Management Decision and Control Systems
Management Decision and Control Systems CSC 546
Popular in Course
Popular in ComputerScienence
This 31 page Class Notes was uploaded by Jaden Jakubowski on Thursday October 15, 2015. The Class Notes belongs to CSC 546 at North Carolina State University taught by Thomas Honeycutt in Fall. Since its upload, it has received 22 views. For similar materials see /class/223830/csc-546-north-carolina-state-university in ComputerScienence at North Carolina State University.
Reviews for Management Decision and Control Systems
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 10/15/15
A Knowledge Insight Model framework for Knowledge Discovery and Data Mining Sarat M Kocherlakota Dept of Computer Science North Carolina State University Raleigh NC 27695 9195134663 smkocherunityncsuedu ABSTRACT Techniques for effective knowledge discovery and data mining primarily involve the analysis of large datasets and the discovery of unique values and relationships within these datasets pertinent to an application a speci c query or a task Statistical models machine learning algorithms and scienti c visualization are but some of the many areas of research that focus on developing suitable techniques for data analysis and exploration Speci c techniques may be effective to speci c areas of application However the existence of a general framework for data exploration would greatly enhance the effectiveness of the wide range of existing techniques for knowledge discovery and data Such a framework or model can support analysis and exploration tasks in a variety of domains and at the same time allow the application of multiple techniques to a variety of information sets We propose such a model called the Knowledge Insight Model KIM for the task of knowledge discovery and data mining The core of KIM consists of four major functions or states called the Framer Maker Finder and Sharer Each of these states or functions is designed to aid the various stages that need to be followed for knowledge exploration and discovery For instance the Framer state could encompass the design of an effective framework the Maker state could involve using effective techniques for developing the design into a working model for knowledge exploration and discovery the Finder state may involve testing and evaluation of the model for its effectiveness and nally the Sharer state could involve interaction between the various modules of the system and continual re nement of the model for higher ef ciency of the system Each state could thus be used to develop a certain aspect of the general system Additionally each state would include a broad range of effective techniques that pertain to that aspect of the system The paper explores the design of KIM with respect to developing a general model for Knowledge Discovery General Terms Management Design Keywords Knowledge Management Knowledge Discovery Data Mining Thomas L Honeycutt Dept of Computer Science North Carolina State University Raleigh NC 27695 9195157001 honeycuteosncsuedu 1 INTRODUCTION In this information age the ability to manage and utilize vast amounts of information is becoming increasingly important Managing information and knowledge has become the focus of many studies seeking to harness information in useful and ef cient ways While advances in use of data base systems focus on not just the ways in which data is stored but also the way in which it is retrieved research areas ranging from statistics to arti cial intelligence and machine learning to scienti c and information visualization focus on gaining new insight from multiple sets of information The terms Knowledge Discovery and Data Mining have gained new signi cance given the importance of managing information effectively in today s world Effective knowledge discovery techniques allow us to see new relationships and gather relevant information from various data sets This is essential given the large amounts of information that modern data mining systems have to deal with From tracking typhoons to understanding spread of diseases medical diagnosis or even unearthing relationships between unrelated data sets and many other applications knowledge discovery and data mining are fast becoming an integral part of any modern application or system The use of various methods and techniques to unearth useful knowledge or information quickly and ef ciently from information sets is fairly common Speci c techniques and methods have been devised to target speci c domain areas It is evident that there are individual complexities that are inherent in individual domain areas making it dif cult to create a generic knowledge discovery system However there seems to exist a certain lac of a general framework or model on which knowledge management or data mining systems can be based on Additionally the presence of various domain dependent applications makes it dif th to perform a broad level data mining operation or task combining techniques pertaining to multiple domain areas without a framework providing guidance as to which solution can be applied to which problem area 2 GENERAL FRAMEWORK The presence of a general framework for knowledge discovery from information sets has many advantages The chief advantage is the ability to possess various techniques and methods that pertain to various domain areas under a single framework thus making it easier for a system to decide which technique may be applied to which task A model would also allow systems to perform data mining tasks combining different techniques from multiple domain areas into a single solution It would also make it possible to develop a solution or application for a speci c data task and formulate the design for the solution The framework could also be used to oversee the development of the application or solution the testing of the solution and applying the solution to the problem or task We shall attempt to describe one such framework or model known as the Knowledge Insight Model KIM that can be used to provide a suitable framework for knowledge discovery and data mining solutions or applications 3 KNOWLEDGE INSIGHT MODEL KIlVI The Knowledge Insight Model KIM chie y comprises of four major functions These functions are labeled as the Framer problem for which a solution has to be proposed and developed and implemented the four functions handle the various tasks required to come up with the solution 31 FRAME The Framer is the function responsible for understanding the problem and designing a framework hence the name Framer within which a solution can be developed The Framer identi es the requirements of a problem and is also responsible in guiding the process of developing a solution The Framer proposes a design for the solution With respect to knowledge discovery the Framer identi es the requirements of the data mining task and the operations that need to be performed on an information set For this purpose the