R. Ching, Ph.D. MIS California State University, Sacramento 1 Week 7 Monday, October 10 Knowledge ManagementKnowledge Management Managing OperationsManaging.

R. Ching, Ph.D. • MIS • California State University, Sacramento 1

Week 7Week 7Monday, October 10Monday, October 10

• Knowledge ManagementKnowledge Management• Managing OperationsManaging Operations


Knowledge and Expertise as a Knowledge and Expertise as a ResourceResource

Capturing Knowledge through ITCapturing Knowledge through IT


Knowledge as a ResourceKnowledge as a Resource

• Information that is contextual, relevant and actionableInformation that is contextual, relevant and actionable

• Used for problem solving Used for problem solving

– Applied in a context Applied in a context

– Relevant to the taskRelevant to the task

– Used to support an actionUsed to support an action

• Fuzzy and loosely coupledFuzzy and loosely coupled

– Must be understood within a contextMust be understood within a context

– Foundational resourceFoundational resource

– Implied understanding of its useImplied understanding of its use

Meaningful and valuable,Meaningful and valuable,but ephemeral (i.e., transitory)but ephemeral (i.e., transitory)


Types of KnowledgeTypes of Knowledge

• Descriptive – knowing whatDescriptive – knowing what

• Procedural – knowing howProcedural – knowing how

• Reasoning – knowing whyReasoning – knowing why

• Presentation – knowing how to Presentation – knowing how to communicate or deliver knowledgecommunicate or deliver knowledge

– Linguistic – knowing how to Linguistic – knowing how to interpret communicationinterpret communication

• Assimilative – knowing how to Assimilative – knowing how to maintain knowledge by improving maintain knowledge by improving existing knowledgeexisting knowledge

Basic knowledgeBasic knowledge

Communicating, Communicating, understanding and understanding and learninglearning


Characteristics of KnowledgeCharacteristics of Knowledge

• Ground truth grained from experience, not theoryGround truth grained from experience, not theory

• Complexity as applied to problem solvingComplexity as applied to problem solving

– Simplification of problem spaceSimplification of problem space

• Judgment puts knowledge in actionable formJudgment puts knowledge in actionable form

• Heuristic and intuitive approaches to problem solvingHeuristic and intuitive approaches to problem solving

• Value and beliefs applied to defining the problem spaceValue and beliefs applied to defining the problem space


Decision MakingDecision Making

• Modeling the problemModeling the problem

Inputs Inputs Process Process Output Output

– Represents the real thingRepresents the real thing

– Identifies the components and their interactionsIdentifies the components and their interactions

– Specifies the behaviorSpecifies the behavior

– Predicts the outcome based on an established process given Predicts the outcome based on an established process given a set of inputsa set of inputs


What is an Expert System?What is an Expert System?

• ““An An expert systemexpert system is an intelligent computer is an intelligent computer programprogram that uses that uses knowledge and inference procedures to solve knowledge and inference procedures to solve problems that problems that are difficult enough to require significant human expertise for are difficult enough to require significant human expertise for their solutiontheir solution. The knowledge to perform at such a level plus . The knowledge to perform at such a level plus the inference procedures used, can be thought of as a model of the inference procedures used, can be thought of as a model of the expertise of the best practitioners of the field.”the expertise of the best practitioners of the field.”

Feigenbaum (1983)Feigenbaum (1983)


Benefits of Expert Systems Benefits of Expert Systems ((Turban)Turban)

• Increased output and productivityIncreased output and productivity

• Increased qualityIncreased quality

• Reduced down timeReduced down time

• Capturing scarce expertiseCapturing scarce expertise

• FlexibilityFlexibility

• Equipment operationEquipment operation

• Using less expensive equipmentUsing less expensive equipment

• Operation in hazardous environmentsOperation in hazardous environments

• ReliabilityReliability


Benefits of Expert SystemsBenefits of Expert Systems

Cont.Cont.

• Response timeResponse time

• Integration of several experts' opinionsIntegration of several experts' opinions

• Working with incomplete and uncertain informationWorking with incomplete and uncertain information

Turban (1990)Turban (1990)


Gate Assignments Expert SystemGate Assignments Expert System

Small Small aircraftaircraft

Gates are a resource of an airline.Gates are a resource of an airline.

