CHAPTER 9

CHAPTER 9

Knowledge Management

IntroductionWhat do we mean by knowledge? Class Discussion

Drucker (1994): “The knowledge society will be more competitive than anything that we have seen so far.”

Why? With knowledge being universally accessible there will be no reason for por performance.

Cyert (1991): “The most crucial variable in economic development is Knowledge.”

Introduction

Leonard-Barton (1995): “Organizations that are successful innovators are those that build and manage knowledge effectively through activities as developing shared problem-solving skill, experimentation, integrating knowledge across functional boundaries, and importing expertise from external sources.”


Ancient Collaboration at the

organizational level Could revolutionize

collaboration and computing

Opening Vignette: Knowledge Management Gives

Mitre a Sharper Edge

Mitre - knowledge management system (KMS) to leverage organizational knowledge effectively throughout the organization

Internal marketing during development Supported at the highest level Provided an important application

Organizational culture shift was critical Saved $54.91 million / invested $7.19 million


Leverages intellectual assets Delivers appropriate solutions to

anyone, anywhere Good managers have always done

this Ancient concept

DSS Insights- GEM: A DSS for Workload-Planning Decisions

Overview:

. GEM a large stevedoring company

. Schedules developed a week ahead

. Each ship is expected to arrive within 10 days. Unexpected conditions cause the schedule to be re- written


System Description:* means very important variable

Ships

. ETA

. Cargo Information

. Ship’s workload per location

. DWT

. Permitted berths

. Maximum number of elevators

. ETD*


Berth

. Equipment information

. Availability of equipment

. Maximum permitted length

. Maximum permitted draught

. Maximum permitted DWT

DSS Insights- GEM: A DSS for Workload-Planning Decisions Other characteristics

.The planner can override the system

.Each ship has a max number of elevators which can be set by the planner

System operation.Run planning scenario with no penalties.Study results.If there are ships in an unfavorable position (ETD) - manipulate penalties to improve ships position.Repeat until satisfactory

Class discussion!!!!!!


Helps organizations Identify Select Organize Disseminate TransferImportant information and expertise

within the organizational memory in an unstructured manner

Knowledge

As a form of capital, must be exchangeable among persons, and must be able to grow

Intellectual Capital- as the competence of an individual and the commitment of the individual to the organization’s goals

(competence * commitment)


Requires a major transformation in organizational culture to

create a desire to share

Knowledge

Information that is contextual, relevant, and actionable

Knowledge is INFORMATION IN ACTION

Higher than data and information

Knowledge Types

Advantaged knowledge Base knowledge Trivial knowledge Explicit knowledge Tacit knowledge

Knowledge Types

Advantaged Knowledge- Knowledge that provides competitive advantageBase Knowledge- Knowledge that is integral to an organization, providing it with short-term solutions (i.e. best practices)Trivial knowledge- knowledge that has no impact on the organization

Explicit Knowledge

Objective, rational, technical Easily documented Easily transferred / taught /

learned

Tacit Knowledge

Subjective, cognitive, experiential learning

Hard to document Hard to transfer / teach / learn

Involves a lot of human interpretation

DATA

ProcessedINFORMATION

Relevant and actionable

KNOWLEDGE

Relevant and actionable data

Data, Information and Knowledge

Knowledge Has

Extraordinary leverage and increasing returns

Fragmentation, leakage, and the need to refresh

Uncertain value Uncertain value sharing

Organizational Learning and

Organizational Memory

Group memory Learning The learning organization Organizational memory Organizational learning Organizational culture

Organizational Memory

Individual wells Information well Culture well Transformation well Structural well Ecology well

Organizational Learning Focuses

Knowledge source Product-process focus Documentation mode Dissemination mode Learning focus Value chain focus Skill development focus

Organizational Culture

Culture is a pattern of shared basic assumptions

Most important aspect of KM success

Why don’t people share knowledge?

Knowledge Management (KM)

A process of elicitation, transformation, and diffusion of knowledge throughout an enterprise so that it can be shared and thus REUSED

Helps organizations find, select, organize, disseminate, and transfer important information and expertise

Transforms data / information into actionable knowledge to be used effectively anywhere in the organization by anyone

How Core Competency is Linked to Explicit and Tacit Knowledge

TacitKnowledge

Policies, patents,decisions,stra tegies, IS, whitepapers, etc.

