《大型决策支持系统总结.docx》由会员分享，可在线阅读，更多相关《大型决策支持系统总结.docx（27页珍藏版）》请在三一办公上搜索。
1、Chapter 1Management Support Systems:An Overview1. Productivity: The ratio of outputs to inputs that measures the degree ofqi innoQQ nt an arnn1711an anrl itq inrli/irli 11 rtc successoianoiganizaiionanuiisinuiviuuaipaiis2. Factors Affecting Decision-Making1) New technologies and better information d
2、istribution have resulted in more alternatives for management.2) Complex operations have increased the costs of errors, causing a chain reaction throughout the organization.3) Rapidly changing global economies and markets are producing greater uncertainty and requiring faster response in order to ma
3、intain competitive advantages.4) Increasing governmental regulation coupled with political destabilization have caused great uncertainty.3. What do Decision Support Systems Offer?1) Quick computations at a lower cost2) Group collaboration and communication3) Increased productivity4) Ready access to
4、information stored in multiple databases and data warehouse5) Ability to analyze multiple alternatives and apply risk management6) Enterprise resource management7) Tools to obtain and maintain competitive advantage4. Cognitive Limits1) The human mind has limited processing and storage capabilities.2
5、) Any single person is therefore limited in their decision making abilities.3) Collaboration with others allows for a wider range of possible answers, but will often be faced with communications problems.4) Computers improve the coordination of these activities.5) This knowledge sharing is enhanced
6、through the use of GSS, KMS, and EIS.5. Management Support SystemsThe support of management tasks by the application of technologies-Sometimes called Decision Support Systems or BusinessIntelligence6. Management Support Systems Tools1) DSS2) Management Science3) Data Mining4) Data Warehouse5) Busine
7、ss Intelligence6) OLAP7) CASE tools8) GSS9) EIS10) EIP11) ERM12) ERP13) CRM14) SCM15) KMS16) ESFigure 103 Structure of an Expert SystemConsultatidn EnvironmentDevelopment EnvirDnmentLLsm jnteHaceFacts about the specific incidentKnowledge baseFacts： What i$ known about Che domain area； Rulms： Logical
8、 reference (e,gr,；between symptoms andcausestExplanationfacilitytKnowledge engineer17) ANN18) I胸岫feAgenk19) E-pe-ofDecisionr-OperationalTycommerceInference engine draws conclusionsI ManagerialControl| InberpretEr Scheduler； ConsistencyStrategic PlanningDocumentedknowledgeStructured : (Programmed)ana
9、lysis, Investments,Knowledgeme： 口 uisitirmmistructuredAccounts receivable, accounts payable, entry Production S口 lutCfedit PDblmMerg( scheduling, inventory lontrolbudgetj -preparation,compensation, QA, -j projectHRpolicyshort-termwarehouseBiMfrecaiilingfkpimqocations,二order personnel tepoTts distrib
10、ution KiitJmlMqc： refinement andcenters jers evaluatioi,sbipl；acquisitio productns, new *planning,;rewards systemsUnstructured (Unprogrammed)Buying software, approving loans, help deskNegotiations, recruitment, hardware purchasingR&Dplanning,technology development, social responsibility plansschedul
11、ing planningHRExpertknowledge7. Enterprise Information Systems1) Evolved from Executive Information Systems combined with Web technologies2) EIPs view information across entire organizations3) Provide rapid access to detailed information through drill-down.4) Provide user-friendly interfaces through
12、 portals.5) Identifies opportunities and threats6) Specialized systems include ERM, ERP, CRM, and SCM7) Provides timely and effective corporate level tracking and control.8) Filter, compress, and track critical data and information.8. Knowledge Management Systems1) Knowledge that is organized and st
13、ored in a repository for use by an organization9. Expert Systems1) Technologies that apply reasoning methodologies in a specific domain2) Attempts to mimic human experts problem solving10. Emerging Technologies1) Grid computing2) Improved GUIs3) Model-driven architectures with code reuse4) M-based a
14、nd L-based wireless computing5) Intelligent agents6) Genetic algorithms7) Heuristics and new problem-solving techniquesChapter 2Decision-Making Systems, Models, and Support1. Decision Making1) Process of choosing amongst alternative courses of action for the purpose of attaining a goal or goals.2) T
