(2011) neumann_knowledge management at individual level to support logistics problem solving

7/29/2019 (2011) Neumann_Knowledge Management at Individual Level to Support Logistics Problem Solving

1/20

83

POSLOVNA IZVRSNOST ZAGREB, GOD. V (2011) BR. 1 Neumann G.: Knowledge management at individual level to support...

KNOWLEDGE MANAGEMENT AT INDIVIDUAL LEVELTO SUPPORT LOGISTICS PROBLEM SOLVING1

Gaby Neumann2

UDC /UDK: 001:658JEL classification / JEL klasifikacija: 030

Review / Pregledni radReceived / Primljeno: September 29, 2011 / 29. rujna 2011Accepted or publishing / Prihvaeno za tisak: November 10, 2011 / 10. studenoga 2011.

Summary

Logistics problem solving is a knowledge-intensive process which usually requires alarge amount of experience with the problem-solving person. Consequently, a proper man-agement of this knowledge and experience at individual level is a chance and a challengefor improving a persons problem-solving capability. Against this background the paper

aims to help knowledge in becoming productive within logistics problem-solving processes.Based upon the statement by logistics company managers characterising knowledge as akey resource in their processes that mainly evolves through projects, the paper identifieslogistics projects as problem solving processes and specifies the type of problems logisticshas to deal with. Further on, chances from applying knowledge management approachesin logistics problem solving are discussed to answer questions for (i) how to encourage andsupport persons to bring in their knowledge including to raise awareness of the knowledgethey have already, (ii) how to mediate between different types of knowledge stakeholdersand repositories, and (iii) how to enable and support the access to and use of knowledge.Tis is illustrated by examples from logistics simulation projects where simulation modelsare to be seen as knowledge repository and an intelligent human-computer dialogue isrequired for accessing this knowledge.

Key words: knowledge management, problem solving, logistics simulation, knowl-edge repository, knowledge sharing.

1 Tis paper was presented at International Conerence Managing Services in the Knowledge Economy,(MSKE 2011) which was held rom July 13th to 15th 2011, Vila Nova de Famalico, Portugal.

2 Gaby Neumann, Engineering Logistics, echnical University o Applied Sciences Wildau, Germany, E-

mail: [email protected]


2/20

84


1. INTRODUCTION AND MOTIVATION

In general, knowledge management describes the conscious, systematic andstrategically anchored handling o knowledge considering people, organization andtechnology in order to encourage individual, social and organizational learning proc-esses (Reinmann-Rothmeier et al. 2001). Whereas knowledge management is typicallydiscussed as organizational matter, this paper ocuses on the application o knowledgemanagement concepts, methods and approaches at personal level. More specificallythe challenge consists in managing a persons knowledge in a way unlocking its devel-opmental potential and giving it value. In order to transorm knowledge into a valu-able organizational (or individual) asset, knowledge management aims at ormalizingand accessing experience, knowledge, and expertise that create new capabilities, enablesuperior perormance, encourage innovation, and enhance customer value (Beckman

1999). For this, knowledge management provides a wide variety o methods and toolsor creating and acquiring, storing and retrieving, exchanging, or using and re-usingknowledge as a strategic resource. Here, knowledge is generally defined as reasoningabout inormation and data to actively enable perormance, problem-solving, decision-making, learning, and teaching (Beckman 1999).

Figure 1: Influences and spheres o logistics problem solving exemplarilyrelated to logistics planning

PROBLEM COMPETENCE

system/process

SOLUTION

additional

EXPERIENCE

general knowledgeabout planning

planning knowledge specificto the organisation

knowledge onthe problem

planner model

general knowledgeabout planning (increased)

planning knowledge specificto the organisation (increased)

knowledge onthe solution

knowledge onthe planner

the problem's

point of view

the planner's

point of view

problemsolvingprocess

learningprocess

objective

elements/influences

subjective

elements/influences

problem solvingmethods and tools

collaborationand communicationmethods and tools

knowledgemanagement

methods and tools

methods and toolsfor teaching, learning

and acquisition ofknowledge

thetools'

pointofview

knowledge about actions

knowledge about tools

the knowledge

point of view


3/20

85


Tere is no doubt about the act that problem solving is a knowledge-intensivepiece o brain work. On one hand problem solving especially successul, effi cient,effective problem solving requires to bring in and apply general and domain-spe-

cific knowledge the problem-solving person has gained rom learning, but also romproblem-solving experience. On the other hand each problem-solving process also isa knowledge-creating process as new knowledge grows rom understanding the prob-lem and its sources, rom finding a suitable or the best solution to this problem androm choosing the right method(s) or tool(s) and correctly applying it (them) in acertain problem-solving step (see Figure 1). Te amount, type and level o knowledgerequired to solve a particular problem and acquired rom dealing with it depends onboth the problems degree o diffi culty and complexity and the level o knowledge andexperience the problem-solving person has. Tereore, each problem is settled downin and linked to a certain domain (objective component) and depends on the prob-

lem solving persons background (subjective component). Te latter particularly canbe characterized by the individual competences in identiying the problem, choosinga suitable problem-solving path, applying appropriate methods and tools, evaluatingcandidate solutions and sel-reflecting the problem-solving process as run through inorder to derive experiences gained and lessons learned (see Figure 2).