Framer would have access to information relating to knowledge discovery and data mining across multiple domain areas Based on this information the Framer identi es the problem as one that requires a single method or technique or one that would need a combination of multiple techniques in a speci c order This knowledge about the problem is transformed into a design or a lan 32 MAKER Once the framework has been established the Maker function is responsible for developing the solution or application based on the design put forth by the Framer The Maker functions at an implementation level wherein it puts together the working modules and components as required by the design developed by the Framer 33 FINDER Finder is a function responsible for nding resources to supplement the Maker s efforts In cases where ere exists a technique or application that performs the required operation the Finder procures this solution for the Maker thereby saving time since the application need not be developed from scratch The Finder function could also produce suitable tests that validate the solution developed by the Maker along with suggestions on where the solution might have failed If such a situation arises the Framer may have to redesign the solution if the Finder Maker combination cannot produce the complete solution With relation to Knowledge Discovery the Finder is called by the Maker to procure information regarding existing solutions or techniques The Finder also produces test cases based on knowledge about the techniques used and the relevant domain areas 34 SHARER The Sharer function as the name suggests is responsible for ensuring that all other functions can work effectively by sharing their information freely and that knowledge ows eely between all modules The Sharing function is instrumental in re ning the model and the application being developed particularly in situations where the application does not perform satisfactorily Once a satisfactory solution is developed the Sharer applies the solution to the problem area The Sharer performs best when it is not ubiquitous in nature and allows the procedures to interact with each other effortlessly With respect to the knowledge discovery and data mining one of the key functions of the Sharer is maintaining a comprehensive knowledge base containing information about the various domain areas that the model hopes to address and the various knowledge discovery techniques tools applications and solutions that exist within each domain area For instance the Framer would use the Sharer and the knowledge base to obtain information regarding existing methods and solutions to aid the Maker Additionally the Framer also uses this knowledge to generate test cases At the very least the knowledge base should possess information about how to access such information An effective knowledge base would allow easy access to required information as desired by the various functions within the mode 4 PLAN DO CHECK ACT PDCA In order to use the functions to develop solutions or applications KIM also contains a development cycle called the PlanDo Check Act PDCA During the Plan phase the problem is analyzed and a solution is designed The Framer function is most active during this phase In the Do phase the Maker develops the design into a possible solution which is tested for its validity in the Check phase of the cycle and the Finder seeks for aws in the solution In the Act phase the Sharer applies the solution to in its area of application Each cycle of understanding the problem designing and testing a solution to implementing the solution as an application could in turn traverse one or many intemal PDCA cycles The development cycle of a product or solution for knowledge discovery purposes may not necessarily end with a single PDCA cycle In case the task is not performed satisfactory re ning the solution to work more accurately becomes very important The re nement process could thus involve another PDCA cycle and 0 on until the requirements are suitably met An individual effort at solution nding would involve all four functions Framer Maker Finder Sharer working in different phases of the duration of the project where each project or task would encompass a PDCA cycle The objective of providing a broad framework is to allow multiple techniques from various domain areas to be applied to various data mining tasks or problems Although the Framer is responsible for providing a suitable framework or design it is the task of the Sharer to ensure that the original framework is adhered to The Sharer can also suggest that plans be modi ed in case modi cations or re nements are necessary The Sharer thus manages the model and its individual functions solutions within the knowledge base that solve the particul problem or tools and applications that can best perform the required task If such a solution is discovered the Maker will adapt or modify this solution to t the requirements In case a solution is not discovered the Maker is responsible for developing the appropriate application based on a plan designed by the Framer The Maker function can utilize the Finder to look for existing ar The Finder function as mentioned earlier is responsible for procuring appropriate resources for the application to make e job of the Maker much easier These resources may also include various tests and evaluations necessary in determining whether a solution developed by the Maker functions satisfactorily or not The newly developed solution is then applied to the problem area by the Sharer All this would constitute a single PDCA cycle for developing a solution 41 IMPORTANCE OF SHARER In several cases though a single PDCA cycle may not be enough to ensure a satisfactory solution or product As we mentioned earlier more PDCA cycles become necessary when re nements or adjustments need to be carried out This is particularly required if results from initial application of the solution show errors even a er a carefully managed PDCA development cycle At this juncture the Sharer assumes a more dominant role Since the Sharer manages the whole process it initiates another PDCA cycle in order to re ne the solution This would mean that the into account newer ensures that the requirements are met and the solution works in the actual scenario for which it was designed If the requirements are not met the Sharer could initiate another re nement cycle or decide that the solution has failed and that a new solution must be designed Sharing thus emerges as a multidimensional function one that can perform many roles or operations as require 5 OTHER USES OF KIM KIM is model that could very easily apply to a team of individuals assigned to a project The four individual functions could be actual people or a group of people within a team or an organization incharge of product development For example with relation to software development the Framer is the person responsible for setting the framework