High capacity High capacity flightsflights


Boeing 747: 347 passengers

Boeing 767: 193-244 passengers


Boeing 757: 181 passengers

Airbus A319/320: 120-138 passengers



Chicago O’Hare (ORD)Chicago O’Hare (ORD)


Los Angeles (LAX)Los Angeles (LAX)

Large aircraft (B747, B777)Large aircraft (B747, B777)

Small aircraft Small aircraft (B737, A319)(B737, A319)

Med-size aircraft (B757, B767)Med-size aircraft (B757, B767)

Med-size aircraft Med-size aircraft (B757, A320)(B757, A320)


ConstraintsConstraints

• Certain aircraft can fit only certain gatesCertain aircraft can fit only certain gates

• Minimize traffic in an areaMinimize traffic in an area

• Minimize distance between aircraft for connecting flightsMinimize distance between aircraft for connecting flights

• Delays due to weather, mechanical problems, aircraft Delays due to weather, mechanical problems, aircraft servicing, arrivals and others have an impact on the servicing, arrivals and others have an impact on the assignmentsassignments

• International flights cannot park at a domestic gateInternational flights cannot park at a domestic gate


When to Use an Expert SystemWhen to Use an Expert System

• High potential payoff or significantly reduced downside riskHigh potential payoff or significantly reduced downside risk

• Ability to capture and preserve irreplaceable human expertiseAbility to capture and preserve irreplaceable human expertise

• Ability to develop a system more consistent than human Ability to develop a system more consistent than human expertsexperts

• Expertise is needed at a number of locations at the same timeExpertise is needed at a number of locations at the same time

• Expertise is needed in a hostile environment that is dangerous Expertise is needed in a hostile environment that is dangerous to human healthto human health


When to Use an Expert SystemWhen to Use an Expert System

Cont.Cont.

• The expert system solution can be developed faster than the The expert system solution can be developed faster than the solution from human expertssolution from human experts

• An expert system is needed for training and development so as An expert system is needed for training and development so as to share the wisdom and experience of human experts with a to share the wisdom and experience of human experts with a large number of peoplelarge number of people

Turban, 1995Turban, 1995


Problems in Knowledge AcquisitionProblems in Knowledge Acquisition

• Expressing the knowledgeExpressing the knowledge

• Transfer to machineTransfer to machine

• Number of participantsNumber of participants– ExpertExpert

– Knowledge engineerKnowledge engineer

– System designerSystem designer

– UserUser

• Structuring knowledgeStructuring knowledge– Representing knowledgeRepresenting knowledge



Cont.Cont.

• OthersOthers– Experts may lack time or may be unwilling to cooperateExperts may lack time or may be unwilling to cooperate

– Testing and refining knowledge is complicatedTesting and refining knowledge is complicated

– Methods for knowledge elicitation may be poorly definedMethods for knowledge elicitation may be poorly defined

– System builders have a tendency to collect knowledge from one System builders have a tendency to collect knowledge from one source, but the relevant knowledge may be scattered across source, but the relevant knowledge may be scattered across several sourcesseveral sources

– Builders may attempt to collect documented knowledge rather Builders may attempt to collect documented knowledge rather than using expertsthan using experts



Cont.Cont.

• OthersOthers– It is difficult to recognize specific knowledge when it is mixed It is difficult to recognize specific knowledge when it is mixed

up with irrelevant dataup with irrelevant data

– Experts may change their behavior when they are being observed Experts may change their behavior when they are being observed and/or interviewedand/or interviewed

– Problematic interpersonal communication factors may exist Problematic interpersonal communication factors may exist between the KE and the expertbetween the KE and the expert



Knowledge Engineering BottleneckKnowledge Engineering Bottleneck

• Eliciting expertise from an expert:Eliciting expertise from an expert:

– CostlyCostly

– Error ProneError Prone

– Time ConsumingTime Consuming

• "Gaps" in the expert's knowledge"Gaps" in the expert's knowledge

• Inability to articulate knowledgeInability to articulate knowledge– "The ability to function at an expert level in a task domain "The ability to function at an expert level in a task domain

does not necessarily confer a corresponding ability to does not necessarily confer a corresponding ability to articulate this know-how."articulate this know-how."

Quinlan (1987)Quinlan (1987)


Rule InductionRule Induction

• "...process of reasoning from the specific to the general. "...process of reasoning from the specific to the general. In ES terminology it refers to the process in which rules are In ES terminology it refers to the process in which rules are generated by a computer program from example cases."generated by a computer program from example cases."