Conver t tacit knowledge intoarticulated and measurableexplicit knowledge

Core Competencies of the Organization

Explicit Knowledge

Expertise, know-how, ideas,organization culture, values, etc.

Process of explicationmay generate new tacitknowledge

TacitKnowledge

KM Objectives

Create knowledge repositories Improve knowledge access Enhance the knowledge

environment Manage knowledge as an asset

KMS Manage

Knowledge creation through learning Knowledge capture and explication Knowledge sharing and

communication through collaboration Knowledge access Knowledge use and reuse Knowledge archiving

Knowledge Repository

Not a database Not a knowledge base (like for

ES)

A collection of internal and external knowledge

Knowledge Repository Types

External Structured internal knowledge Informal internal knowledge

KM Activities

Externalization Internalization Intermediation Cognition

KM Features

Create a knowledge culture Capture knowledge Generate knowledge Explicate (and digitize)

knowledge Share and reuse knowledge Renew knowledge

Cyclic Model of KM

Create knowledge Capture knowledge Refine knowledge Store knowledge Manage knowledge Disseminate knowledge

Cyclic Model of KM

ManageKnowledge

StoreKnowledge

DisseminateKnowledge

RefineKnowledge

Create Knowledge

CaptureKnowledge

KM Examples Mitre Dow Chemical Company Xerox Chrysler Monsanto Chevron Buckman Laboratories KPMG Ernst & Young Arthur Andersen Andersen Consulting

Why Adopt KM

Cost savings Better performance Demonstrated success Share Best Practices Competitive advantage

Chief Knowledge Officer (CKO)

Maximize firm’s knowledge assets

Design and implement KM strategies

Effectively exchange knowledge assets

Promote system use

KM Development

Need a knowledge strategy Identify knowledge assets

KM Development

1. Identify the problem2. Prepare for change3. Create the team4. Map out the knowledge5. Create a feedback mechanism6. Define the building blocks7. Integrate existing information

systems

Strategies for Successful KM

Implementation

1. Establish a KM methodology2. Designate a pointperson3. Empower knowledge workers4. Manage customer-centric

knowledge5. Manage core competencies

More Strategies

6. Foster collaboration and innovation

7. Learn from best practices8. Extend knowledge sourcing9. Interconnect communities of

expertise (communities of practice)

10.Report the measured value of knowledge assets

KM Methods, Technologies, and

Tools Email or messaging Document management Search engines Enterprise information portal Data warehouse Groupware Workflow management Web-based training Others

How to KM

Integrate the technologies to manage knowledge effectively

KM Tool Categories

Information architecture Technical architecture Application architecture

KM Software Knowware still developing but…

DecisionSuite Wincite DataWare KnowledgeX Knowledge Share

KM Success

Economic performance Technical and organizational infrastructure Standard, flexible knowledge structure Knowledge-friendly culture Clear purpose and language Change in motivational practices Multiple channels for knowledge transfer Worthwhile level of process orientation Nontrivial motivational encouragement Senior management support

Measuring Success

Balanced Scorecard Skandia Navigator Economic Value Added Inclusive Valuation

Methodology

KM Failure Causes

1. Unclear definition of knowledge2. Overemphasis on knowledge

stock, not flow3. Belief that knowledge exists

outside people’s heads4. Not recognizing the importance

of managing knowledge5. Failure to manage tacit

knowledge

More Failure Causes

6. Failure to disentangle knowledge from its uses

7. Downplaying reason and thinking8. Focusing on the past and present, not

the future9. Failure to recognize the importance of

experimentation10.Substituting technology contact for

human interface11.Overemphasis on measuring

knowledge, not its outcomes

KM and AI

Can use AI in KM Can use KM in AI Data mining can create

knowledge

Electronic Document Management

A KM for documents Everyone is on the same page Documents are up to date Simple example: corporate

phonebook

Lower costs Better performance

The Knowledge-Based View of Decision Making

Accepting Messages: see next slide

A decision maker (human-being) can accept stimuli from the environment

The stimuli are messages that carry knowledge (information)

Some messages have a direct and immediate impact on the decisions being manufactured

Other messages can be:. discarded. passed along to others and or other places. stored for future use


Issuing Messages

The decision maker can issue messages to the surroundings:

. other people

. documents/storage vessels

The message may also be the Manufactured Decision


Assimilating Knowledge Figure 4.2, page 99.