and the guidelines of the solutions Framers are persons capable of analyzing problems at a glance and being able to comprehend the bigger picture and the grand scheme of things Framers have the ability to understand how to apply old solutions to new problems Framers could be thought of as good designers Framers could also be regarded as those in policymaking positions Within that perspective Makers are persons adept at being able to implement designs effectively and ef ciently Makers could be thought of as good programmers Finders are persons with the ability to nd resources and solutions to problems the Makers might encounter They could be persons assisting with effective hardware support and persons involved in testing solutions developed by the Maker for faults or aws within the implementation And nally Sharers are effective project managers They could also be Framers but not necessarily so Their skills correspond to ensuring that all 0 er team members carry out their tasks ef ciently and ll in positions of the individual team members wherever necessary Thus Sharing emerges as a multi dimensional function The software development process would be carried out using one or more PDCA cycles Each stage of development could itself be modeled on an internal PDCA cycle as per the needs of the development team or organization 6 CONCLUSIONS In this paper we introduced the concept of the Knowledge Insight Model or KIM which could be used to develop a general framework for knowledge discovery applications or tasks We hope that this framework could be used to not only develop domain speci c applications but also knowledge discovery and data mining applications that function effectively across multiple domains Such applications designed using KIM would greatly enhance the effectiveness of recent advances made in knowledge management and knowledge discovery Knowledge discovery applications designed using KIM could hope to tackle large problems encompassing multiple domain areas Individual techniques and tasks could also be further re ned under this model to work more effectively We also mentioned how KIM could also be applied to the composition and working of a software development team in generating effective products KIM could also very easily be applied to a larger organization The various areas of application and levels of application only illustrate the exibility of KIM TABLE OF CONTENTS Tntrndnotinn Pattern quot39 Are Patterns Really All That Important The Virtues Of Pattern Recognition Pattern Recognition amp Decision quot1 ino What Knowledge Patterns Are There Meta K 39 Ag Playbook Patterns Must Fit The onteszt OOCUIUIampUJUJ The Limitations Of Automated Pattern Our Knowledge Pattern Taxonomy Description of Our Knowledge Patterns C 39 39 P P FPI PI I PQ AnnendiY rr To understand is to perceive patterns Isaiah Berlin Introduction Given the fast changing and ever increasing complex nature of the world gaining insight into how patterns are forming and structures are developing represents the most powerful way of managing in the new economy Winslow Farrell How Hits Happen In the rapidly emerging knowledgebased economy knowledge is a fundamental factor input and it is big Knowledge is an awesome new resource for value creation Accordingly we have begun to pay more attention to knowledge as a fuel It is now too precious to waste Too important to leave to ad hoc management A new discipline knowledge management has begun to evolve related to the methods tools and strategies for harnessing knowledge intellectual capital and intangible assets Many schools of thought exist in relation to how knowledge can be successfully harnessed for productive uses The problem is that with all the hype the clamour and the noise how does one make sense of the dissonance and begin to clearly understand the tested and proven pathways to knowledge success We think that growing an understanding of knowledge patterns and internalizing the inner logic of these frameworks is a great way to move forward Phil Jackson the famous basketball coach gets it right when he observes The idea was to code the image of a successful move into my visual memory so that when a similar situation emerged in a game it would seem to paraphrase Yogi Berra like deja vu all over again We have come to the conclusion from our research that knowledge pattern recognition is an especially critical and requisite skill for smart innovative knowledge strategy This is a key strategy for making better sense of the growing knowledge puzzle Pattern Recognition Pattern recognition has proven itself to be a seriously powerful tool for guiding action in many other fields In the military domain pattern recognition can be used to identify enemy submarines by their acoustic signature the particular pattern made by their propulsion systems In the Gulf war we saw vivid pictures of the tomahawk cruise missile in action It s guidance system is programmed with an image map of the target It locks onto this pattern in the precise delivery of its payload In modern fish plants pattern recognition is used to sort fish by type In some plyboard manufacturing plants sheets of board are graded and sorted by their pattern Financial services companies use software equipped with pattern 39 39 J to spot 39 trends and thereby intercept impending fraud Other typical applications are automatic sorting of bank bills recognition of a speaker by his or her speech recognition of abnormal electro cardiograms signals optical reading of written documents visual inspection of manufactured products for quality control Today after the September 11 terrorist attacks in the United States there is a growing deployment of facial recognition technology to h had aproyen ef cacy aeross awlde speetrunn ofhuman aeuyltles The double hellx of n ttt our eontenuon that there ls agreat future forusmg sueh an approaeh wlth lt one ean deeode We 39 wd l w our ellents Wlth the erattlng oflnnoyauye knowledge strategy Are Patterns ReaIIyAII That Important 7 1n the quotes below you ean see evldence of the faetthatlnthe nnlnds ofleadmg thnkers and praeuuoners today patterns are extremely lmponant rhls ls our argument from authonty Here are afew fayounte examples w Bnan Arthur A leadlng Eeononnlst from The Santa Fe Insutute mdgteaz quot Mcheal Mosehen one of the world39 s greatest Jugglers Mame s Law whxch states that enmpnmpmeessthgpnwet W111 dnnhze Every 18 minths Lhaught 1 was hethg ndtenznnsquot Nathan Myhryol d form er Chlef Teehnology Of cer Mlcroso Corp 7712 thud revalutum 15 mated th bxalagy dhd sezfntgdhrzthg system the search390 a sehse afpattemquot James Balley Lhey canmbule ta Lhe system as an Integrated whalequot Bn an Goodwm a Bntlsn Blol oglst patternsquot Gary Hamel Management suateglst TM m w ofknowledge pattern reeognluon as a eore eompeteney The Virtues Of Pattern Recogn ion to eommu complex moyes wltln tremendous slmpllelty elanty speed and power For examplel may