IT and Rule InductionIT and Rule Induction

Case Case Classified Classified Through Through

DeductionDeduction

Rules Induced Rules Induced From Example From Example

CasesCases

Individual Cases Individual Cases Applied to the Applied to the

RulesRules

InductionInduction(Inductive Logic)(Inductive Logic)

DeductionDeduction(Deductive Logic)(Deductive Logic)


Advantages of InductionAdvantages of Induction

• Discovers rules from examplesDiscovers rules from examples

• Avoids knowledge elicitation problemsAvoids knowledge elicitation problems

• Can produce new knowledgeCan produce new knowledge

• Can uncover critical decision factorsCan uncover critical decision factors

• Can eliminate irrelevant decision factorsCan eliminate irrelevant decision factors

Durkin, 1991Durkin, 1991


Pay or Reject Pay or Reject Type of AccountType of AccountCredit Credit RatingRating

Overdraft for Overdraft for Single or Multiple Single or Multiple

Checks Checks PayPay RegularRegular GoodGood MultipleMultiplePayPay StudentStudent UnknownUnknown SingleSingle

RejectReject StudentStudent PoorPoor SingleSingleRejectReject StudentStudent GoodGood MultipleMultiplePayPay StudentStudent GoodGood SingleSingle

DecisionDecision Decision AttributesDecision Attributes

Check Overdraft CasesCheck Overdraft Cases

PayPay RegularRegular UnknownUnknown MultipleMultiplePayPay RegularRegular GoodGood SingleSingle

RejectReject RegularRegular PoorPoor SingleSingleRejectReject StudentStudent UnknownUnknown MultipleMultipleRejectReject RegularRegular UnknownUnknown MultipleMultiple


Pay or Reject Pay or Reject Type of AccountType of AccountCredit Credit RatingRating

Overdraft for Overdraft for Single or Multiple Single or Multiple

Checks Checks

??

DecisionDecision Decision AttributesDecision Attributes

Pay or Reject?Pay or Reject?

RegularRegular UnknownUnknown SingleSingle


Decision TreesDecision Trees

• A decision tree is a representation of a decision procedure for A decision tree is a representation of a decision procedure for determining the class of a given instance. Each node of the determining the class of a given instance. Each node of the tree specifies either a class name or a specific test that tree specifies either a class name or a specific test that partitions the space of instances at the node according to the partitions the space of instances at the node according to the possible outcomes of the test. Each subset of the partition possible outcomes of the test. Each subset of the partition corresponds to a classification subproblem for that subspace of corresponds to a classification subproblem for that subspace of the instances which is solved by a tree.the instances which is solved by a tree.

Utgoff, 1989Utgoff, 1989


Bank Overdraft ApplicationBank Overdraft Application

• 340 Cases of check overdrafts340 Cases of check overdrafts• Classification Variable:Classification Variable:

– Check not paid (0) vs. paid (1)Check not paid (0) vs. paid (1)• Seven Decision Attributes:Seven Decision Attributes:

– Difference between amount of check and amount in checking Difference between amount of check and amount in checking account (DIFF)account (DIFF)

– Type of account (student, regular) (ACT)Type of account (student, regular) (ACT)– Credit rating (good, unknown, poor) (CR)Credit rating (good, unknown, poor) (CR)– Overdraft of single or multiple checks (COV) Overdraft of single or multiple checks (COV) – Interaction between DIFF and ACTInteraction between DIFF and ACT– Interaction between DIFF and CRInteraction between DIFF and CR– Interaction between DIFF and COVInteraction between DIFF and COV

Dummy coded Dummy coded (CR(CR11, CR, CR22): ):

00 (unknown), 00 (unknown), 10 (good),10 (good),01 (poor)01 (poor)


