelstoincreasediffusion/transfer
Communication and transmission channels are necessary to accelerate knowledge transfer. They are an
importantbasisfor:

Elaborationofasharedcontextforinterpretation

Legitimizationofcapturedknowledgeasbestpractice

Evolutionofcodifiedknowledgethroughsocialinteractions.

Inthegridbelow,wesynthesisekeypointstoanalyseforcodifiedknowledgetransferefficiency.
Table1:Analysisgridforcodifiedknowledgetransfer

Codifiedknowledge

Readersprofiles

Exchangechannels

ActivitiesforefficiencyofcodifiedKnowledgetransfer
Complexknowledge
identifysetsofknowledgewithhighlevelofcomplexity
explicitandillustrateassociatedroutines
createasharedcontextforinterpretation(developinteractionsbetweenexpertsand
knowledgereceivers)
Specificknowledge
Identifysetsofknowledgewithhighdegreeofdependencewiththeknowledgesources
contextofuse.
Explicitgeneralprinciplesassociatedtospecificknowledge
Identifyandillustrateotherpossiblecontextsofuse
Background
Professionalbackground
levelofexpertiseofthereaderintheknowledgedomain
degreeoffamiliaritywithknowledgeengineeringmodels
Contextsofuse
Identifyvariousworksituationswherethecodifiedknowledgewouldbeusefulforthe
reader.
Definepreferencesforlogicalstructuring
Definepreferencesforhisunderstandingprofile
Visualrepresentationofknowledge?
Textualrepresentationofknowledge?
Audiopreference(multimedia)?
Illustrationwithconcretecasestudies?
Identifyexistingcommunicationandtransmissionchannels
Stimulatesocialinteractionsbetweenknowledgesources(experts)andreaders

5. Methodologyforknowledgetransferefficiency
Wepropose,here,anempiricalmethodologyfortransferandappropriationofcodifiedknowledgereferential
at individual and organizational levels. It is a two steps approache (rewriting and sharing), guided by the
previousanalysisgrid(table1)andsupportedbyasetofmethodologicaltoolstestedinseveralcompanies
andinvariouscontextswiththeKnowledgeManagementClub,inFrance.

www.ejkm.com

273

ISSN14794411

ElectronicJournalofKnowledgeManagementVolume11Issue32013

Executing these two steps transfer methodology supposes, first, that identification and codification of tacit
knowledgearewellperformed.

5.1 Fromknowledgemappingtotacitknowledgecodification
Weidentifytacitknowledgetocaptureusingacartographicapproachtoanalyseknowledgeareasinthefirm.
Then,withknowledgeengineeringtechniques,wecaptureandcodifytacitknowledge.
5.1.1 MappingandEvaluationofknowledgedomains
Werefertothedefinitionofknowledgecartographygivenby(Speel1999):knowledgemappingisdefinedas
theprocess,methodsandtoolsforanalyzingknowledgeareasinordertodiscoverfeaturesormeaningandto
visualize them in a comprehensive, transparent form such that the businessrelevant features are clearly
highlighted.

WehaveaDomainorientedapproach:wemakeananalysisfromamassofinformationinordertoorganize
itinlogicdifferentfromthefunctionalapproach.Infact,thegoalistoignorethefunctionalstructureofthe
firm,groupingactivitiesintoknowledgedomains.Thistaskdemandsanimportantcapacityofanalysisbecause
itsnotanaturalprocess.

Knowledgedomainsmapisavisualrepresentationbyoperationalactorsofknowledgedomainstheyconsider
essential for their activities. They are grouped according to a common finality on the same theme of
knowledge.Accordingtotheprecisionrequired,adomaincanbedividedintosubdomainsandathemeinto
subthemes.
For each knowledgedomain, we make a synthesis ofthecollective perception(of operational actors) about
theknowledgedomaincriticality.Itistheresultofaqualitative(collectedarguments)andquantitativeanalysis
relyingonaCriticalKnowledgefactorsgrid(Tounkara,IsckiaandErmine2009).Thisgridhasbeenperformed
and validatedin many French and Forein companies. The Critical Knowledge Factor grid contains20 criteria
regroupedinfourthematicaxes(table2)
Table2:Thecriticalknowledgefactorsgrid
Thematicaxes
Rarity

Utility

Difficultytocapture
knowledge

Natureofknowledge

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.