tell you about eompames that are bullttolast ulltto se 3quot or bulltto mpquot al wltln a few words 1 ean eommunleate to you plctures ofthree dlfferent strategy nnr h w m wouldbe true l we spoke aboutleaslng timershanng and outsourelng as buslness models In a few r 39T39h r F r t l Vrlmw t l pl quotm t operauonal yalue Pattern Recogn ion amp Decisionmidng Pattern reeognltlon also supports effectlve andrapld deelslon maklng And one tlung ls a lt wld t r at r lemty r ean enable enueal deelslon maklng whenreverknowledge moblllzatlon ls an urgent lssue U L L othe hum wwwd suddenly ordered hls men out of an lnfemo they were battllng Klem says Thls lneldent Tu my ease dn39 39 p n n h ordered hls rnen out of me bulldlng quot 1n thls parneular case1ust as me erewreaened llne street he llvlng roorn oor ofllne house llney had been rn cavedln Had llney not uld Twwwr ml h w w h h r old supported deersrons Cogmnme sclennst Andy Can of Washlngton Unlverslty ealls people fast pattem completersquot We are really wonderful as humans at completmg patterns We get a hlnt and llnen ll m the rest We see ablacktall swlshlng around a spnng nearby Ourvery survlval depends on th5 way ofthlnkmg says Wlnslow Farrell m How Hlts Happen What Knowledge Patterns Are There Satelllte Irnageryquot u d F rVw d M w w m we sourees m h a to trylng to dercode the knowledge pattem m use on the ollner hand we have analyzed the know nowquot for yanous functlons A thrd souree ofpattem rdeas are he rnany schools ofthoughtquot eaeln wth thelr own approaeln to knowledge applreanon Vl hatwe T F W m basls ofthls paper In surnrnary we vlew best knowledge practlces m buslness nn l n n nd y Meta Knowledge Playbook Our t n th y r r V only a theoretrea1porrrt ofvxew Together they eorrrprrse a sort ofMetaKnowledge Playbook Wehave begun to rrrtegrate thrs understandmg we have aehreyed of knowledge patterrrs rrrto our strategy and praetree playbook As we do so rt wru y t ht h r H r H w 39T h mm P rexample mangle PFquot 4quot vhtr l tt as flverman tat eht The basre tdeats to orehestxate the ow of movementm orderto lure t 4 4 4 4 4 4 W W name from one ofthe most eommor pattems ofmovement the srdehrre mangle the Jordan era Chreago Bulls Wmmng performance Patterns Mist FR The Context TwFrmm Tmtr DHFWV therruse offuel strategy Dependmg on the traek the eompetrtrye lme up atthe start of the raee and other related factors ears may be put on a one two or three stop strategy for rerfuellmg A earmay be p1aeed on a one stop fuel patrtem However xf m approprrate pattem mustbe adaptedto surt the eorrtext If orre39s repertorre and 4t wru p Tu W 1 quotW b to be R d A A h m w ofexpenence Judgment andtaert knowledge In the game of golf dAfferent clubs are 77 So similarly in the knowledge game knowledge patterns are in our view the key clubs one has to have in one s bag and to learn to play And it will take a lifetime of playing and practice to truly master all dimensions of the game and to learn to play all available approach shots So we are not suggesting that applying patterns in practice is an automatic static or exact science Rather one has to have a feel for knowledge patterns and to internalize an understanding of them so that knowledge pattern recognition is fast uid and adaptive When one has achieved mastery of knowledge patterns one will have truly become knowledge wise The opposite is also true Without such mastery one is not seriously playing the game but only fooling around It is interesting that even in the sports arena today software is playing a supporting role involving pattern analysis There was for example a recent article in Strategy amp Business about the attempt by the New York Nicks basketball team to use new software developed by IBM named Scout This software is deployed in order to analyze players patterns of play Ifyou can determine that every time player x goes to the basket he uses his right hand then that s a pattern of play that could be very useful to know about The logic of developing a meta knowledge playbook based on well understood knowledge patterns therefore makes sense and has exciting application potential Yet at the end of the day the players on the court still have to play the game and execute plays well The Nicks may have used Scout However the technology can only enable and assist performance Insights have to be treated as suggestive not definitive Using pattern recognition therefore will never be the whole story to the achievement of championship level performance However it can clearly be a highly potent contributor to winning The Limitations Of Automated Pattern Recognition There has been a recent argument about pattern recognition in business as a key to learning and profitability The book Pro t Patterns by Slywotzy Morrison Moser Mundt amp Quella makes this case in a very compelling way They argue that The art of identifying understanding and exploiting patterns needs to become part of the mental process of every decision maker interested in creating sustained profit growth This we would entirely agree with They also decipher and discuss three forms of knowledge profit patterns While these are very useful knowledge patterns are not their main focus The three knowledgetoprofit patterns they cite are E Product To Customer Knowledge 7 my product business teaches me about my customer Operations To Knowledge assets to essence 3 Knowledge To Product 7 expertise crystallized E1 linl Their analysis does beg for a more exhaustive treatment Their contribution is nevertheless valuable and insightful For the purposes of our present discussion let us simply note and agree that there can be a powerful link between the ability to recognize knowledge patterns and pro tability So this gives us all the more reason and incentive to pay active attention to the mapping and application of knowledge patterns m buslness A fuller dlscusslon of knowledgertorprofltpatterns are exploranons for a subsequent wnrte paper now m development our Knowledge Pattern Taxonomy exhaustlve m mh nd tlnern Tnere ls no ranlnng or pnonty to tlne lrst The eategory desenpnons are rnere A v Th avallable for deployment dependlng othe appropnateness oftneeontent Our eunent Worklng eore taxonorny ofknowledge patterns ls tlnerefore as follows w l tr camquot new 3 J lt5 ennuonnnnan Figure 4 Knowledge Meta Pattern Array 1 Knowledge Leadershlp Knowledge Harvesting StoryTelling Intelligent AgentKnowbot Knowledge Reuse Knowledge Mining Developing Work ow Process Assets Social Network Analysis Competency Management Competitive Intelligence amp Time To Knowledge Mind To Market Acceleration Growing Customer Capital Knowledge Substitution Knowledge Network Knowledge Centre Learning Organization Communities Of Practice Knowledge Aggregation Knowledge Based Engineering Knowledge EnvironmentWorldsEcologies Smart Products amp Services Knowledge Mapping Knowledge Arbitrage Intellectual Property Development Reputational Capital IC Measurement