Once the Decision maker has established the meaning of an incoming message it can be assimilated with the DM’s knowledge store

When new knowledge is assimilated, it alters the knowledge store:. just be added. cause existing knowledge to be altered, discarded, or marked as being obsolete. It may cause fundamental alterations of

the knowledge store


Recognizing the Need for a Decision

May be very obvious: . Highly structured . Happens frequently

May be take many repetitions of the event/stimuli to initiate action, thus it is:. unstructured. novel. by observing conditions (economic, political, mechanical) we may come to recognize that: - a problem exits - a solution is required


Manufacturing a Decision: The manufacturing process produces new knowledge from knowledge.

The sources of raw materials is the decision maker’s storehouse of knowledge (experience, facts, rules, etc).

Knowledge is extracted on an as needed basis and manipulated by Cognitive abilities to produce solutions for the flow of problems that constitutes the KNOWLEDGE Manufacturing Process.

The solution that is the product of the process is the NEW KNOWLEDGE.

1.

A Manufacturing AnalogyMaterial Product Decision1. The process begins in reaction to a customer order or anticipated order.

The process begins in reaction to a recognizedneed or opportunity.

2. The process draws on an inventory of raw materials.

The process draws on an inventory ofKNOWLEDGE.

3. Items entering inventory are subject to quality testing controls.

Knowledge is assimilated into inventory only ifit is expected to be usable.

4. Abilities for manipulating materials transform/assemble raw materials into final products.

Abilities for manipulating knowledgetransformation/assemble existing knowledgeinto new knowledge about what to do.

5. During the process may yield material by- products that are stored in inventory or discarded.

During the process there are intermediatepieces of knowledge called problem solutions.

6. The process may yield material by-products that are stored in inventory or are discarded.

The process may yield knowledge by-productsthat are stored in inventory or are discarded.

7. The manufacture may be an individual or have multiple participants.

The decision maker may be an individual or consist of multiple participants

8. The finished product is packaged fordistribution.

The decision is packaged for distribution.

From Holsapple & Whinston, 1996, page 100.

Defining KnowledgeThree Views of Knowledge

Knowledge Representations:If a system has and can use a representation of somethingthen the system itself can also be said to have KNOWLEDGE.

The textbook can be a representation of knowledge if it can be read.

Representation is pattern of Symbols => an abstraction

It embodies knowledge:

Defining Knowledge

A useable representation of something

From a DSS point of view we must be concerned with

the computer memory and how it

processes knowledge represents knowledge

Clean data and defined objects are required

Knowledge States

A set of states ranging from raw data to decisions

Six states of Knowledge:

data

Information

structured information

insight

judgment

decision: The highest state

Defining Knowledge

One state of knowledge can be used to generate different states of knowledge

DSS help in: Acquiring knowledge

deriving knowledge

Knowledge Production

The result of a productivity activity (i.e. LEARNING) involving acquisition and/or derivation The flows and stock of knowledge

Figure 4.3, page 106

Stocks are the inventories of knowledge

The flows are the messages that tell the stock to do something

Knowledge Sources

The decision makers store house of Knowledge: Internal & External

The DM can be active or Passive about acquiring knowledge

Active: Message can be emitted to invoke a response

Passive the DM observes without invoking reactions

Knowledge Sources

The Decision to Acquire/ Derive Knowledge

General a mixture of acquiring and derivation of knowledge

Acquiring knowledge may tax:cognitive abilities, time, economic limits

There are tradeoffs

DSS tens to promote greater reliance on internal production of knowledge

Knowledge Sources

Reliability of Knowledge: Do we get the same knowledge from internal and external sources?

If there are multiple external sources- to they yield the same result

DSS

Without a DSS it may be infeasible to produce the it internally

Use the DSS in parallel with the knowledge acquisition to check the source reliability

Knowledge Sources

Knowledge Qualities DSS

accurately retaining knowledge

flagging inconsistencies

analyzing uncertainties

tracking multiple sets of knowledge

Knowledge Sources

Utility of Knowledge: Usefulness

Knowledge can be useful to different people

To a history professor knowledge about particle physics is probably not useful

Figure 4.4

DSS: Present what is relevant to a specific DM

Provide high quality knowledge


Techniques Text ManagementForms ManagementBusiness GraphicsSolver ManagementRule ManagementDatabase ManagementReport GenerationSpreadsheet AnalysisProgrammingMessage Management