ID3 Decision TreeID3 Decision Tree

176130

60125

5957

501

21

01

20

480

956

556

40

354 2

2 01

168

053

115

10

015

154

142

12

1011

690

321

320

01

Pay Reject Pay

Reject Pay

Reject Reject Reject Pay Reject Pay

Pay Reject Pay Reject

DIFF<20.5

DIFF<10.5

DIFF<9.4DIFF<40.3

DIFF<42.2

CR *DIFF<6.5

CR *DIFF<.035

DIFF<1.65

ACT*DIFF<.175

CR*DIFF<5.5

COV*DIFF<1.5

DIFF<5.55

ACT*DIFF<3

Yes No

Yes No Yes No

ACT*DIFF<19.6

Overall Classification Rate: 97.7%Overall Classification Rate: 97.7%1

2

1165


Estimated Estimated error rateerror rate

176

130

116

5

60

125

59

57

50

1

2

1

0

1

2

0

48

0

9

56

5

56

4

0

3

542

2

2

1

0

1

1

68

0

531

15

1

0

0

15

15

4

14

2

1

2

101

1

69

032

1

32

0

0

1

Pay Reject Pay

Reject Pay



DIFF<20.5

DIFF<10.5

DIFF<9.4DIFF<40.3

DIFF<42.2

CR *DIFF<6.5

CR *DIFF<.035

DIFF<1.65

ACT*DIFF<.175

CR*DIFF<5.5

COV*DIFF<1.5

DIFF<5.55

ACT*DIFF<3

Yes No

Yes No Yes No

ACT*DIFF<19.6

176

130

116

5

60

125

59

57

50

1

2

1

0

1

2

0

48

0

9

56

5

56

4

0

3

542

2

2

1

0

1

1

68

0

531

15

1

0

0

15

15

4

14

2

1

2

101

1

69

032

1

32

0

0

1

Pay Reject Pay

Reject Pay



DIFF<20.5

DIFF<10.5

DIFF<9.4DIFF<40.3

DIFF<42.2

CR *DIFF<6.5

CR *DIFF<.035

DIFF<1.65

ACT*DIFF<.175

CR*DIFF<5.5

COV*DIFF<1.5

DIFF<5.55

ACT*DIFF<3

Yes No

Yes No Yes No

ACT*DIFF<19.6

Pruning: Finding the Optimal TreePruning: Finding the Optimal Tree

956

5957

60125

176130

40

556

501

168

1165

Reject

CR *DIF

Paid

CR *DIF

Reject

COV*DIF

Paid Reject

Yes No

Yes No

Yes No

Yes No

2

2

CR *DIF 1

176130

CR *DIF 2




Pruning Using the Estimated True Pruning Using the Estimated True Error RateError Rate

Number of nodesNumber of nodes

Error rateError rate Minimum Minimum error rate error rate

Optimal Optimal number of number of

nodesnodes

IncreaseIncrease

Incr

ease

Incr

ease

22 33 44 55 66

••••••

••

••


RPA TreeRPA Tree

956

5957

60125

176130

40

556

501

168

1165

Reject

CR *DIF

Paid

CR *DIF

Reject

COV*DIF

Paid Reject

Yes No

Yes No

Yes No

Yes No

2

2

CR *DIF 1

Overall Correct Classification Rate: 96.1%Overall Correct Classification Rate: 96.1%


RPA TreeRPA Tree

956

5957

60125

176130

40

556

501

168

1165

Reject

CR *DIF

Paid

CR *DIF

Reject

COV*DIF

Paid Reject

Yes No

Yes No

Yes No

Yes No

2

2

CR *DIF 1

Overall Correct Classification Rate: 96.1%Overall Correct Classification Rate: 96.1%

For example:For example:If credit rating is ‘poor’ and If credit rating is ‘poor’ and difference > 20difference > 20 then action = ‘reject’then action = ‘reject’ else examine furtherelse examine further = ‘yes’= ‘yes’

If credit rating is ‘good’ andIf credit rating is ‘good’ and difference > 1difference > 1 then action = ‘pay’then action = ‘pay’ else examine further else examine further = ‘yes’= ‘yes’

1 x 201 x 20

< 20< 20 2020

1 x 11 x 1

CRCR11 CR CR22

UnknownUnknown 00 00GoodGood 11 00PoorPoor 00 11

0 x any value0 x any value

0 x any value0 x any value


Another Example…Another Example…


Data Mining: Knowledge Discovery and Data Mining: Knowledge Discovery and ClusteringClustering

Credit card useCredit card use


Knowledge Discovery and ClusteringKnowledge Discovery and Clustering


• Males (75%)Males (75%)• Age 41-45 (37%), 46-50 (35%)Age 41-45 (37%), 46-50 (35%)• College educated (90%), College educated (90%), • Annual income NT$1-2 million Annual income NT$1-2 million

(71%) (US$29,000-58,000)(71%) (US$29,000-58,000)• Primary purchases: Primary purchases:

Books/magazines (65%), Food Books/magazines (65%), Food (37%), Plane tickets (34%)(37%), Plane tickets (34%)

0000 0000 0000 00000000 0000 0000 0000

PlatinumPlatinum0000 0000 0000 00000000 0000 0000 0000

PlatinumPlatinum

55%55% 0000 0000 0000 00000000 0000 0000 0000

GoldGold0000 0000 0000 00000000 0000 0000 0000

GoldGold

0000 0000 0000 00000000 0000 0000 0000

RegularRegular0000 0000 0000 00000000 0000 0000 0000

RegularRegular30%30%

15%15%




• Males (87%)Males (87%)• Age 51-55 (36%), 56-60 (30%)Age 51-55 (36%), 56-60 (30%)• College educated (66%)College educated (66%)• Annual income NT$4-5 million Annual income NT$4-5 million