Criteria
Numberandavailabilityofexperts
Externalization
Leadership
Originality
Confidentiality
Adequacywithstrategicobjectives
Valuecreation
Emergence
Adaptability
Use
identificationofknowledgesources
Mobilizationofnetworks
Tacitknowledge
Importanceoftangibleknowledgesources
Rapidityofobsolescence
Depth
Complexity
Difficultyofappropriation
Importanceofpastexperiences
Environmentdependency

Eachcriterionisevaluatedaccordingtoascalecomposedof4levels,representingthedegreeofrealizationof
thecriterion.Eachevaluationofacriterionisbasedononequestion.Eachlevelisexpressedbyaclearand
syntheticsentencebyavoidingthevaguetermsandwhichleadtoconfusion(ratingdescription)

Last,welistinatable,knowledgedomainsconcernedbyspecificitiesitcouldbeinterestingtohighlightwhen
consideringtheoperationalactorspointsofview:domainswithgreatexpertise,domainstobevalorised,very
vulnerable domains or domains that need to improve/adapt methods for training courses and knowledge

www.ejkm.com

274

ACPIL

ThiernoTounkara

transfer. This table is a basis for a more refined analysis and for identification of suitable knowledge
managementactions:

Codificationtransferwhenitisaboutactionsforacquisition,preservationorsharing

Organizationwhenitwasmanagerialactions

TrainingRecruitment when actions are dealing with learning systems, recruitment for new
competencies

Innovationwhenactionsaredealingwithcreativity,environmentscanning,etc.

5.1.2 Capturingandcodifyingtacitknowledgedomain
Firstweidentifyexpertstointerviewfortacitknowledgecapturing:theyareknowledgesourcesandwillbe
authorsofthecodifiedknowledgereferential.Wemakeindividualinterviews.

We define goal and scope of the codification sessions. During these 2 hours sessions, there are strong
interactions with experts to identify and formalize the different types of tacit sets of knowledge. Using
knowledgeengineeringtechniques(asCommonKads,forexample)andtheirassociatedknowledgemodelswe
codifytacitknowledgemodels(4.1)

Thecodifiedknowledgereferentialisthenreadbyotherexpertswhowilladdcommentsandthenrevalidated
byauthors(knowledgesources)ofthecodifiedknowledgereferential.

Theresultisacodifiedknowledgereferential:

Reflectingtheknowledgedomainandthetacitexperienceofoneormanyexperts;

Structuredintochapterscorrespondingtocrucialtacitsetsofknowledgeidentifiedwithexperts.

5.2 Adaptingknowledgereferentialtoreaders(rewritingapproach)
Thisapproachreliesontwosteps:

Characterisationofreaders:

It is an important step for defining readers profiles (Background, context of uses, preferences for logical
structuring).

Elaborationofspecificationsforrewriting

Thegoal,here,istodefine:

Additionalcontentsforthedescriptionofhighlycomplexandspecificknowledge

Additionalillustrations(casestudies,videos)toelaborate

Alogicalstructuringforthecodifiedknowledgereferential

Rewritingandvalidationofthenewknowledgereferential

5.3 Sharingtheknowledgereferential(sharingapproach)
Oursharingapproachhasonemaingoal:createasharedcontextforknowledgeinterpretationtomakeeasier
andaccelerateorganizationallearning.

Werelyonthreeprinciples:

Aclearvisionofactorsinvolvedintheprocessoftransfer

Wecanidentifythreegroupsofkeyactors:

Knowledge sources who are experts or specialists interviewed to capture tacit knowledge; they are
authorsofthecodifiedknowledgereferential.

Knowledge readers are knowledge receivers selected to contribute to the adaptation of the codified
knowledgereferential;theyareapoolofrewriters.

www.ejkm.com

275

ISSN14794411

ElectronicJournalofKnowledgeManagementVolume11Issue32013

Otherknowledgereceiverswhoarepotentialfutureusers(otherteammembers,newemployees,etc.)

Anadequatestructuringofexchangesbetweengroupsofactors

The goal is to formalize situations of exchange which will lead to a collective good appropriation and a
legitimization of captured knowledge.Clear and precise objectives must be defined for each formalized
situation(table3).