Knowledge ManagementReasons for Understanding Knowledge

ManagementPosition or integrate Knowledge into a decisionExtend the role of supporting participants “from mere production to the processing, storage, retrieval, dissemination, utilization and general management of knowledge.”Facilitate and develop a philosophy and methodologies for handling knowledgeShift the role of supporting participants “from producing certainty and complete knowledge to structuring ignorance and managing uncertainty

Lohuizen & Kochen, 1986 in Holsapple & Whinston, 1996, page 112


Five types of Knowledge

1. Practical2. Intellectual3. Pastime4. Spiritual5. Unwanted


Three Primary Types of Knowledge

1. Descriptive: Includes descriptions of past, present, future, and hypothetical situations. DATA and INFORMATION- To Know What

2. Procedural Knowledge: The how to do; Step-by-Step

3. Reasoning Knowledge: To know Why- Approaches to problem solving


Three Secondary Types of Knowledge

1. Linguistic: Vocabulary and grammar, body language, meaning of gestures.

2. Assimilation Knowledge: The basis for controlling changes to the knowledge store.

A filtering mechanism

3. Presentation Knowledge: The basis for packaging outgoing responses


The Decision Maker possess knowledge

The DSS has processing abilities that can supplement the Decision Maker

The Cognitive Basis for Knowledge

Declarative Knowledge factual information that is static in

nature it is usually describable to us history- events, facts flexible- it can be reorganized to suite

our purposes Knowing That

Cognitive- Knowledge

Some knowledge can be encoded in a declarative format which can later be transformed into a procedural format as we become familiar with the information.Examples: Reading Windows for Dummies Reading a Golf technique book then truing on the PC/ Play golf

Cognitive Knowledge

Attention: the concentration and focusing of mental activityPaying attention seems to accentuate, or enhance, sensory input that has been focused on

The Cognitive Basis for Knowledge

Procedural Knowledge the underlying skillful actions we possess it is dynamic it is not very describable the acquisition of a skill involves making

and detecting errors (skiing, bike riding, ballet)

with additional experience we improve Knowing How

Information ProcessingSensory system: where specific aspects of the environment are detected and organized-into cognitive codeThe code is passed into memoryMemory Working memory: a workbench for

cognitive codes (short term memory) Permanent memory: long term storage of

declarative and procedural knowledge

Memory

External InputSensory RegisterShort-term StorageLong-term Storage

Sensory Register

Where our feature detection and pattern recognition process produce a cognitive code that can be stored for a short time.The Sensory register does not depend on resource allocation- we do not have to pay attention to incoming stimuli in order to have this cognitive code created.

Sensory Register

It must have a large storage capacityIt is modality specific: has difference storages for audio, visual

The code in storage Decays over timeResources must be allocated to transfer the code to STM or LTM

Short-term Memory

Limited capacity (RAM)Storage is organized by sensory component: acoustic, verbal, linguistic

Storage duration of unrehearsed material is about 30 secondsMaterial that is not elaborated or transferred decays.

Long-term Memory

To go from STM to LTM requires rehearsalRehearsal: procedures that maintain the vitality of the

code STS code will last indefinitely if it is

occasionally refreshed by rehearsal. Rehearsal duplicates and augments the

code for long-term storage (associations/links are created),

Technology Infrastructure

Organizational Infrastructure

The Architecture of Knowledge Repositories --- "Pipeline"

Repository* Content

* Structure

Knowledge Views


* Content* Packaging/Format

* Accessing/Distribution

Aquisition RefinementStorage

RetrievalDistributed Presented

SUPPLIERS

USERS

Process Platform

Discussion Participants

The Architecture of Interactive Knowledge Repositories --- "Virtual"

Repository* Content

* Structure

Knowledge Views


* Content* Packaging/Format

* Accessing/Distribution

Aquisition RefinementStorage

RetrievalDistributed Presented

Knowledge Engineering and Acquisition

Taxonomy of Knowledge Types

Primary Descriptive: data, information, descriptions of past,

present, and future situations Procedural: how to do something Reasoning: codes of conduct, regulations, policies,