(54%) (US$116,000-145,000)(54%) (US$116,000-145,000)• Primary purchases: Food Primary purchases: Food

(63%), Clothing (33%)(63%), Clothing (33%)

0000 0000 0000 00000000 0000 0000 0000


PlatinumPlatinum

10%10% 0000 0000 0000 00000000 0000 0000 0000

GoldGold0000 0000 0000 00000000 0000 0000 0000

GoldGold

0000 0000 0000 00000000 0000 0000 0000

RegularRegular0000 0000 0000 00000000 0000 0000 0000


35%35%




• Males (54%), Males (54%), • Age under 25 (32%), 26-30 Age under 25 (32%), 26-30

(30%)(30%)• College educated (34%), College educated (34%), • Annual income under Annual income under

NT$500,000 (60%) NT$500,000 (60%) (US$14,500)(US$14,500)

• Primary purchases: Primary purchases: Entertainment (54%), Clothing Entertainment (54%), Clothing (19%), Others (19%)(19%), Others (19%)

0000 0000 0000 00000000 0000 0000 0000


PlatinumPlatinum

83%83% 0000 0000 0000 00000000 0000 0000 0000

GoldGold0000 0000 0000 00000000 0000 0000 0000

GoldGold

0000 0000 0000 00000000 0000 0000 0000

RegularRegular0000 0000 0000 00000000 0000 0000 0000


3%3%




• Males (58%)Males (58%)• Age 31-35 (44%), 26-30 (31%)Age 31-35 (44%), 26-30 (31%)• College educated (46%)College educated (46%)• Annual income under Annual income under

NT$500,000-1 million (69%) NT$500,000-1 million (69%) (US$14,500-29,000)(US$14,500-29,000)

• Primary purchases: Clothing Primary purchases: Clothing (77%), Food (64%), (77%), Food (64%), Entertainment (33%)Entertainment (33%)

0000 0000 0000 00000000 0000 0000 0000


PlatinumPlatinum

73%73% 0000 0000 0000 00000000 0000 0000 0000

GoldGold0000 0000 0000 00000000 0000 0000 0000

GoldGold

0000 0000 0000 00000000 0000 0000 0000

RegularRegular0000 0000 0000 00000000 0000 0000 0000

RegularRegular

18%18%9%9%




• Females (78%)Females (78%)• Age 36-40 (66%), 41-45 (24%)Age 36-40 (66%), 41-45 (24%)• College educated (57%), College educated (57%), • Annual income under Annual income under

NT$500,000-1 million (75%) NT$500,000-1 million (75%) (US$14,500-29,000)(US$14,500-29,000)

• Primary purchases: Daily Primary purchases: Daily necessities (68%), Others necessities (68%), Others (49%), Food (21%)(49%), Food (21%)

0000 0000 0000 00000000 0000 0000 0000


PlatinumPlatinum

56%56% 0000 0000 0000 00000000 0000 0000 0000

GoldGold0000 0000 0000 00000000 0000 0000 0000

GoldGold

0000 0000 0000 00000000 0000 0000 0000

RegularRegular0000 0000 0000 00000000 0000 0000 0000

RegularRegular

33%33%11%11%




• Females (83%), Females (83%), • Age 41-45 (38%), 46-50 (38%)Age 41-45 (38%), 46-50 (38%)• College educated (73%), College educated (73%), • Annual income under NT$2-4 Annual income under NT$2-4

million (81%) (US$58,000-million (81%) (US$58,000-116,000)116,000)

• Primary purchases: Clothing Primary purchases: Clothing (71%), Jewelry (59%), Food (71%), Jewelry (59%), Food (33%)(33%)

0000 0000 0000 00000000 0000 0000 0000


PlatinumPlatinum

19%19% 0000 0000 0000 00000000 0000 0000 0000

GoldGold0000 0000 0000 00000000 0000 0000 0000

GoldGold

0000 0000 0000 00000000 0000 0000 0000

RegularRegular0000 0000 0000 00000000 0000 0000 0000

RegularRegular

49%49%32%32%


Managing Islands of InformationManaging Islands of Information

• Networked IT infrastructureNetworked IT infrastructure

– Centralized vs. distributedCentralized vs. distributed

• TrendsTrends

– Centralized data centerCentralized data center

– Decentralized processing to business operating unitsDecentralized processing to business operating units

– Client-server architectureClient-server architecture

• Fat-client vs. thin-clientFat-client vs. thin-client

– Networked computing architecture and reengineeringNetworked computing architecture and reengineering