Usingadequatechannelsinregardtothepurposeoftheknowledgetransfer

Foreachsituationofexchange,werecommendtoselectthemostsuitabletransmissionchannel(Informalor
formal,personalorimpersonal)toincreaseappropriationandtransferability(table3).
Table3:Formalizationofexchangestodevelopasharedcontextforknowledgeinterpretation
Situationof
exchange
Presentationof
thecodified
referentialto
knowledgereaders

Exchanges
between
knowledgereaders
toadaptthe
codifiedreferential

Sharingthere
writedreferential
withpotential
futureusers

Sharinglearned
lessonswhenusing
ofthecodified
referential

Objectives
Thegoalis,forknowledgereaders,tounderstandtheobjectives,
scopeandcontentofthecapturedknowledge.Knowledgesources
(experts/specialists)present,commentthereferentialandclarifysets
ofcomplex/specificknowledgebygivingexamplesanddifferent
contextsofuse.
Thispresentationinitiatestheprocessofelaborationofashared
contextforknowledgeinterpretationandisimportantforitssuccess.
Manysessionscanbeusefultohaveacollectiveunderstandingofthe
codifiedreferential.
Knowledgereadersareinvolvedinacollaborativeworkwhichwilllead
totheadaptationofthecodifiedknowledgereferential.
Here,theyidentifycomplexandspecificsetsofknowledgeandtryto
makethemmoreexplicit:
buildingcasestudiescollectively
illustratingothercontextsofuserelyingontheirownexperiences
changingthelogicalstrutcturingofsomechapters
etc.
Thegoalistosharetherewritedandstabilizedcodifiedreferential
withotherknowledgereceiverswhoarepotentialfutureusers.The
poolofrewritershastodefineapprenticeshepobjectivesanddelimit
thetheappropriatesetsofknowledgetheywillhavetofocuson.
Trainingsessionscanbeappropriatetoexchangewithpotentialfuture
users.
Theobjectiveistofaciltatefutureevolutionsofthecodified
knowledgereferentialbycapitalizinglearnedlessonsofactorsusingit.
Exhanges(eveninformal)betweenthedifferentgroupsofactorsmust
beorganizedperiodicallytoidentify:
newwaysofdoingmoreefficient(evolutions)
newapplications/newcontextsofuse
difficultiesmet
etc.

Transmission
Channel
Formalseminars

Formalseminars

Trainingsessions
Onlinetraining
sessions

Informalseminars
Coffeebreak
conversations
Onlineforums
Formalseminars

Thebelowfiguresynthetisesourtwostepsmethodologyforknowledgetransferefficiency.

6. HydroQuebeccasestudy
HydroQuebecisoneofthebiggestelectricityproduceranddistributorinNorthAmerica.HydroQuebecisa
publiccompanyandhisprincipalshareholderistheQuebecGovernment.

TheHydroQuebecstudyispartofabiggerprojectsupportforcriticalknowledgecaptureleadedbythe
Human Resources Department. The study lasted 2 months and concerned one operational unity. It was
motivatedbythefutureretirementofoneoftheunitysexperts.

Interviewing this expert and using knowledge engineering techniques, a codified knowledge referential was
elaboratedinafirststage.

www.ejkm.com

276

ACPIL

ThiernoTounkara

Input or output for an activity


Analysis grid
for codified
knowledge
transfer

Characterisation
of readers

B is a sub activity of A
Actor

Readersprofiles

Elaboration of
specifications for
re-rewriting

Codified
knowledge
referential

Redefinition of
content

-Identify complex knowledge and


complete describtion
-Identify specific knowledge and
complete describtion

Illustration

-Identify needed illustrations


-Develop concrete case
studies

Reorganization of
content

-Additional
contents and
illustrations
- Logical
structuring

Re-writing

New Codified
knowledge
referential

-Define a logical structuring


- Define a good balance between
texts, visual models and
multimedia
Exchange
channels
(informal or
formal)

Sharing of new
codified knowledge
referential

Experts

Readers

Figure1:Methodologyforcodifiedknowledgetransfer
Twoobjectiveswereassignedtoourstudy:

Optimisetheuseofthecodifiedknowledgereferential

Acceleratethetransferoftheknowledgereferentialtofivefutureexpertsofthesameunity

6.1 Approach
Wemixedquestionnairesandinterviewstoimplementourknowledgetransfermethodology.

Questionnaires were used with the 5 future experts (readers of the codified knowledge referential) to
characterizetheirprofiles(ProfessionalandstudyBackground,contextsofuseofthereferential,etc.).Fourof
themhadatechnicalbackgroundandthefifthhadamanagerialbackground.