Scorecard Product Knowledge Development Knowledge RepresentationVisualization Knowledge Agility Sense amp Respond Ideation eKnowledge ExchangeKnowledge Market Local or Indigenous Knowledge The Knowledge Toll Squeeze Description of Our Knowledge Patterns 1 Knowledge Leadership This attends to the development of teams of highly capable knowledge leaders It could involve the development of a wide range of positions Chief Knowledge Of cer knowledge architect knowledge steward and so on It could also involve the nurture of grassroots knowledge activists and champions 2 Knowledge Harvesting This approach is one of eliciting knowledge from knowledge workers so that it can be recorded and codi ed as a corporate asset It usually can involve interviewing and observing a knowledge subject and making explicit the knowhow about how to do a task that is currently undocumented A subpattems of this technique is called after actionreview and is used by the US Military and others to extract lessons learned from action carried out in the eld 3 Storytelling Storytelling is an ancient communication art We have used it to share knowledge with others throughout history It is used as Moliere says to simultaneously please instruct and educate As a tool for the socialization and extemalization of all aspects of knowledge it can be most effective David Snowden from IBM s Cynefin Centre For Organizational Complexity is a leading proponent of the use of storytelling He suggests that the combination in stories of the use of metaphor pictures and images is a more enduring way to build common understanding and focused thinking in a knowledge management program than other linear and literal communication methods Steve Denning has also championed the use of storytelling with great effectiveness at the World Bank 4 Intelligent AgentsKnowbots This involves the use of arti cially intelligent software agents as surrogate knowledge agents There are many types of agents available for use 5 Knowledge Reuse This is an approach whereby knowledge is codified shared and made available for reuse The benefit to an organization is that time can be saved by not having to reinvent the wheel It eliminates redundancy It can also be used to lift competencies across the organization by spreading knowhow where it s strong to areas where capabilities could be strengthened Examples of organizations with application case histories where this has proven to be effective would be Texas Instruments and IBM At Texas Instruments over 80 of the software code written was reported as being reused At IBM templates for responding to Requests For Proposals cut down dramatically on the time and resources needed for sales people to respond to potential customers 6 Knowledge Mining Knowledge mining is the analysis of large amounts of transaction data and information contained in knowledge bases for the extraction of useful insight in the form of developing trends patterns exploitable opportunities or anomalies 7 Developing Work ow Process Assets This would involve the mapping and embedding in software of rules roles and routing paths associated with a particular process This process knowledge then becomes a process asset a production script that is a part of an organization s infrastructural capital 8 Social Network Analysis This approach is based on insights drawn from the field of anthropology It is predicated on the notion that there are informal social networks through which much important peer to peer knowledge sharing occurs Attempts are made to understand the trusted networks and lines of in uence that are normally hidden from view This is important for nurturing change that does not run afoul of these hidden networks A leading proponent of this technique is Karen Stephenson of Imperial College London There is also beginning to emerge complimentary social network analysis software which can be used to assist analysts in such a project 9 Competencies Competency management is another tactic that has been used to address the question of what knowledge do we have or do we need In this approach job positions are profiled in terms of the knowledge requirements for various positions This can very useful for recruitment purposes You can have a better basis for matching people with required skill sets Internally it can be used to locate people who have knowledge that can useful to other areas of the enterprise It can pinpoint corporate knowledge strengths and magnify gaps that exits It can be used to help plan for meeting future needs It can also give a clearer idea of areas where training should be funded and encouraged Increasingly ERP software systems such as SAP PeopleSoft JD Edwards and Meta4 offer functionality to help HR units to do competency management 10 Competitive Intelligence Competitive Intelligence has been described as one of the fastest growing new departments among the Fortune 500 Competitive Intelligence is growing as a discipline and as a profession according to the Society Of Competitive Intelligence Professionals httpwwwsciporg It is ultimately centered on effective knowledge acquisition It focuses on scanning a company s competitive landscape for threats opportunities risks and advantages Information is gathered in a coherent fashion analyzed and interpreted in a way that will support strategic decision making There is special use software developed for assisting this function such as Knowlede X which was purchased by IBM Cipher s Knowledge Assist and Wincite Systems Wincite There is also special software being developed to support the building of war rooms storyboarding and the running of simulations Larry Kahaner s book Competitive Intelligence is a basic primer covering this field The following Radar Chart gives an example of how an organization s Knowledge Pattern capabilities can be represented graphically l Knmk at Lm t39wv A mekdgehll mmg w amnettanetnelnene w a ltnr n NHan Amlws Woulnwmaeoss Assays n Kvnw edyan39tlng elnntrntntr Figure 5 Radar Chart 7 Knowledge Performance Lndeators 7 Group A l KnowledgeLeaderslnp 2 nowledgeHarvesang 3 Storerelllng 4 IntelllgentAgentKnowbot 5 Knowledge euse o Knowledgeang 7 DeveloplngWork owProeessAssets 8 SoclalNetworkAnalysls 9 Co peteneyManagernent 10 Cornpetltwe Intelllgenee Other Knowledge Patterns we have deeoded are ll Mud To Market Aeeeleraaon w n 0 tb by wlneln ldeas go from eoneept to produet or servlce and on to the eustorner Roger D Blaekwell39s book ealled Frorn Mmd To Marketquot 1997 ls one eneellent treatrnent of ths approaeln 12 Growlng Customer Capltal rneeang tnelr ongolng needs Unlque eustornerknowledge andlnslglnt ean be D n havlng ll knowledge othe eustorner ean be golden Custornerknowledge ls not only about anticipating and serving their existing needs It can also be the foundation for capturing insight into future needs for goods and services Moreover if one has a high degree of customer interaction and intimacy the customer may be willing