– Distributed IT architecture and the Internet Distributed IT architecture and the Internet


OperationsOperations

• Managing operations includesManaging operations includes

– Solving operational problemsSolving operational problems

– Defining operational measuresDefining operational measures

– Applying management principles to managing the IT Applying management principles to managing the IT resourcesresources

Ensuring the IT resources are accessible to and supporting Ensuring the IT resources are accessible to and supporting the right people in and outside the organizationthe right people in and outside the organization


Major Operational IssuesMajor Operational Issues

• Outsourcing IS functionsOutsourcing IS functions

• Information security in the Internet ageInformation security in the Internet age

• Planning for business continuityPlanning for business continuity


Managing IT OutsourcingManaging IT Outsourcing


Managing IT OutsourcingManaging IT Outsourcing

• ““Goal is to improve return on assets by moving these assets Goal is to improve return on assets by moving these assets off the financial books while retaining control over their use.”off the financial books while retaining control over their use.”

Sprague and McNurlin, 1993Sprague and McNurlin, 1993

• Drivers: Focus and ValueDrivers: Focus and Value• Reasons for outsourcing:Reasons for outsourcing:

– Cost-effective access to specialized or occasionally needed Cost-effective access to specialized or occasionally needed computing power (operational) or systems development computing power (operational) or systems development skills (development)skills (development)

– Avoidance of building in-house IT skills and skill setsAvoidance of building in-house IT skills and skill sets– Access to special functional capabilitiesAccess to special functional capabilities


Market ChangesMarket Changes

• Changing customer and vendor relationshipsChanging customer and vendor relationships

• Changes that occur with movement to outsourcingChanges that occur with movement to outsourcing

– IS/IT management loses an increasing amount of control IS/IT management loses an increasing amount of control because more of the activities are turned over to outsidersbecause more of the activities are turned over to outsiders

– Providers take more risks as they offer more optionsProviders take more risks as they offer more options

– Providers’ margins improve as they offer more servicesProviders’ margins improve as they offer more services

– The importance of choosing the right provider becomes The importance of choosing the right provider becomes more important due to the increased risk with outsourcingmore important due to the increased risk with outsourcing


Types of OutsourcingTypes of Outsourcing

• IT outsourcing – Moving entire IT function to an outsourcerIT outsourcing – Moving entire IT function to an outsourcer

• Transitional outsourcing – Project-base outsourcingTransitional outsourcing – Project-base outsourcing

• Best-of-breed outsourcing – Selective outsourcingBest-of-breed outsourcing – Selective outsourcing

• Shared services – Insourcing Shared services – Insourcing

• Business process outsourcing – Outsourcing all or most of a Business process outsourcing – Outsourcing all or most of a reengineered process that has a large IT componentreengineered process that has a large IT component

• e-Business outsourcing – e-business functione-Business outsourcing – e-business function


Another view on outsourcingAnother view on outsourcing


Need to compete effectivelyNeed to compete effectivelyNeed to compete effectivelyNeed to compete effectively

Motives to OutsourceMotives to Outsource

• Two major factors:Two major factors:

– Recognition of strategic alliancesRecognition of strategic alliances

• Strengthening the Strengthening the weakest link in the chainweakest link in the chain

– Finding a strong partner that complements an Finding a strong partner that complements an organization’s skillsorganization’s skills

– Changes in the technology environmentChanges in the technology environment

• New technology, new strategiesNew technology, new strategies


InformationInformationRequirementsRequirementsInformationInformation

RequirementsRequirementsEnvironmentalEnvironmental

ShiftsShiftsEnvironmentalEnvironmental

ShiftsShifts

HighHigh

LowLow

HighHighLowLow

Impact of Impact of Existing IT Existing IT

applicationsapplications

Impact of Future IT Impact of Future IT applicationsapplications

FactoryFactoryOperational ITOperational IT

SupportSupportBasic elementsBasic elements

TurnaroundTurnaroundGradual adoptionGradual adoption

StrategicStrategicStrategic IT plan, Strategic IT plan,

initiativesinitiatives

The evolution of The evolution of technology often technology often changes the strategic changes the strategic relevance of IT relevance of IT service to a firmservice to a firm

Environmental ChangesEnvironmental Changes


Drivers of OutsourcingDrivers of Outsourcing

• Concerns about Cost and QualityConcerns about Cost and Quality– High inhouse costs and less High inhouse costs and less sensitivesensitive (to (to

internal requirements) IT personnelinternal requirements) IT personnel• Lacking in current technology skillsLacking in current technology skills• Capacity utilization problemsCapacity utilization problems