Interviewingthefivefutureexpertshelpedusidentifyadditionalneedsforthecodifiedknowledgereferential
more.Interviewswereguidedbyouranalysisgridforcodifiedknowledgetransfer.

6.2 Results
Ourstudyleadedtonewspecificationsforthecodifiedknowledgereferentialtoincreaseitstransferability(
table4).
Table4:Synthesisofnewspecificationsforthecodifiedknowledgereferential

AbouttheContent

Aboutthestructuring

www.ejkm.com

Newspecificationsforthecodifiedknowledgereferential
alistofadditionaldescriptionstowrite
alistofmissingcriticalknowledgetointegrateinthereferential(needingnewinterviews
oftheexpert)
alistofsetsofknowledgetocompletebyconcretecasestudies

Despitethemultiplicityofmodelsinthecodifiedknowledgereferential,readersfind
modelsreadablebecauseeachthemwasassociatedwithatextualdescription.
Forthelogicalstructuringofthereferential,theypreferredaproblemsolvingapproach.

277

ISSN14794411

ElectronicJournalofKnowledgeManagementVolume11Issue32013

Aboutthenewreferential
sharing

Newspecificationsforthecodifiedknowledgereferential
Itwasproposed:
acollectiveandcollaborativerewritingofthecodifiedknowledgereferential(withthe
expert);
aKnowledgemanagementfacilitatorwhowilldrivethesharingstepandcoordinate
theevolutionofthereferential.

TheprincipalimplicationofHydroQuebecCaseStudyisthatthemethodologyforknowledgetransfermustbe
performedasacollectiveandcollaborativeprocesswhichinvolvesthreecategoriesofactors:

knowledgesources(experts/specialists)

knowledgemanagementfacilitators

Knowledgereaders

7. Conclusion
Inourarticle,wefirstunderliefactorswhichaffectefficiencyofcodifiedknowledgetransfer.Thenwepropose
anoperationalmethodologytooptimizecontinuitybetweenknowledgecodificationandknowledgetransfer.
Thismethodologyreliesontworobusttheoreticalframeworks:theSECImodelofNonakaandTakeuchiand
theBoisotISpaceKMmodel(3.2).

Integratedwithknowledgeengineeringtechniques,themethodologycanenhanceknowledgecodificationby
leadingtotheelaborationofapertinentsharedcontextforknowledgeinterpretation.

TheHydroQuebeccasestudyhighlightstheimportanceofdefininganappropriateorganizationtosupportthe
knowledgetransferprocess.

Economic aspect can be a limitation of the knowledge transfer model we propose. Implementing such a
processtransfercanbeaheavyinvestmentforcompanies:

Anorganization(itcanbeaformalcommunity)mustbesettledandthisinvolvesidentifyingactorsand
definingforthemrolesandresponsbilities

Actors must be available for the codification, the rewriting and the evolution of the knowledge
referential.

In the next step of our research, we will focus on the evaluation of knowledge transfer efficiency after an
implementationoftheproposedmethodology.

Acknowledgments
ThankstoJeanLucRichard,theconsultantwithwhomIperformedthecasestudyatHydroQuebec

Thanks to the Knowledge Management Club (in France) which gave us real opportunities to exchange with
variousbusinesscompanies.