Figure 2: Problem-solving steps and problem-solving personality

Consequently, knowledge management is expected to provide a tremendouscontribution to the improvement o problem solving capability. o learn rom previ-ous experience, to know about valuable experts, their location and expertise or gettingthem involved in running projects or new tasks, to better understand pre-conditions,settings and decisions rom the past and with relevance to todays activities, to easilyidentiy and get access to the right knowledge exactly when it is needed these aresources or increasing a persons, a teams or an organizations problem-solving com-petence and perormance. Despite o this understanding and although problem-solv-ing and knowledge management methods and tools are available in growing amountand more and more easy-accessing way, the developmental potential o knowledge

problemfinding

problemsolving

assessmentand evaluation

implementationof the solution

problemsensibility

versatile knowledge,flexible in thinking

ability to assesscritically

ability toassert oneself

problem solving process

features of the personality


4/20

86


in problem solving is ofen not recognised and remains locked in the different typeso (potential) knowledge sources. Adopting a statement on human needs or compu-ter technology by Shneiderman (2002) the link between knowledge management and

problem solving can generally be described as ollows: the old discussion about howto support problem solving is about what (knowledge management) tools can do; thenew discussion about how to support problem solving is (and must be) about whatkind o problem-solving support people really need.

Against this background the paper discusses needs and chances or knowledge-based and knowledge-ocussed support to problem solving particularly with regard tochallenges in a complex problem scenario. Here, logistics is being used as applicationarea as logistics problems are quite ofen o very complex nature leading to challengingproblem-solving processes with always new, multi-step procedures and the need or

a variety o methods and tools. In addition to this, logistics problem solving requiresthe ability to operate in an ill-structured situation. Te ollowing sections will intro-duce into problems and problem solving in logistics (Section 2), beore approaches orunlocking the developmental potential o knowledge in logistics problem solving arediscussed. For this, logistics simulation projects are used as examples to illustrate howinternal knowledge rom methods, models and solutions can be accessed (Section 3),how the use o knowledge or purposeul experimentation might be supported (Sec-tion 4), and how existent knowledge might be distributed through knowledge exchangeand stakeholder development (Section5). Section 6 summarizes discussions and drawsconclusions on barriers and chances or helping knowledge in becoming productive.

2. PROBLEMS AND PROBLEM SOLVING IN LOGISTICS

In general, a person aces a problem when a current situation is not satisying,but or some reasons it momentarily cannot be changed into a desired one. Due to thecomplicated structure o potential solutions, problems related to logistics planning andoperation but also corresponding problem-solving processes are o complex nature.Tey are subject to a variety o influences, demands and circumstances entailing stepsto configure, dimension or evaluate in dependence on what is given and what has to be

defined. According to this, our classes o problems o increasing level o diffi culty andcomplexity can be identified:

asks are challenges that require applying known methods within a clearlydefined procedure in order to change a well-structured current situation intoa known target situation. Te problem-solving person has all knowledge andcompetence necessary to understand the problem and got or the solution.

Interpolation problems are characterized by an unknown procedure or howto deal with them although current as well as target situations are knownand a collection o problem-solving methods (or tools) is available. Here,analytical thinking is required by the problem-solving person in order tofind a suitable way to come rom the problem to the solution by purposeullyapplying the methods.


5/20

87


For synthesis problems both, methods/tools and problem-solving path, areunknown and the current situation might be ill-structured. Te problem-sol-ving person has just a clear idea and understanding o the target situation, i.e.

s/he knows where s/he wants to go to. o get there mainly creative thinkingis required in order to understand the current situation, select appropriatemethods and find a working procedure or solving the problem.

Dialectic problems orm the most challenging type o problems. Here, neithercurrent nor target situations are clearly defined and neither methods nor aprocedure solving the problem are known. Tereore, the problem-solvingperson must combine analytical and creative thinking capabilities in orderto approach and find a solution.

Figure 3: Steps and challenges in the problem solving process

Logistics problems might belong to any o those classes depending on thesituation and on the persons level o knowledge, competence and experience. Tat iswhy logistics problem solving typically can be characterised as a phased process oloops, as a process developing variants or and versions o both the logistics processand the logistics system. Analysing steps and creative, evaluating steps or synthesiswith partially changing cognitive problems and views take turns. As a rule searchingor appropriate and suitable components or solving the problem comes first. Afer thatthe defined components must be rendered consistent with one another in order orthe over-all unctionality as given in the problem specification to be realised. Methodsavailable or this process are many and diverse; they are selected according to the typeo problem (i.e. design, decision-making, parameter-setting, selection, and evaluation),

method'sselection

cognitivetarget

solvingmethod

use of themethod results

cognitionanalysis of results

conclusions

methods poolfor solving

the problem

analytical calculations

statistical calculations

selections

evaluations (ranking)

problem solv ing processprocessstructure

system'sstructure

processparameter

system'sparameter

evaluation of theselection

evaluation of themethod's use

evaluation ofthe solution


6/20

88


the current state-o-the-solution and available input data and inormation. Te stocko methods includes simple deterministic as well as complicated stochastic calcu-lation models in the orm o analytical ormulae and extends as ar as simulation

models (see Figure 3). Applying the appropriate method in the right way brings outresults which need to be analysed to derive findings that match the identified targetor in other words solves the initial problem. But apart rom the challenge o methodselection and application another challenge consists in dealing with a dynamicallychanging environment logistics problem-solving is embedded in. ime pressure andlimitation o resources define restrictions to be taken into consideration. Even at anearly planning stage and on the basis o little inormation only, a maximum o reli-ability and quality must be reached with minimal effort and in minimal time. In theend the challenge usually does not consist in searching or the optimal solution intheory, but or an appropriate, practicable and realizable one which solves the identi-

fied problem properly and effi ciently.