15、he four phases of the decision process are:一 Intelligence一 Design一 Choice一 implementation2. Systems1) Structure-Inputs-Processes-Outputs-Feedback from output to decision maker2) Separated from environment by boundary3) Surrounded by environment3. System Types1) Closed system-Independent-Takes no inp
16、uts-Delivers no outputs to the environment-Black Box2) Open system-Accepts inputs-Delivers outputs to environment4. Models Used for DSS1) Iconic-Small physical replication of system2) Analog-Behavioral representation of system-May not look like system3) Quantitative (mathematical)-Demonstrates relat
17、ionships between systems5. Phases of Decision-Making1) Simons original three phases:-Intelligence-Design-Choice2) He added fourth phase later:-Implementation3) Book adds fifth stage:-Monitoring6. Decision-Making Intelligence Phase1) Scan the environment2) Analyze organizational goals3) Collect data4
18、) Identify problem5) Categorize problem-Programmed and non-programmed-Decomposed into smaller parts6) Assess ownership and responsibility for problem resolution7. Decision-Making Design Phase1) Develop alternative courses of action2) Analyze potential solutions3) Create model4) Test for feasibility5
19、) Validate results6) Select a principle of choice-Incorporate into models-Risk assessment and acceptance8. Decision-Making Choice Phase1) Principle of choice-Describes acceptability of a solution approach2) Normative Models-Optimization-Rationalization.Courses of action are known quantity.Options ra
20、nked from best to worse-Suboptimization.Decisions made in separate parts of organization without consideration of whole3) Descriptive Models(1) Describe how things are believed to be(2) Typically, mathematically based(3) Applies single set of alternatives(4) Examples:-Simulations-What-if scenarios-C
21、ognitive map-Narratives4) Developing Alternatives(1)Generation of alternatives-May be automatic or manual-May be legion, leading to information overload-Scenarios-Evaluate with heuristics-Outcome measured by goal attainment9. Decision-Making Choice Phase1) Decision making with commitment to act2) De
22、termine courses of action-Analytical techniques-Algorithms-Heuristics-Blind searches3) Analyze for robustness10. Decision-Making Implementation Phase1) Putting solution to work2) Vague boundaries which include:-Dealing with resistance to change-User training-Upper management support11. Decision Supp
23、ort Systems1) Intelligence Phase-Automatic2) Data Mining-Expert systems, CRM, neural networks-Manual3) Design Phase-Financial and forecasting models-Generation of alternatives by expert system-Relationship identification through OLAP and data mining-Business process models from CRM, RMS, ERP, and SC
24、M4) Choice Phase-Identification of best alternative-Identification of good enough alternative-What-if analysis-May use KMS, GSS, CRM, ERP, and SCM systems5) Implementation Phase-Improved communications-Collaboration-Training-Supported by KMS, expert systems, GSSANNMISData Mining, OLAPEIS ERRGSS EIS
25、SCM CRM ERP KMS Management ScienceI DSS ESCRMJ SCM1) A DSS is a methodology that supports decision-making.2) It is:-Flexible;-Adaptive;-Interactive;-GUI-based;Iterative;Employs modeling.Data: external and internalupoOthercomputer-basedsystemsDatamanagementModelmanagementKnowledge-basedsubsystemsUser
26、interfaceInternet,intranets,extranetsExternalmodels3) Components of 曲s Subsystems:Cota manacamardManager (user)M-Model management M-User interface-Knowledge Management and organizational knowledge base2. Data Management Subsystem1) Components:- Database-Database management system-Data directory-Quer
27、y facility2) DatabaseInterrelated data extracted from various sources, stored for use by the organization, and queried-Internal data, usually from TPS-External data from government agencies, trade associations, market research firms, forecasting firms3)Data Directoryanagey DBM y MBMCatalog of all da
28、ta-Contains data definitions-Answers questions about the availability of data items-Source-Meaning-Allows for additions, removals, and alterationsPrivate, personalrganizational Knowled9& BsseI Decision support databaseDatabase mansgement system Retrieval Inquiry Update Report generation DeleteI Corp
29、orate data warehouseInterfacemanagennentModelmanagementHierarchical Network Relational Object orientedK nowledge-basedsubsystemFigure 5.1DatabaseStructuresCustomer Recordsb. HisrarthicslNameQuaiHitVNutB成UVaShfircution integrationsand commanCl Nstwcrlc nagement ents: toduct e management System1Custon