to lend their knowledge to the project of cocreation of new products So from the standpoint of continuous improvement and innovation there can be a signi cant competitive advantage to having the customer within your business web Customer Relationship Management software which has been evolving rapidly can be harnessed to the project of growing customer knowledge capital 13 Knowledge Substitution Knowledge substitution would for example attempt to swap smart logistics for physical product storage By knowing the timing required to service an operation one can choreograph a movement of goods or services to minimize holding patterns Essentially inventory is replaced by realtime or justin time knowledge 14 Knowledge Network or Knowledge Centre This type of knowledge pattern usually is found in professional services rms The idea is to have centralized repositories or pointers to knowledge resources so that the rm can harness all it knows in response to client needs It is an integrating technology network usually supported by an intranet or portal It functions as a clearinghouse for connecting knowledge seekers with knowledge providers Approaches used methodologies case histories lessons learned are collated so that they may be readily available to all members of the rm 15 The Learning Organization The stock of knowledge in the world is now doubling every three years or less according to some experts Every organization is being challenged to ensure that it s people learn continuously in order to keep up and to stay ahead of the competition A learning organization strategy is one where leadership is assigned responsibility for coordinating learning efforts Every attempt is made to foster a culture of active learning and to provide learners with the technology nancial resources and time to engage in learning that supports the mission of the company Every effort is made to ensure that acquisition of new knowledge and knowledge sharing and is appropriately enabled and supported As Charles Handy says The learning organization can mean two things it can mean an organization which learns andor an organization which encourages learning in its people It should mean both The Age Of Unreason l6 Communities Of Practice These are informal peer networks of knowledge workers who connect around common group needs and goals The idea is to cultivate a culture where learning is socialized and tacit plus explicit knowledge can be exchanged in a trusted community network The World Bank for example has deliberately nurtured the spawning of such Communities in recognition of their ability to socialize knowledge Inherent in the idea of communities of practice are the principles of selforganization networking and learning 17 Knowledge Aggregation Knowledge aggregation involves the building of a deep knowledge base that can serve to attract serve and sustain members of an on line community Amazoncom for example began by aggregating and organizing knowledge about books including where they could be found and what members thought about particular books 18 Knowledge Based Engineering This type of knowledge approach is most to be found in manufacturing environments Knowledge about engineering design and what was done on a particular project is documented in a knowledge base so that this knowledge is not lost when people leave It is basically the codification of specialized and complex engineering knowledge in shareable repositories 19 Knowledge EnvironmentsEcologiesWorlds This approach focuses on surrounding the knowledge worker with environments that are conducive and supportive in doing knowledge work 20 Smart Products amp Services A smart product or service is one which is intelligent It contains congealed knowhow There is wide spectrum of such products and services A bookseller who uses a permission based approach to profile customers and then alerts them when a new book on their favourite topic might be described as pursuing a smart service strategy A product such as an Otis elevator equipped with selfdiagnostics which represent the best knowledge the company has available and dials out to request service help well before onsite facilities management people are even aware of a developing problem That might be classed as a smart product Products that are designed to adapt and learn and give the user feedback using neural network technology might also be deemed to be intelligent This is a strategy that can be used for varying purposes To maintain customer loyalty to better adapt and fit customer needs to learn from customers to differentiate one s product from those of competitors to achieve higher degrees of reliability These are among the possible benefits Service amp Support subpattem This approach essentially involves enabling front line knowledge workers to have the right answers for internal or external customers This may involve the use of diagnostic knowledge bases where the accumulated knowledge about problems is stored for easy retrieval It could encompass at a more sophisticated level the use of performance support and buddy systems Tlnelatter are dreetones to experts who may have tlne answers to help the knowledge worker resolve a questron they are unable to answer on than own H MdenMam D 1y Gummy ustnmatcapual s Suhimmng Knowledge 14 Knowledge Artom r u Kv lwllagt based Engl mevl ls lea39wrm Crgnmmrmr w Kuwasdge Aagmgallcw Vb wmmn vlhe 01 Pvazllce hammers Figure 5 Radar cnart rKnowledge Perforrnanee Indeators 7 Group B 11 Mud To Market Aeeelerauon 12 Growmg Customer Caprtal 13 Knowledge Subsututlon 17 Knowled eAggregauon Based Englneenng 19 Knowle geEnvrronrnentWorldsEeologres srnartproduets amp Servlces other Knowledge Patterns we have deeoded lnclude 21 KnowledgeMapplng K wledge rnapprng ean be averv powerful teelnnrque We vlew the Human Genome Project whereby all tlne known genes m tlne human body are belng rnapped as a wrlH r organrzatron and Where rt ls ean be very a very powerful and useful rersource 1t ean n t M n eontentprovrders Allled wrtln the use ofknowledge portal workflow and doeurnent dlr Vrl lolth 1 V Wdr h eorporate knowledge rersources Thls ean save urne reduee eost and enhance the qualrtv 16 of knowledge performance In London England taxi drivers have a guidebook called The Knowledge that is a compendium of knowledge about their city that is a time honoured navigational reference resource 22 Knowledge Arbitrage Global rms on the other hand form multicultural teams that work across borders and within product lines in order to gain economies of scale and scope They also engage in what is best called knowledge arbitrage arbitrage the efficient sourcing and distribution of ideas and products drawing on the best ideas and lowest priced inputs from around the globe This means looking at the world as one economic unit not a matrix of business divisions focused on countries and regions John Thornton was chairman Goldman Sachs Asia and was selected as a Global Leader