Within these constraints discrete event simulation is in many cases an appro-priate method and tool allowing the chronological reproduction o real processes andsystems realistically and accurately in any detail. Processes can saely be shown andspeeded up; they are as repeatable and variable as desired. Pre-condition or this is anappropriate, valid simulation model representing an existing or planned logistics solu-tion at the required level o detail. Input data such as time parameters, other quantita-tive parameters or process rules are either obtained rom analysing real systems (or

aim of the simulation

questions addressed to the simulation

simulationmodel

knowledge/experience

of the simulation user

simulation

system

simulationoutput

methodical

support

findings fromsimulation

Figure 4: Impact o the simulation user on the outcome o a simulation project


7/20

89


modelling an existing system) or derived rom settings and expert assumptions (when

modelling a planned system). Purposeul experimentation with the model produces a

wide variety o output data that need to be transerred into results by context-sensitive

and problem-specific interpretation on the basis o statistical compression and graphi-cal representations. Additionally, results can clearly be visualized by means o anima-

tion which supports model validation, experimentation, understanding o output and

presentation o results to the same extent. Simulation methodology is to be applied to

logistics problem solving whenever in reality not (yet) extant logistics systems or con-

texts are to be investigated, cause-effect chains are highly complex in nature, uture or

visionary scenarios are to be observed, alternative design variants are to be analysed,

or a long-term analysis o system behaviour needs to be run.

Due to its prominent role in logistics problem solving logistics simulation is

used as example to discuss and demonstrate the needs and chance or knowledge-based support. Tis is even more useul as one o todays challenges regarding simu-

lation consists in seeing it in the context o human-centred processes. Tis requires

understanding simulation as a complex problem-solving, knowledge-generation and

learning process but simultaneously as a tool to support teaching and as the subject

o knowledge application. Furthermore, human resources involved in a simulation

project are the key actors or its success and effi ciency. As shown in Figure 4 the

simulation user impacts the outcome o a simulation project in different ways and at

different stages. Te final appearance o a simulation model (being implemented by

use o a particular simulation package in order to answer certain questions address tothe simulation) strongly depends on the modelling experience and philosophy o the

modelling person. Based upon this model and ollowing the users experimentation

strategy simulation output is being produced, the interpretation and understanding o

which again requires direct involvement o and input by the user.

Te clue to the successul implementation o those knowledge management

procedures is ofen an appropriate (supporting) environment and climate in the organ-

ization. Concerning this, there is a greater need or a cultural shif than or additional

sofware tools and I solutions. Tis is also confirmed by the results rom a survey on

the role o knowledge in logistics companies (see Neumann and om 2006). Here,nearly all (o the not so many) respondents declared knowledge was a key resource or

the companys perormance: more than 75% o the responding companies (at least in

the logistics managers opinions) characterize knowledge management as supportive

to all activities and helpul to better perorm at the market; urthermore knowledge

changes and improves all the time (in more than 60% o the cases) and/or develops

with particular activities such as projects (in 50% o the cases). Consequently, it is

worth to go into detail when discussing about how knowledge-based support can be

provided to logistics problem-solving in general and logistics simulation in particu-

lar. For this typical knowledge management activities have been selected: accessing

knowledge rom a repository, using knowledge or planning, running and analysing

experiments, and distributing knowledge through knowledge exchange and urther

development o knowledge stakeholders.


8/20

90


3. ACCESSING KNOWLEDGE: SIMULATION MODELS ASKNOWLEDGE REPOSITORIES

A simulation model is more than just a tool necessary to achieve certain objec-tives o experimentation and cognition. In the course o a simulation project the simu-lation model is developed, modified, used, evaluated and extended within an ongoingprocess. Tereore, it is also a kind o dynamic repository containing knowledge aboutparameters, causal relations and decision rules gathered through purposeul experi-ments. In the end, knowledge stored in the simulation model can be considered prov-en, independently o whether it was developed by the domain expert him- or herselor by a consultant simulation expert. Unortunately, this knowledge is usually not verywell documented and thereore does exist implicitly only inside the simulation model.

Knowledge represented by or stored in a simulation model is o both explic-it and implicit natures. According to Nonaka and akeuchi (1995) implicit or tacitknowledge is that kind o knowledge that a person carries in his or her mind ofen noteven being aware o it, that the person cannot express and that is thereore not directlyexternalizable. It is typically based on experiences and expressed in the orm o actionpatterns and eeling-based decisions. As a consequence the simulation model as theresult o applying this implicit knowledge somehow carries at least parts o it withoutallowing direct access. Indirect access might eventually be possible via analyzing apersons model building and implementation behaviour and interpreting observations.Tis way o accessing simulation knowledge then is more a kind o knowledge explica-

tion strategy helping a person to recognize and discover him/hersel intuitive decisionsand actions. But this is diffi cult and very challenging i possible at all.