References
Alavi,M.andLeidner,D.E.(2001)Review:KnowledgeManagementandKnowledgeManagementSystems:Conceptual
FoundationsandResearchIssues,MISQuarterly,vol.25,no.1,pp.107136.
Argote,L.,Ingram,P.,Levine,J.M.,andMoreland,R.L.(2000)KnowledgeTransferinOrganizations:Learningfromthe
ExperienceofOthers,OrganizationalBehaviorandHumanDecisionProcesses,vol.82,no.1,18.
Boisot,M.(1998)KnowledgeAssets:SecuringCompetitiveAdvantage,intheInformationEconomy,Oxford:Oxford
UniversityPress.
Cohen,WM.andLevinthal,D.A.(1990)AbsorptiveCapacity:ANewPerspectiveonLearningandInnovation,Administrative
ScienceQuarterty(35),pp,128152,
Dalkir,K.(2011)KnowledgeManagementinTheoryandPractice,seccondedition,TheMITpressCambridge,
Massachussets.
Dieng,R.,Corby,O.,Giboin,A.,GolebiIowsSka,J.,Matta,N.andRibiereM.(2000)Mthodesetoutilspourlagestiondes
connaissances,Dunod.
Fahey,L.andPrusak,L.(1998)TheElevenDeadliestSinsofKnowledgeManagement,CaiiforniaManagementReview{40.3),
pp.265276.

www.ejkm.com

278

ACPIL

ThiernoTounkara

Garud,R.(1997)Onthedistinctionbetweenknowhow,knowwhat,andknowwhy,InA.HuffandJ.Walsheds.,Advances
inStrategicManagement,Greenwich,CT,JAIPress.
Gupta,A.andGovindarajan,V.(2000)KnowledgeFlowswithinMultinationalCorporations,StrategicManagementJournal
(21),pp.473496.
Harrison,A.andHu,Q.(2012)KnowledgeTransferwithinOrganizations:ASocialNetworkPerspective,45thHawaii
InternationalConferenceonSystemSciences.
Holtham,C.andCourtney,N.(1998)TheExecutiveLearningLadder:AKnowledgeCreationProcessGroundedinthe
StrategicInformationSystemsDomain,inProceedingsoftheFourthAmericasConferenceonInformationSystems.E.
HoadleyandI.Benbasat(eds.),Baltimore,MD,pp.594597.
Nonaka,I,,andTakeuchi,H.(1995)TheKnowledgeCreatingCompany:HowJapaneseCompaniesCreatetheDynamicsof
Innovation.OxfordUniversityPress,NewYork.
Polanyi,M.(1967)TheTacitDimension,RoutledgeandKeoanPaul,London.
Reed,R.&DeFillippi,R.J.(1990)Causalambiguity,barrierstoimitation,andsustainablecompetitiveadvantage,Academy
ofManagementReview,15(1),pp.88102.
Simonin,B.L.(1999)AmbiguityandtheProcessofKnowledgeTransferinStrategicAlliance,StrategicManagementJournal,
pp.595623.
Speel,PH.,Shadbolt,N.,DeVries,W.,VanDam,PH.andOhara,K.(1999)KnowledgeMappingforindustrialpurpose.
October99,Banff,Canada.ConfrenceKAW99.
Szulanski,G.(1996)Exploringinternalstickiness:Impedimentstothetransferofbestpracticewithinthefirm,Strategic
ManagementJournal,vol.17,pp.2743.
Teece,D.(1977)TechnologyTransferbyMultinationalFirms:theResourceCostofInternationalTechnologyTransfer,
EconomicJournal,Vol.87,pp.242261.
Tounkara,T.,Isckia,T.andErmine,J.L.(2009)FromStrategytoKnowledgeManagementPlan:howtocreatestrategic
alignment?ICICKM2009,Montreal,12October2009.
Tounkara,T.,Matta,N.,Ermine,J.L.andCoppens,C.(2002)LappropriationdesconnaissancesavecMASK,EGC2002,
Montpellier2123.
Vance,D.andEynon,J.(1998)OntheRequirementsofKnowledgeTransferUsingIS:ASchemaWherebySuchTransferis
Enhanced."InProceedingsoftheFourthAmericasConferenceonInformationSystems,E,HoadleyandI.Benbasat
(eds),Baltimore,MD,August.pp,632634.
Zack,M.H.(1999)Managingcodifiedknowledge,SloanManagementReview,vol.40,no.4,1999,pp.4558.
Zander,U.andKogut,B.(1995)KnowledgeandtheSpeedoftheTransferandImitationofOrganizationCapabilities,
OrganizationScience,Vol.6,pp.7692,1995.

www.ejkm.com

279

ISSN14794411