– High standards (and quality) to complete High standards (and quality) to complete effectivelyeffectively

– High High stakes game stakes game to hire specialized IT to hire specialized IT people people

– Lower costs of outsourcers (generally)Lower costs of outsourcers (generally)– Greater responsiveness from outsourcer to Greater responsiveness from outsourcer to

IT needs of the organizationIT needs of the organization

American Airlines American Airlines and Sabre (co-and Sabre (co-hosting)hosting)



• Breakdown in IT performanceBreakdown in IT performance– Need to retool lacking technologyNeed to retool lacking technology

• Intense supplier pressuresIntense supplier pressures– Sales of surplus supplier capacitySales of surplus supplier capacity

• Simplified general management agendaSimplified general management agenda– Outsource non-core competence operationsOutsource non-core competence operations

• Financial factorsFinancial factors– Reduce sporadic capital investments in ITReduce sporadic capital investments in IT– Downsizing IT operating costsDownsizing IT operating costs– Greater organizational awareness of IT’s costsGreater organizational awareness of IT’s costs– More appealing for takeoversMore appealing for takeovers


• Corporate cultureCorporate culture

– Resistance to change within the organizationResistance to change within the organization

– Labor unionsLabor unions

• Eliminating an internal irritantEliminating an internal irritant

– Conflicts between users and IT staffConflicts between users and IT staff

• Other factorsOther factors

– Quick access to current technology and skillsQuick access to current technology and skills

– Need to quickly response to changes in the marketNeed to quickly response to changes in the market



Framework for OutsourcingFramework for Outsourcing

1.1. Position on the strategic gridPosition on the strategic grid

HighHigh

LowLow

HighHighLowLow







StrategicStrategicStrategic IT plan, Strategic IT plan,

initiativesinitiatives

YesYes

YesYes

DependsDepends

DependsDepends

Product differentiationProduct differentiationProduct differentiationProduct differentiation


Strategic Grid: OutsourcingStrategic Grid: Outsourcing

HighHigh

LowLowHighHighLowLow







StrategicStrategicStrategic IT plan, initiativesStrategic IT plan, initiatives

• Economies of scaleEconomies of scale• Higher-quality service and Higher-quality service and

backupbackup• Management focus Management focus

facilitatedfacilitated

• Correct internal problemCorrect internal problem• Tap cash sourceTap cash source• Cost flexibilityCost flexibility• DivestitureDivestiture

• Access to IT professionalsAccess to IT professionals• Focus on core Focus on core

competenciescompetencies• Access to current ITAccess to current IT• Reduce risk in IT Reduce risk in IT

investmentsinvestments

• Internal IT shortfallsInternal IT shortfalls• Internal IT development Internal IT development

skill shortfallsskill shortfalls



1.1. Position on strategic grid (Position on strategic grid (cont.cont.))

– Outsource operational activitiesOutsource operational activities

• More operationally dependent organizationsMore operationally dependent organizations

– Need for greater analysis when large IT budgets involvedNeed for greater analysis when large IT budgets involved

2.2. Development portfolioDevelopment portfolio

– Maintenance vs. development projectsMaintenance vs. development projects

• High structured vs. low structured development workHigh structured vs. low structured development work



3.3. Operational learningOperational learning

– Organizational assimilation of technologyOrganizational assimilation of technology

4.4. Organization’s IT architecture and infrastructureOrganization’s IT architecture and infrastructure

– Currency of architectureCurrency of architecture

5.5. Current technology in the organizationCurrent technology in the organization

– Segregated operations more easily outsourcedSegregated operations more easily outsourced


Structuring the Alliance between Structuring the Alliance between Outsourcer and “Outsourcee” Outsourcer and “Outsourcee” (Customer)(Customer)

• FactorsFactors

– Contract flexibilityContract flexibility

– Standards and controlStandards and control

– Areas to outsourceAreas to outsource

– Cost savingsCost savings

– Supplier stability and qualitySupplier stability and quality

– Management fitManagement fit

– Conversion problemsConversion problems

AllianceAlliance



• Contract flexibilityContract flexibility

– Accommodating changes in the environmentAccommodating changes in the environment

• Information needsInformation needs

• Competitive needsCompetitive needs

• Advances in ITAdvances in IT

• Standards and controlStandards and control

– Risk (i.e., lost of control, disruptions) in operationsRisk (i.e., lost of control, disruptions) in operations

– Risk in introducing innovations to the organizationRisk in introducing innovations to the organization

– Risk in revealing internal Risk in revealing internal secretssecrets



• Areas to outsourceAreas to outsource

– DetermineDetermine

• Are operations segregated or tightly embedded?Are operations segregated or tightly embedded?