for Tomorrow by the World Economic Forum in 1993 23 Developing Intellectual Property Assets The best case where this applies is where companies hold or are generating significant Intellectual Property assets Dow Chemical is a well known example of a company that has an extensive patent portfolio By better identification organization classification valuation and management of it s intellectual capital assets it was able to use them to create more value than existed previously Essentially an intellectual property strategy is one involving paying more critical attention to intellectual property assets from creation to disposition It involves taking steps to protect and extract latent value from such assets There are software vendors such as Aurigin Systems now called Micropatents highly focused on developing enablers to facilitate the management of such assets The CEO of Aurigin recently coauthored with David Kline a relevant book on the subject book called Rembrandts In The Attic 24 Reputational Capital This is an approach where attention is focused on ensuring that the image brand and reputation of the rm is carefully developed enhanced and maintained 251 II ICapitallf 0 1 The balanced scorecard is an approach articulated by Kaplan and Norton It is one performance measurement approach that moves beyond the gap that results when only traditional indicators are used In this scheme knowledge is included as a factor in organ izational metrics It s based on the concept that what is measured gets done Therefore by having a measurement system for organizations which also takes into account their effectiveness in terms of their use of intellectual and human capital it s designed to en sure that there is a consistent alignment in the way attention is paid to harnessing knowledge along with other factor inputs Several vendors have developed software to help companies monitor their business performance using a balanced scorecard approach Emmples are Open Ratings Inc and Corvu 25 Knowledge Discovery amp Innmu39on domain Ihich can39be ponedco anoam knowledge domain 27 Knowledge RepresentationVisualization Iigul39c 7 mm1c zumm made Lr inmliiup nlm nmliun wurcu 1mm minimum LUH L quot 39 IntemLeminn LL 39 L neneupLeL 39 39 nLLnLLLnnn its implications 23 Knowledge AgilitySense amp Respond n L L LL This requires an Agility ofthought in seeing me new patterns responding to ever L L N Hwy Working to make mom for me new DL Charles Savage 29 Ideatlon Td tr n we glv s lnto the future Thls lnvolves lrnaglneenng runnlng seenanos eoneeptuahzrng future or n r ernerglng 01 0me anu rpat wsuallze e ln the marketplace by knowlng more and seemg ffectlve knowledge ereatlon and knowledge aequlsltlon Software or mindrmapplng mindrscapmg wsuallzauon rnodellng and slrnulauon ls lnereaslngly avallable to asslst sueh efforts et ready for and ereate an advantag ore elearly ltls a eornblnauon ofe 30 erKnowledge Markets Flgure 8 Knowledge ows dynamlcally through ermarkets souree Kaeteur Instltute 2004 TM ofbetter eornrneree m knowledge andlntelleetual eapltal It ls the erBay forldeas39 model There u as the knowledge store the erleamlng exchange the quesuon and answer V m H wt h w m 1m ld uhquot t M u m at our Meta Portal to erKnowledge Marketsquot at httg waw klkm org 31 Loeal or Indlgenous Knowledge enmronrnent or eulture whleh eould be extremely valuable Ethnorbotanlsts are n 4 v l A see http www desertknowledge eom au 32 The Knowledge Toll Squeeze u an 41 mm 39 77 Kmmmgambvagg m r KMMMH Exn innarllume 23 lnruernnl many new l N leerpl mmmesl twp 2s wenmaml mm as KnnvE gtAjlhl rm LJFlml nmun 2r morons DlLLaVHV l rmnnlvm n knroleessamnmarm Figures Radar cnart rKnowledge Performance Lndeators 7 Group c 21 Knowledge Mapplng 22 Knowl dgeArbrtrage 23 Intelleetual Property Development 24 Keputatronal Caprtal c Measurement Seoreeard 2o ProduetKnowledge Development 27 Knowledge RepresentanonVrsualrzanon 28 Knowledge AgrlrtySense ampRespond 29 ldeatron 30 erKnowledge ExchangeKnowledge Market Conclusions t l t game 1 mmum h w r lulu wd M quotW w H r asa assessments and knowledge strategy assessments for elrents In our andts we ask tlne n e t 10 p tnnall r m m lwn rt quotd touse knowledge strengths where it s known to count based on the experience of others The three radar charts above are examples of the kind of visual knowledge pattern profiling we produce of an organization39s knowledge strengths and weaknesses It provides senior management with a clear baseline metric for understanding and for taking further action In Knowledge Management it is important that we begin to better understand recognize internalize and harness the fundamental knowledge patterns applicable to the game Knowledge pattern recognition is now a critical skill and core competency Our Knowledge Assessment and Knowledge Strategy services based on our ongoing knowledge pattern research are a low risk high bene t intelligent rapid and affordable way to move forward References Arthur B 1998 In Fast Company Issue 18 p 93 Bailey J 1996 After Thought The Computer Challenge To Human Intelligence Basic Books New York Farrell W 1998 How Hits Happen Harper Collins New York Hamel G 2000 Leading The Revolution Harvard Business School Press Boston Jackson P Delahanty H 1995 Sacred Hoops Hyperion New York Kahaner L 1996 Competitive Intelligence Simon amp Schuster New York Kline G 2000 In Fast Company Issue 38 p 296 Moschen M 1997 In Fast Company Issue 11 p 174 Myhrvold N 1998 In Fast Company Issue 18 p 93 Rivette K G Kline D 2000 Rembrandts In The Attic Harvard Business School Press Boston Slywotzy Morrison Moser Mundt Quella 1999 Pro t Patterns Times Business Random House New York Appendix Related Resources 1 Knowledge Leadership Entovation Global Knowledge Leadership Map httpwwwentovationcomkleadmapindeXhtm Amidon D The 7 CS Of Knowledge Leadership 11111 A AI cuul 39 39 J 39 391 7cshtm 2 Knowledge Harvesting Larry Todd Wilson is a cognitive psychologist and one of the leading practitioners in this field httpwwwk 39 39 39 w tincr nm 21 3 StoryTelling httpwwwstevedenningcom Denning S 2000 The Springboard How Storytelling Ignites Action in KnowledgeEra Organizations Butterworth Heinemann October 2000 David Snowden of The IBM Cynefin Centre For Organization CompleXity is another leading theorist and practitioner He has developed a series of pioneering methods including the use of anthropological techniques for knowledge disclosure through the ASHEN model the use of stories as an advanced form of knowledge repository and the Cynefin model of formal and informal communitiesquot See also The Knowledge Socialialization project At IBM httpwwwresearchibmcomknowsoc The Storytelling Centre 7 httpwwwstorytellingcenternet 4 Intelligent Agent Knowbot Software Agents Group MIT Media Lab httpagentsmediamitedu Botspot httpwwwbotspotcom 5 Knowledge Reuse Software httpwwwcerebytecom Quality Information and Knowledge Management le KuanTsae Huang Pleasantville