Instead, we want to ocus on a more promising way to access simulation knowl-edge that is o explicit nature. Explicit knowledge is that kind o knowledge that ex-ists independent o a person, e. g. in the orm o any document. More precisely it isknowledge that has been or can be articulated, codified, stored and accessed by otherpersons. I not articulated yet, this knowledge is still carried by the person, but couldbe externalized by use o the right means. In case o simulation modeling this knowl-edge lays with the model building or implementing person and also with the devel-

oper o simulation sofware. It reers to modeling settings and assumptions like sys-tem boundaries and interaction with its environment, system elements and their links,their level o detail in representation, input data, probability distributions etc. And italso covers model implementation decisions, such as the simulation tool to be used,model components to be used, modeling tricks to be applied etc. o get access to thisknowledge as finally represented by the simulation model structured documentationand documentation support are helpul, but there is also the need or understandingwhich knowledge is behind or inside the simulation tool used. On the other hand,explicit simulation knowledge might directly be available rom the simulation modelin the orm o model structure and parameters or simulation parameters and eventu-

ally comments, annotations, speaking names o model components or variables etc.Tereore, it should be subject to automatic extraction into a project report or modeldescription. In the end, both orms o explicit knowledge need to be put together speci-


9/20

91


ying HOW does a simulation model look like in greater detail and relating it to WHYdoes it look like this.

o access explicit knowledge stored in a simulation model that has been imple-mented by use o particular simulation sofware, it might be a good starting point tolook at the conceptual model developed beore. When this model has been representedin a ormalized way and careully validated, it correctly conceptualizes the system andprocesses to be modelled including components, dependencies and partly also param-eters.

Tus, logistics simulation knowledge combines aspects rom a wide variety osubjects and logistics simulation projects require a respective collection o interdisci-plinary expertise. Furthermore, logistics simulation knowledge is dynamic and alwaysevolves in the course o a project. Tat is why it has to be documented continuously

and consequently. Tis requires a common and even ormalizable documentationramework that is adjustable not just to the specific simulation problem, but especiallyto the current state-o-knowledge with each aspect and thus can simultaneously hostdescriptions o different levels-o-detail or different aspects.

Te proposal or how such a documentation ramework (in terms o both,structure and procedure) might look like is presented in Figure 5. Here, a simulationproject meeting is used as example or speciying knowledge and input to be docu-mented: apart rom categories with typical inormation on a meeting, such as loca-tion, time, participants etc., this ramework allows collecting detailed knowledge about

the particular simulation project. For the latter, subject-related knowledge was clearlyseparated rom procedure-related knowledge and individual sub-categories or bothknowledge aspects have been defined. Following the basic structure o a viewpointdescription each knowledge category contains documentation and criticism parts.Whereas the documentation part o procedure-related knowledge represents the mainaspects o a simulation-based problem-solving process in general, the documentationpart o the subject-related knowledge was specifically structured according to the ap-plication area o logistics. Consequently, on the first level it is urther divided intoobject, system, process and environment categories which are suitable to completelydescribe a logistics or supply chain problem and solution. I the proposed documenta-

tion ramework is to be applied to any other application area o simulation, this specificpart would have to be adapted to the relevant elements o problems and their solutionsin this particular area.

I in each project meeting one copy o this template is filled with the particularcontents and results, a growing set o documents with structured simulation knowl-edge is produced. In addition to this, urther documents, written or oral communica-tion outside the meetings and urther external knowledge sources used within theproject can be analyzed in a similar way and thus increase the project-specific knowl-edge collection. From composing all o these individual documents step-by-step andaccording to project progress, a project-accompanying structured documentation oboth the state-o-the-problem (and solution) and the state-o-the-problem-solving(including the state-o-development o the simulation model, methods and tools usedor decisions taken) is derived.


10/20

92


Figure 5: Structure or documenting a simulation project meeting

Source: Neumann 2006

At the same time the state-o-documentation resulting rom each newly addedsource o knowledge also represents the particular state the project had reached at thatpoint in time. With this, not only moment-specific representations o project knowl-edge, but also period-related representations o project progress become storable in aormalized way. Due to the act that knowledge on the problem or solution and knowl-

edge on the project procedure are always jointly documented, not only a purposeulreflection o taken decisions is required, but also a clear explanation o all modifica-tions to the model, procedure or solution initiated by those decisions. ogether withthe corresponding files containing the simulation model this would principally allow

Problem /

Solution

Problem

Solving

Process

GeneralInformation

Minutes of the Meeting

.................

Documentation

Objects

System

Process

Environment

CriticismProblems

Tasks

Documentation

Models and

Methods

Tools

Procedure

Analysing problem

Model building

Validating model

Planning experiments

Running experiments

Interpreting outcomes

Presenting results

Project

Organization

Objectives

Partners

Simulation

customer

Simulation

service

provider

Volume

Schedule

ProjectEnvironment

CriticismProblems

Tasks

Topic

Location

Time

Participants

Agreements

Next meeting


11/20

93


to return to any point in the problem-solving process and continue rom there towardsa different path whenever this seems to be necessary or appropriate.

4. USING KNOWLEDGE: KNOWLEDGEBASEDEXPERIMENTATION

In general, input inormation or a simulation project usually come with thetender specification or are to be identified and generated in the problem definition anddata collection phases o the simulation (Figure 6). Here, the user decides (and bringsin) what is to be taken into consideration or model building and which inormation isrequired or the investigation.