30、ner NumberDustcxrier NaProduct NumberProduct NamaGreenBrownBlackWhiteM.1S.1T.1U 1NutBottWasherScrewA. RclBtianAlCustomer NamePBuCtNumberQuantityGreenM.110BrownS.1300Gceen1170VWiiteS.150Green&1250Brown11120GrawnU 150Usage AecordsBncwnttT1S 1U 1*1WasherBoltScrewilli即300 I503. Model Ma1) Compor-Model b
31、as-Model bas-Modeling language -Model directory-Model exe2) ModelsStrategic-Supports top management decisionsTactical-Used primarily by middle management to allocate resources Operational-Supports daily activitiesAnalytical-Used to perform analysis of data3) Functions:-Model creation-Model updates-M
32、odel data manipulation-Generation of new routines4) Model directory:-Catalog of models-Definitions5) Model Management ActivitiesModel execution-Controls running of modelModel command processor-Receives model instructions from user interface-Routes instructions to MBMS or module execution or integrat
33、ion functionsModel integration-Combines several models operations4. User Interface System1) GUI2) Natural language processor3) Interacts with model management and data management subsystems4) Examples-Speech recognition-Display panel- Tactile interfaces5. Knowledge-Based Management System1) Expert o
34、r intelligent agent system component2) Complex problem solving3) Enhances operations of other components4) May consist of several systems5) Often text-oriented DSS6. DSS Classifications1) Alter-Extent to which outputs can directly support or determine the decision-Data oriented or model oriented2) H
35、olsapple and Whinston-Text oriented, database oriented, spreadsheet oriented, solver oriented, rule oriented, or compound3) Donovan and Madnick-Institutional-Problems of recurring nature4) GSS v. Individual DSS-Decisions made by entire group or by lone decision maker5) Custom made v. vendor ready ma
36、de-Generic DSS may be modified for useFigure 3.7Summary of DSS CapabilitiesCreate variety H D6S Applicawns specifc DSS quickly and 朝必FscifitHte rtRraitrtile rlemign pracEssFar rntfiriE use, e 硕 icflUjn, and CDruruCLiDrt of DBStn 日 d data sources, types, and 汩rwits for & woriety 6 pWtiTS und concerts
37、Access to & /aridty .心.report/diEpiaC. ww/BffcJttjnt (fefs handling4 IMu池1S fnanagemEHt (unGtfdr*5. Voxty 航 1叫g dK4 WWSand: Analysis7 Irachngof drAA l攻景DSSlg厢 and ndcjw 曲La supportModvls1. Library B models to cirestitute b mndpl banfie. insn也icsb. mtai心.catalog.integrateC. tdnriKipreprograi nmEd) li
38、brary?Injrjng fotilrty3. MndRl-mjiriipRjlaCicm snd ueb laxity4 Modsl bBGB manaqoiiurx ftjnctiDns5 Model1 dCKurnC*rtWr*iEk Iracking ! rrTjdul uSOgv7 Flexible ahd edpti 陌 mndet support2. DSS Models1) Algorithm-based models2) Statistic-based models3) Linear programming models4) Graphical models5) Quant
39、itative models6) Qualitative models7) Simulation models3. Problem Identification1) Environmental scanning and analysis2) Business intelligence3) Identify variables and relationships-Influence diagrams-Cognitive maps4) Forecasting-Fueled by e-commerce-Increased amounts of information available throug
40、h technology4. Static Models1) Single photograph of situation2) Single interval3) Time can be rolled forward, a photo at a time4) Usually repeatable5) Steady state-Optimal operating parameters-Continuous-Unvarying-Primary tool for process design5. Dynamic Model1) Represent changing situations2) Time
41、 dependent3) Varying conditions4) Generate and use trends5) Occurrence may not repeat6. Decision-Making1) Certainty-Assume complete knowledge-All potential outcomes known-Easy to develop-Resolution determined easilyUncertainty-Several outcomes for each decision-Probability of occurrence of each outc
42、ome unknown-Insufficient information2) Probabilistic Decision-Making-Decision under risk-Probability of each of several possible outcomes occurring-Risk analysis7. Influence Diagrams1) Graphical representation of model2) Provides relationship framework3) Examines dependencies of variables4) Any leve
43、l of detail5) Shows impact of change6) Shows what-if analysisVariables:IntermediateDecision C ) or uncontrollableArrows indicate type of relationship and direction of influenceCertaintyUncertaintyFigure 4.2 An Influence Diagram for the Profit ModelincomeAmount used in advertisementProfitModeling with Spreadsheets and easy to use7) Flexible8) End-user modeling tool9) Allows linear programming and regression analysis10) Features what-if analysis, data managem