diagnostic rules

Secondary Linguistic: vocabulary, grammar, knowledge of

gestures Assimilative: permissible contents, retention cycles,

relevancy filters Presentation: modes of communication, graphing,

messaging, inverse of Linguistic knowledge

Conceptual Model of Knowledge Engineering Process

Validation and Verification Feedback of Performance

and Knowledge

Knowledge ElicitationTools

Knowledge Structuring

Tools

Knowledge ModelingTools

PERSONExpert performance of

some task in some domain

COMPUTEREmulation of expert

performance of some task in some domain

Psychology of person

Personal construct psychology of person as an anticipatory system

Ontology of computer

knowledge representation and operationalizing an

anticipatory system

Psychological model of skilled

performanceRepresentation of skill in

terms of conceptual structures

Computational model of skilled

performanceRepresentation of skill in

terms of logicalstructures

Required Expertise Transfer

Elicitation

Feedback

Unification of psychological and

computational representation

Knowledge Acquisition Dimensions

StrategicStrategicKE-driven

Expert-drivenMachine-driven

InterviewsProtocol analysis

Repertory Grid

TacticalTactical

Five-Stage General Process of Knowledge Acquisition

Identification

Identify Problem Characteristics Conceptualization

Find ConceptsTo

RepresentKnowledge

Formalization

Design Structure to Organize Knowledge

Implementation

Formulate Rules, Frames, etc., to

Embody Knowledge

Testing

Validate Rules that Organize Knowledge

Refinements

Redesigns

Reformulations

Requirements

Concepts

Structure

Rules

Basic Pre-Interview Checklist

Decide what you need to know. Ask yourself why this information is

needed. Determine that an interview is the best

method for obtaining this information. Determine the appropriate degree of

structure for the interview. Consider the method in which the

answers to your questions will be coded and analyzed.

Necessary Task Conditions for Successful Concurrent Protocols

The sample of cases employed must be highly representative of the task under study.

Each task must have a clearly defined conclusion or point of completion.

The task must be able to be completed in one protocol session.

All data must be presented to the expert in a familiar form.

A test case should be given to the expert prior to the collection of protocols so that he or she may become familiar and comfortable with the verbalization process.

Typical Repertory Grid Structure

Constructs Element 1 Element 2 Element 3Distinction 1Constraint 1,1Constraint 1,2Constraint 1,3Distinction 2Constraint 2,1Constraint 2,2Constraint 2,3Distinction 3Constraint 3,1Constraint 3,2Constraint 3,3Distinction 4Constraint 4,1Constraint 4,2Constraint 4,3Distinction 5Constraint 5,1Constraint 5,2Constraint 5,3

Individual Individual Individual Individual Individual Individual Individual

Concept

Concept

Concept

Concept

ConceptConcept

ConceptConcept

Concept

Concept Concept

Concept Concept Concept

Concept

Knowledge Base Validation Measures and Techniques

Accuracy: How well does the system reflect reality. How correct is the knowledge in the knowledge base.

Adaptability: Possibilities for future development or changes. Adequacy: The portion of the necessary knowledge that is included in the knowledge base. Appeal: How well the knowledge base matches intuition and stimulates thought and

practicability Breadth: How well is the domain covered. Depth: The degree of the detailed knowledge. Face Validity: How credible is the knowledge. Generality: Capability of a knowledge base to be used with a broad range of similar

problems. Precision: Capability of the system to replicate particular system parameters. Consistency

of advice and coverage of variables in the knowledge base. Realism: Accounting for the relevant variables and relations. Similarity to reality. Reliability: The frequency of system predictions that are correct. Robustness: Sensitivity of conclusions to model structure. Sensitivity: The impact of changes in the knowledge base on the quality of outputs. Technical/Operational: Goodness of the assumptions, context, constraints, and conditions. Turing test: Ability of a human evaluator to identify if a given conclusion is made by a real

expert or a computer. Usefulness: How adequate the knowledge is (in terms of parameters and relationships) for

solving correctly. Validity: The capability of the knowledge base for producing empirically correct predictions.

KM – The Future

Not a fad Impact is immense Research on organizational

culture How to do each step Are they the right steps?


The definition is clear The concepts are clear The challenges are

Clear Surmountable

The benefits are clear (and can be huge)

The tools and technologies are viable

Knowledge ManagementKey Issues

Organizational culture Executive sponsorship Measuring success

The future: Comprehensive standardized KM packages

Knowledge Mangement“The wise see knowledge and action

as one” (Bhagvad Gita)Intelligent organizations recognize

that knowledge is an asset, perhaps the only one that grows over time, and when harnessed effectively can sustain the ability to continuously compete and innovate.

CHAPTER 9

Documents

organizational knowledge

actionable knowledge

knowledge society

ships workload

re written dss insights

equipment information

maximum permitted length

maximum permitted draught