Figure 6: Sources and evolution in simulation knowledge

Simulation experiments, or example, to support logistics planning and opera-tion might be oriented towards modifications in either unctionality or structure orparameters o a model and its components or even in a combination o those varia-tions leading to more complex fields o experiments. o correctly interpret simulationoutput it is necessary to understand what the objectives, parameters and procedures oa certain series o experiments were and to relate this to the results and findings. Here,

analyzing

model buildingimplementing

validating

experimentinginterpreting

simulation

model(s)

experiments

results

reports

presen-

tations

planning

knowledge

op

erating

knowledge

tender

specification

observations

interviews

meetings

talks

e-mail

phone

simulationproject

knowledgeaboutproblemandsolution

application

knowledge

methodological

knowledge

management

knowledge


12/20

94


the simulation user aces the ever challenging task to interpret numerous and diversedata in a way being correct with respect to the underlying subject o the simulationstudy and directly meeting its context. Tese data are usually produced and more or

less clearly presented by the simulation tool in the orm o trace files, condensed statis-tics and perormance measures derived rom them, graphical representations or ani-mation. Problems mainly consist in

1) clearly speciying questions the simulation customer needs to get answered,

2) purposeully choosing measures and selecting data enabling the simulationservice provider to reply to the customers questions, or

3) processing and interpreting data and measures according to the applicationarea and simulation problem.

o overcome these problems and give support in defining simulation goals andunderstanding simulation results, methods and tools are required that are easy to useand able to mediate between knowledge and understanding o the simulation cus-tomer (the logistics expert planning or operating that process and system to be simu-lated) and the simulation expert (the expert rom the point o view o data and theirrepresentation inside computers).

Within this context, it is worth thinking in more detail about what a simulationcustomer (the logistics expert) might look or when analyzing the outcome o simula-tion experiments (Neumann 2005):

ypical events. Te logistics expert specifically looks or moments at whicha defined situation occurs. Tis kind o query can be related, or example, tothe point in time at which the first or last or a specific object enters or leavesthe system as a whole or an element in particular. Other enquiry might beoriented towards identiying the moment when a particular state or combi-nation o states is reached or conditions change as defined.

ypical phases. Te logistics expert is especially interested in periods cha-racterized by a particular situation. In this case s/he asks or the duration othe warm-up period, or the period o time the system, an element or object

is in a particular state, or how long a change o state takes.

Statements. Te logistics expert looks or the global characteristics o proces-ses, system dynamics or object flows such as process type (e.g. steady-state,seasonal changes, terminating/ non-terminating), perormance parameterso resources (e.g. throughput, utilization, availability), parameters o objectflows (e.g. mix o sorts, inter-arrival times, processing times). Tis inorma-tion is usually based on statistics resulting rom trace file analysis and repliesto either a specific or more general enquiry by the user.

When the potential interests o a simulation customer as explained above arecompared, one significant difference emerges: whereas the first two aspects need spe-cific questions ormulated by the logistics expert directly at data level, the last aspectis characterized by usually uzzy questions o principle rom the more global users


13/20

95


point o view. Beore these questions o principle can be answered, they have to betranserred to the data level by explaining them in detail and putting them in terms oconcrete data (Figure 7).

Figure 7: User-data interaction or simulation output analysis

As result o this process o interpretation a set o specific questions is definedwith each o them providing a specific part o the overall answer in which the user isinterested. Questions at data level correspond to results that can be delivered directlyby the simulation even i minor modifications to the simulation model should be re-quired. o derive an answer in principle to a question o principle the respective set ospecific answers needs to be processed urther. Tese steps o additional analysis andcondensing can be understood as a process o re-interpretation to transer results romdata to user level.

analysis of data

interpretation re-interpretation

user level

model level

tracefile

test for data availability

proposal ofwatch dogs

filterfor data

modification of modelsimulation run

converting data

user of the simulation model

data level

?

principlequestion

!

principleanswer

? ??specificquestion

specificquestion

specificquestion ! !!

specificanswer

specificanswer

specificanswer i

nterpretation

layer


14/20

96


All steps o interpretation and re-interpretation aim to link the users (logisticsexperts) point o view to that o the simulation expert. Tey not only require an ap-propriate procedure, but, even more importantly, an interpretative model representing

the application area in which simulation takes place. Tis model needs to be based onknowledge and rules expressed in the users individual expertise, but also in general-ized knowledge o the (logistics) organization regarding design constraints or systembehaviour and the experience o the simulation expert derived rom prior simulations.As this knowledge might not only be o explicit nature but also comprise implicit ortacit knowledge simulation users as individuals or team need to remain involved in thesteps o interpretation and re-interpretation at least. Whereas explicit knowledge mightbe transerred into rules and algorithms, tacit knowledge cannot be separated rom itsowner and thereore requires direct involvement o the knowledge holder in the inter-pretation process. More specifically this means support is required or translating anyquestion o principle into corresponding specific (data-related) questions as well as orderiving answers o principle rom a number o specific (data-related) answers. Althougha set o (standard) translation rules might be known, ormalized and put into the rulebase already, always urther questions remain that are unknown to the rule base yet.Here, the logistics expert needs support in (i) correctly ormulating the right questionand (ii) getting the ull picture rom the puzzle o available data and their analysis.