New York Yang W Lee Brighton Massachusetts Richard Y Wang Chestnut Hill Massachusetts Published October 1998 Prentice Hall PTR ECS Professional http wwwphptrcomptrbooksptr 0130101419html 6 Knowledge Mining VXInsight from Sandia Labs httpwwwcssandiagovprojects VXInsighthtml Knowledgist TM is a comprehensive Knowledge Mining Tool Knowledgist is a powerful personal semantic processing tool that dramatically reduces the amount of time people spend looking for relevant information on the Web an Intranet or their own computer httpwwwinventionmachinecom 7 Work ow Process Assets WARIA 7 httpwwwwariacom INSEAD httpwwwinseadfrCALTEncyclopediaComputerSciencesGroupwareWork ow WFMC 7 httpwwwwfmcorg 8 Social Network Analysis The International Network for Social Network Analysis httpwwwheinzcmueduprojectINSNA Karen Stephenson is a Professor Of management and Corporate Anthropologista and is one of the leading luminaries in this field httpwwwnetformcom Valdis Krebs is another leading practitioner of Organization Network Analysis httpwwworgnetcom 22 Linked The New Science Of Networks 7 httpwwwndedualb 9 Competency Management USNavy 7 Workforce Flaming 7 httpwwwchipsnavymilarchives027springindex27 lesworkforceplanninghtm Agilience software httpwwwagiliencecom Meta4 software httpwwwmeta4com 10 Competitive Intelligence Society For Competitive Intelligence Professionals wwwsciporg Competia httpwwwcompetia com Fuld amp Company Inc httpwwwfuldcom 11 Mind To Market Roger Blackwell httpwwwrogerblackwellcom Commercializing New Technologies Getting from Mind to Market by Vijay K Jolly See M2M 7 httpwwwm2mca Harvard 7 Mind Of The Market Laboratory httpwwwresearchmattersharvardeduprogramphpprogram7idl3 Case RealWorld Knowledge Management What s Working for HoffmanLaRoche httpwwwbusinessinnovationeycommkohtmlcaseistudieshtml See Small Business School 7 httpwwwsmallbusinessschoolorg 12 Growing Customer Capital Don Peppers amp Matha Rogers httpwww ltolcom CRM Forum httpwwwcrmforumcom 13 Knowledge Substitution Dell s BuildToOrder Business Model 7 see Harvard Working Knowledge Article httphbswkhbseduitemjhtmlid3497amptdispatch Mary Eisenhart New Spin On The Supply Chain httpwwwkmmagcomarticlesdefaultaspArticleID346 KnowledgeBased Logistics httpwww almc armv mil nln 39 W 5047htm 14 Knowledge NetworldKnowledge Centre Buckman Labs KnetiX httpwwwknowledgenurturecom Business Week Article 7 Spread The Know How httpwwwbusinessweekcom200000743b370405lhtm Radio Interview Why knowledge management is smart according to Brooke Manville partner at McKinsey amp Company httpwwwpcradiocomotrstrategyhtml 23 15 Learning Organization Peter Senge amp Society For Organizational Learning httpwwwsolonlineorgaboutsolwho Stanford Learning Organization Web http wwwstanfordedugroupSLOW David O Ulrich Professor of Business Administration httpe eced l39ms nmich J 39 r quotJ 39 39 idLasp httpwwwsumtotalcom httpwwwskillssoftcom 16 Communities Of Practice Etienne Wenger httpwwwewengercom George Por httpwwwcoilcom Communities of Practice and Pattern Language James B Smethurst httpwww motavlnr com m tavlu 39 quot7 quotI of nrm tice htm 17 Knowledge Aggregation Zhu Hongwei Michael D Siegel and Stuart E Madnick Information Aggregation A Valueadded EService Working Paper 106 2001 MIT httpebusinessmiteduresearchpapersauthorhtml Brad Hoyt httpwwwkmnewscomEditorialkmhtm 18 Knowledge Based Engineering Knowledge Technologies International KTI httpwwwktiworldcom Unigraphics httpwwwugsolutionscom Coventry University Knowledge Engineering amp Management Centre httpwwwkbecoventryaculd 19 Knowledge EnvironmentWorlds Ecologies Integrating Spatial Semantic and Social Structures for Knowledge Management Chaomei Chen Department of Information Systems amp Computing Brunel University UXbridge UK Davies J Knowledge Management Research BT Laboratories Martlesham Heath Ipswich httpwwwlabsbtcomlibrary archivehicss799il Arian Ward Work frontiers International Victoria Ward httpwww r 39 quot 39 39ve issueS issSfeaShtml Can The Design of Physical Space In uence Collaboration 20 Smart Products amp Services Stan Davis httpwwwstanmdaviscom 24 21 Knowledge Mapping Know Map magazine httpwwwknowmapcom Denham Grey Knowledge Mapping A Practical Overview Mitre httpwwwmitreorgpubsedgeapri1700damorehtm Inspiration software Mindjetcom MindMan Mindmapping software 22 Knowledge Arbitrage EEmployment Is It Time to Change the Way We Work Compelling Arguments for the Next Internet Revolution by Ron Messer MBA CMA CA httpwwwwfsorgmesserhtm 23 Intellectual Property Development httpwwwaurigincom Rembrandts in the Attic Unlocking the Hidden Value of Patents Hardcover Authors Kevin Rivette David Kline httpwwwhbspharvardeduhbspprod detailasp8990 24 Reputational Capital Thought Leader Charles J Frombrun httpwwwreputationorg httpwwwreputationinstituteorg 25 Intellectual Capital Measurement Institute For Intellectual Capital Research httpwwwbusinessmcmastercamktgnbontisic The ICM Group httpwwwicmgroupcom httpwwwknownetorg Case IC At Scandia httpwwwfpmcomcasesel3html 26 Product Knowledge Development httpwwwsopheoncom Coopers amp Edgett Product Development Institute httpwwwproddevcom httpwwwpdinstitutecom 27 Knowledge RepresentationVisualization httpwwwinXightcom Smart Maps httpwwwsmartmoneycom httpwwwidiagramcom 28Knowledge Agility Sense amp Respond Rick Dove Chairman Paradigm Shift International Senior Fellow Agility Forum Dr Charles Savage httpkeeinccomagilityhtm 25 Sense and Respond Capturing Value in the Network Era Authors Stephen P Bradley ed Richard L Nolan ed Adaptive Enterprises Creating and Leading SenseandRespond Organizations Stephan H Haeckel httpwww hbsn harvardeduproductsDWW 39 39 J quot html 29 Ideation Dr John Kao httpwwwjammingcom Peter Schwartz Global Business Network httpwwwgbncom Joey Reiman httpwwwbrighthousecom DaVid Siegel Futurize Now Leif EdVinsson httpwwwfuturepanelcomthesestheses2htm httpwww ericssnn 39 en lish article noV brain html 30 eKnowledge ExchangeeKnowledge Market Kaieteur Institute Meta Portal to eknowledge markets See Taxonomy Of SubPattems amp amp Links at httpwwwkikmorgZportalpage2htm 31 Local or Indigenous Knowledge Shaman Pharmaceuticals was engaged in trying to leverage ethnobotanical knowledge eg see httpwwwnetsciorgScienceSpecialfeaturellhtml httpwwwshamancomA Shamanhtml 32 The Knowledge Toll Squeeze Celera Genomics a private enterprise may be in a control position with regard to knowledge of the human genome for which it may be able to squeeze atoll for access to that knowledge base wwwcelera com 26 Contact Information Bryan Davis Kaieteur Institute For Knowledge Management 67 Alberta Avenue Toronto Ontario M6H 2R7 Tel4l66511837 EMail bdaViskikm org Web Site httpwwwkikmorg 2005 Kaieteur Institute For Knowledge Management This document may not be reproduced or used for any commercial purpose without the express written permission of the author and attribution Version 121 27
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'