• Can specialized competencies be acquired in the long Can specialized competencies be acquired in the long run?run?

• Are operations core to the organization?Are operations core to the organization?

• Cost savingsCost savings

– ObjectiveObjective evaluation of costs and savings evaluation of costs and savings



• Supplier Stability and QualitySupplier Stability and Quality

– Financial stabilityFinancial stability

• Difficult to insourceDifficult to insource

• Difficult to change outsourcersDifficult to change outsourcers

– Incompatibility between the organization and outsourcerIncompatibility between the organization and outsourcer

• TechnologyTechnology

• Organization cultureOrganization culture

• Between technology and organization’s strategyBetween technology and organization’s strategy



• Management fitManagement fit

– Compatibility between management styles and culturesCompatibility between management styles and cultures

• Conversion problemsConversion problems

– Mergers and acquisitionsMergers and acquisitions

• IncompatibilitiesIncompatibilities


Managing the AllianceManaging the Alliance

• Critical areas:Critical areas:

– CIO FunctionCIO Function

• Management (balance between organization and Management (balance between organization and outsourcer)outsourcer)

• Planning (vision)Planning (vision)

• Awareness of emerging technologiesAwareness of emerging technologies

• Continuous adaptation (evolution)Continuous adaptation (evolution)

– Performance measurementsPerformance measurements

• Essential standards, measurements and interpretationsEssential standards, measurements and interpretations


Managing the AllianceManaging the Alliance

– Mix and coordination of TasksMix and coordination of Tasks

• Development versus maintenance (portfolios)Development versus maintenance (portfolios)

– Associated risks inherent to eachAssociated risks inherent to each

– Customer-outsourcer interfaceCustomer-outsourcer interface

• Delegation of authority, not responsibilityDelegation of authority, not responsibility


Information SecurityInformation SecurityProtecting the Information ResourceProtecting the Information Resource

• Five security pillarsFive security pillars– Authentication – verifying the authenticity of the userAuthentication – verifying the authenticity of the user

• Something you know, have or are (i.e., physical attribute)Something you know, have or are (i.e., physical attribute)– Identification – identifying users to grant them appropriate Identification – identifying users to grant them appropriate

accessaccess– Privacy – protecting information from Privacy – protecting information from

being seenbeing seen– Integrity – keeping information in its Integrity – keeping information in its

original formoriginal form– Nonrepudiation – preventing parties from denying actions Nonrepudiation – preventing parties from denying actions

they have takenthey have taken

EncryptionEncryption


Management and Technical Management and Technical CountermeasuresCountermeasures

Management countermeasuresManagement countermeasures

• Evaluate return on their security expendituresEvaluate return on their security expenditures

• Conduct security auditsConduct security audits

• Do not outsource cybersecurityDo not outsource cybersecurity

• Security awareness trainingSecurity awareness training

Technical countermeasuresTechnical countermeasures

• FirewallsFirewalls

• EncryptionEncryption

• Virtual private networks (VPN)Virtual private networks (VPN)

ExpenseExpenseLikelihoodLikelihood

Balancing between expense Balancing between expense and likelihood of a threatand likelihood of a threat


Planning for Business ContinuityPlanning for Business Continuity

• Recognize threatsRecognize threats

• Contingency plans if a threat is realizedContingency plans if a threat is realized

– Alternate workspaces for people to resume workAlternate workspaces for people to resume work

– Backup IT sitesBackup IT sites

– Up-to-date evacuation plansUp-to-date evacuation plans

– Backed up computers and serversBacked up computers and servers

– Helping people cope with disasterHelping people cope with disaster


Internal and External ResourcesInternal and External Resources

InternalInternal

• Multiple data centersMultiple data centers

• Distributed processingDistributed processing

• Backup communications facilitiesBackup communications facilities

• LANsLANs

ExternalExternal

• Integrated disaster recovery servicesIntegrated disaster recovery services

• Specialized disaster recovery servicesSpecialized disaster recovery services

• Online and off-line data storageOnline and off-line data storage


R. Ching, Ph.D. MIS California State University, Sacramento 1 Week 7 Monday, October 10 Knowledge ManagementKnowledge Management Managing OperationsManaging.

Documents

knowledge linguistic

problem space slide

problem solvingheuristic

problem solvingcomplexity

problem solvingused

problem spacevalue

inference procedures

transitory slide