One approach or enabling this could be based on viewpoint descriptions.Viewpoint descriptions were introduced into model validation as a new kind o com-munication and interaction between the human observer o simulation results and the

computer as the simulation model using authority that was called oracle-based modelmodification (Helms and Strothotte 1992). Here, the principle idea is that the userpresents his or her observations (in the animation) as a viewpoint description to thecomputer that initiates a reasoning process. Tis results in definition and realization onecessary changes to the simulation model in an ongoing user-computer dialogue. Temain advantage o this concept lies in the reduced requirements or rule-base defini-tion. Tose aspects that easily can be ormalized (e.g. typical quantitative observationsor unambiguous logical dependencies) are translated into questions to the user (Whatis it s/he is interested in?) or various orms o result presentation (as figures or dia-

grams), whereas those that are non-imaginable yet or individual to the user or simplyhard to ormalize need not to be included to provide meaningul support to the user.Tere is no need to completely speciy all possible situations, views and problems inadvance, because the person who deals with simulation output brings in additionalknowledge, experience and creativity or coping with non-standard challenges. Evenurther, this way the rule-base continuously grows as it learns rom all applicationsand especially rom those that were not involved yet. On the other side the user ben-efits rom prior experience and knowledge represented in the computer by receivinghints on what to look at based upon questions other users had asked or which were ointerest in earlier investigations.

Tis approach helps in designing the interpretation layer or mediating betweensimulation customer and simulation model or output no matter how many data havebeen gathered and how big the trace file grew. As discussed simulation results derived


15/20

97


rom running experiments by use o a particular simulation model are as good as theyfinally respond to the questions the simulation user is interested in. Te challenge con-sists in knowing about questions a user in a specific project might have. Generally, a

certain amount o (standard) questions can be pre-defined in correspondence with theapplication area and another set o questions might be defined by the user when start-ing into simulation modelling and experimentation. Tis might even lead to a specificocus in trace file generation and recording o simulation output data by purposeullyintroducing a cohort o observers to the model that directly correspond to the typeand amount o data required or responding to questions already addressed by the user(olujew 1997).

o understand the message o simulation results ormal trace file analysis is oneimportant step. Te other one is the non-ormal, more creative step o directly answering

all questions that are o interest to the user (in our case the logistics expert). Te precon-dition is to know (and understand) what the questions o the user are, but also the abilityo the user to ask questions relevant to a particular problem. For the latter, the rameworkor trace file analysis and interpretation provides even urther support: ypical ques-tions no matter i they are o generic or specific nature help the user in identiying theproblem or the questions to be asked or the aspects to be investigated. As discussed, thiscan be supported by the approaches or viewpoint description and defining observers orspeciying analysis ocus. Additionally, a pattern combining typical symptoms (i.e. visiblesituations or measurable characteristics) with the underlying problems causing thosesymptoms would be o huge benefit as this might also guide the user in truly understand-

ing what happens in a specific material handling or logistics system.

5. DISTRIBUTING KNOWLEDGE: KNOWLEDGEEXCHANGE AND STAKEHOLDER DEVELOPMENT

In the course o a problem solving process there are bidirectional links betweenactivities or problem solving and knowledge management. On one hand knowledgeavailable with persons, inside organizations and in the orm o technology is (re-)usedto solve a particular problem. On the other hand knowledge about the problems final

solution and the chosen mode o action or its generation characterizes the increasedscientific basis and additional experience o the problem-solving person, team or or-ganization. Usually these links are based upon the persons directly involved in theproblem solving process. Its quite common to make use o own experience, but to ben-efit rom knowledge, experience and lessons learned o other parts o the organizationthat is still not the usual procedure yet. o overcome this and to make knowledge o asuccessul or even unsuccessul problem solving process available to uture planningtasks that is the challenge or knowledge management and its integration into person-alized problem solving.

Drawing this picture it becomes pretty clear that logistics problem solving re-quires a wide variety o knowledge, competences, skills and experiences with the prob-lem-solving person. In particular distinctive problem solving competence is requiredwhich can be related to the ability to learn, to be curious and interested in gaining new


16/20

98


experience and acquiring additional knowledge. Te different levels reached in thisprocess might range rom novice to expert, rom amateurs to proessionals. As shownin Figure 8 (applied to solving logistics planning problems) these different levels not

only differ in the degree o independence an individual handles problems with, butalso in the need or getting supported to compensate particular deficits. Furthermore,these competence levels are not o static nature. For one and the same person theymight vary with respect to the problem to be solved itsel or the field it is associatedwith or the way inormation and knowledge is accessible or even the methods andtools to be used in a particular situation. Due to the act that an individuals problem-solving competence depends on his or her knowledge and experience with respect toa particular problem constellation and problem solving procedure, it cannot really beimproved within an abstract scenario ar rom reality. Instead this process is most e-ficient, effective and successul when associated with a real problem-solving challenge.

Here, the development and improvement o abilities and skills can be achieved (i) bythe purposeul use o appropriate tools, (ii) by building well-balanced teams, or (iii) byinitiating and supporting learning processes. In the end problem-solving capabilitiesand with this the level o problem-solving competence can only develop over time andthrough experience, but this process might be speeded up a bit by benefitting romexperts experience instead o gaining all experiences onesel. Tereore, the questionor how to effi ciently and effectively transer lessons learned rom a more experiencedproblem-solver to a less experienced person in the field remains one o the key ques-tions in nowadays economic societies.

Figure 8: Problem-solving competences and needs or compensation o deficits

novice competent

actor

expertactor inaccordancewith situation

advanced

beginner

ability

to act intuitively

to act with greater awarenesson the basis of an intutive

understanding of situationsto act methodically

on the basis of systematisedexperience

to compare things insituation context for deriving

similarities etc.

to implement clear ruleswithout the need for

paying attention on a context

LevelofCapa

bility

Level of Competence

existingplanning competencies

deficits in competenciesto be compensated


17/20

99


Being aware o this, organizations invest a large amount o money in technol-ogy to better leverage inormation, but ofen the deeper knowledge and expertise thatexists within the organization remains untapped. Te sharing o knowledge remains

limited in most respects, and at least, strained. APQC (2004) sees major reasons orthis in technology that is too complicated and the human nature that poses barri-ers to knowledge sharing. Cultural aspects can enhance an open knowledge transeror inhibit a positive attitude towards knowledge sharing. aking cultural aspects intoconsideration requires letting the knowledge management approach and with thisthe knowledge sharing process in particular fit the culture, instead o making anorganizations culture fitting the knowledge management approach (McDermott andODell 2000).

In a perect world the benefits o accessing and contributing knowledge would

be intrinsic: people who share knowledge are better able to achieve their work objec-tives, can do their jobs more quickly and thoroughly, and receive recognition romtheir peers and mentors as key contributors and experts. Nevertheless, knowledgeis ofen not shared. ODell and Grayson (1998) identified our common reasons orthis:

Ignorance. Tose who have knowledge dont realize others may find it useuland at the same time someone who could benefit rom the knowledge maynot know another person in the company already has it.

No absorptive capacity. Many times, an employee lacks the money, time, and

management resources to seek out inormation they need.

Lack of pre-existing relationship. People ofen absorb knowledge rom otherpeople they know, respect, and like. I two managers dont know each other,they are less likely to incorporate each others experiences into their ownwork.

Lack of motivation. People do not see a clear business reason or pursuing thetranser o knowledge.

o meet these challenges, the discipline o knowledge sharing should continu-

ously be reinorced. For this, there are two different approaches: the organization mighthost visible knowledge-sharing events to reward people directly or contributing toknowledge or the organization might rely on the link between knowledge sharing andeveryday work processes by embedding knowledge sharing into routine work proc-esses. Here, initiating o a close, interpersonal link between a mentor or coach (theexpert) and the novice is a promising way not to rely on enthusiasm only, but to bringin a personal commitment to the process o developing another persons problem-solv-ing competence.

Tose expert-novice links might also be part o learning processes to improvean individuals problem-solving competence in a learning-by-doing scenario. Te ped-agogical ramework or this is ormulated by the cognitive apprenticeship theory (seeCollins et al. 1989): in general an apprentice is a learner who is coached by a master toperorm a specific task. Based on this, the theory transers the traditional apprentice-


18/20

100


ship model as known rom crafs, trade and industry to the cognitive domain. Moreprecise, cognitive apprenticeship aims at externalizing processes that are usually carriedout internally. Tis approach works with methods like modelling, coaching, scaffold-

ing, articulation, reflection and exploration. Against this background coaching is to beunderstood as helping a person in actively creating and successully passing individuallearning processes through guidance-on-demand. It is a highly ocused process thatunlocks potential and maximizes perormance at both the individual and organiza-tional levels. It helps people gain clarity, remove sel-imposed limitations and increasetheir sel-reliance, so they can better leverage their strengths and help others to do thesame. Coaching helps individuals to develop critical insight, bringing a new sense opurpose to their actions. It helps them to see where they are, where they want to go to,and how to get there. It stirs them to contribute more. Coaching is a ormal system thatresults in positive, lasting change. In the end, the coach (i.e. the expert) offers support

in case o diffi culties (i.e. scaffolding), provides hints, eedback and recommendations,and eventually takes over certain steps or solving the given problem. However, thecoach only appears when explicitly being called by the person to be coached (i.e. likea help system) and the scaffolding is gradually ading as the learning novice proceeds.So, coaching aims to develop heuristic strategies through establishing a culture o ex-pertise and with this goes ar beyond pure learning as typically provided in workplacelearning environments.

6. CONCLUSIONHuman beings are still and continue to be the key problem solvers, knowledge

holders and knowledge users in logistics. User-riendly tools designed in accordancewith human needs or computer-based support can help to manage, distribute andprovide access to knowledge on one hand and to solve problems on the other hand.Tey take over routine jobs and deal with those problems the problem-solving personhad converted into algorithmic tasks beorehand. In addition to this, intelligent systemscan work with sub-problems i they had respective knowledge and suitably distributedprocesses based on a sophisticated human-computer dialogue. Pre-condition or this isto gain a clear picture about knowledge, abilities and skills o the problem-solving per-son, to make this knowledge accessible and last but not least to understand a personsproblem-solving competence maniested in this.

Te paper has discussed and presented chances and barriers or helping knowl-edge in becoming productive within problem-solving processes. Although many othem have been known or quite some time already, there is still the need or integrat-ing them into a knowledge-riendly environment supporting knowledge acquisition,using, sharing and development o lasting effect. Results rom a survey amongst logis-tics companies clearly gave proo o the increasing interest and understanding withinthe industrial practice.


19/20

101


REFERENCES:

1. APQC (2004). Failures in knowledge sharing. American Productivity and Quality

Center.2. Beckman, .J. (1999). Te Current State o Knowledge Management. In: Liebowitz,

J. (ed.), Knowledge Management Handbook, CRC Press, Boca Raton, I-1-I22.

3. Collins, A.; Brown, J. S.; Newman, S. E. (1989). Cognitive apprenticeship: eachingthe crafs o reading, writing, and mathematics. In: L. B. Resnick (ed.), Knowing,learning, and instruction. Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 453-494.

4. Helms, C.; Strothotte, . (1992). Oracles and Viewpoint Descriptions or ObjectFlow Investigation. Proc. of the 1stEUROSIM Congress on Modelling and Simulation,Capri (Italy), pp. 47-53.

5. McDermott, R.; ODell, C. (2000). Overcoming the Cultural Barriers to SharingKnowledge. APQC (American Productivity and Quality Center).

6. Neumann, G. (2005). Simulation and Logistics. In B. Page and W. Kreutzer, editors,Te Java Simulation Handbook Simulating Discrete Event Systems in UML andJava, Aachen: Shaker, pp. 435-468.

7. Neumann, G. (2006). Projektwissen in der Logistiksimulation erschlieen und be-wahren: Au dem Weg zu einer neuen Dokumentationskultur. In: S. Wenzel (ed.),Simulation in Production and Logistics 2006, pp. 341-350 (Gaining and storing

project knowledge in logistics simulation: on the way towards a new documenta-tion culture, in German).

8. Neumann, G.; om, E. (2006). Knowledge Management and Logistics: Where weare and where we might go to. In: LogForum 2(2006)3, No. 3.

9. Nonaka, I. and akeuchi, H. (1995). Te Knowledge-Creating Company: How Japa-nese Companies Create the Dynamics of Innovation. Oxord University Press.

10. ODell, C.; Grayson, J. (1998). If only we knew what we know. Free Press.

11. Reinmann-Rothmeier, G.; Mandl, H.; Erlach, C.; Neubauer, A. (2001). Wissensman-

agement lernen, Beltz, Weinheim and Basel. (Learning knowledge management; inGerman)

12. Shneiderman, B. (2002). Leonardos Laptop: Human Needs and the New Computingechnologies. Cambridge: Te MI Press.

13. olujew, J. (1997). Werkzeuge des Simulationsexperten von morgen. Proc. of Simu-lation and Animation 97, SCS - Society or Computer Simulation International,San Diego, pp. 201-210. (ools o the simulation expert o tomorrow, in German)


20/20

102


UPRAVLJANJE ZNANJEM NA INDIVIDUALNOJ RAZINIKAO PODRKA RJEAVANJU LOGISTIKIH PROBLEMA3

Gaby Neumann4

Saetak

Rjeavanje problema u logistiki proces je koji uobiajeno zahtjeva intenzivnokoritenje znanja i iskustva osobe koja rjeava problem. Posljedino, pravilno upravljanjeovim znanjem i iskustvom na individualnoj razini predstavlja priliku i izazov za poboljanjekapaciteta rjeavanja problema dotine osobe. Na temelju ove podloge, kroz rad se nastojidoprinijeti produktivnijem koritenju znanja u procesima rjeavanja logistikih problema.

Na temelju izjave menadera logistikog poduzea koji karakteriziraju znanje kao kljuanresurs u svojim procesima koje uglavnom evoluira kroz razne projekte, u radu se identi-ficiraju logistiki problemi kao procesi rjeavanja problema i navode se tipovi problemakoji se javljaju u logistiki. Nadalje, razmotrene su mogunosti koritenja pristupa upravl-janja znanjem u rjeavanju logistikih problema s ciljem odgovaranja na sljedea pitanja:(i) kako ohrabriti i podrati osobe da doprinesu ukupnom znanju ukljuujui i podizan-je svjesnosti o znanju koje ve posjeduju, (ii) kako posredovati izmeu razliitih tipovaznanja koje posjeduju interesne skupine i drugi nositelji i (iii) kako omoguiti i podratipristup prema znanju i koritenje znanja. Ovo je ilustrirano simuliranim problemima izlogistike gdje su simulacijski modeli prikazani kao nositelji znanja, a inteligentan ljudsko-

raunarski dijalog je potreban da bi se pristupilo ovom znanju.Kljune rijei: upravljanje znanjem, rjeavanje problema, logistike simulacije,

nositelji znanja, dijeljenje znanja.

JEL klasifikacija: 030

3 Rad je predstavljen na meunarodnoj konerenciji Managing Services in the Knowledge Economy,(MSKE 2011) koja je odrana od 13. do 15. srpnja 2011. godine u Vila Nova de Famalico u Portugalu.

4 Gaby Neumann, Inenjerska Logistika, ehniko sveuilite aplikativnih znanosti, Wildau, Njemaka, E-

mail: [email protected]

(2011) neumann_knowledge management at individual level to support logistics problem solving

Documents