This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Chapter 12: Enhancing Decision Making Part 1: Introduc:on to Decision Making Learning Objec:ves • Describe diﬀerent types of decisions and the decision‐ making process • Assess how informa:on systems support the ac:vi:es of managers and management decision making • Demonstrate how decision‐support systems (DSS) diﬀer from MIS and how they provide value to the business • Demonstrate how execu:ve support systems (ESS) help senior managers make beLer decisions • Evaluate the role of informa:on systems in helping people working in a group make decisions more eﬃciently • Discuss the piNalls we as humans face when making decisions, to hopefully avoid them in the future Produced by Brian D. Janz, Ph.D. 1 Decision Making and Informa:on Systems • Business value of improved decision making • Types of decisions: – Improving hundreds of thousands of “small” decisions adds up to large annual value for the business – Structured: Repe::ve and rou:ne; involve deﬁnite procedure for handling so they do not have to be treated each :me as new – Semistructured: Only part of problem has clear‐cut answer provided by accepted procedure – Unstructured: Decision maker must provide judgment, evalua:on, and insight to solve problem Decision Making and Informa:on Systems • Opera:onal managers, rank and ﬁle employees • Middle managers: – Make more structured decisions – e.g., does customer meet criteria for credit? • Senior managers: – Make more structured decisions but these may include unstructured components – e.g., why is order fulﬁllment report showing decline in Minneapolis? – Make many unstructured decisions, – e.g., should we enter a new market? Produced by Brian D. Janz, Ph.D. 2 Simon’s Four Stage Decision Making Model Decision Making and Informa:on Systems • Informa:on systems can only assist in some of the roles played by managers • Classical model of management – Five func:ons of management: Planning, organizing, coordina:ng, deciding, and controlling • More contemporary behavioral models – Actual behavior of managers appears to be less systema:c, more informal, less reﬂec:ve, more reac:ve, and less well organized than in classical model – Mintzberg’s behavioral model of managers deﬁnes 10 managerial roles falling into 3 categories Produced by Brian D. Janz, Ph.D. 3 Mintzberg’s 10 Managerial Roles • Interpersonal roles: – Figurehead – Leader – Liaison • Informa:onal roles: – Nerve center – Disseminator – Spokesperson • Decisional roles: – Entrepreneur – Disturbance handler – Resource allocator – Nego:ator Decision Making and Informa:on Systems • Three main reasons why investments in informa:on technology do not always produce posi:ve results – Informa:on quality • High‐quality decisions require high‐quality informa:on – Management ﬁlters • Managers have selec:ve aLen:on and have variety of biases that reject informa:on that does not conform to prior concep:ons • Strong forces within organiza:ons resist making decisions calling for major change – Organiza:onal culture Produced by Brian D. Janz, Ph.D. 4 Systems for Decision Support • Management informa:on systems (MIS) • Decision support systems (DSS) • Execu:ve support systems (ESS) • Group decision support systems (GDSS) Chapter 12: Enhancing Decision Making Part 2: Decision Support Technologies Produced by Brian D. Janz, Ph.D. 5 Management Informa:on Systems (MIS) • Help managers monitor and control business by providing informa:on on ﬁrm’s performance and address structured problems • Typically produce ﬁxed, regularly scheduled reports based on data from TPS • Example: California Pizza Kitchen MIS – e.g., excep:on reports: Highligh:ng excep:onal condi:ons, such as sales quotas below an:cipated level – For each restaurant, compares amount of ingredients used per ordered menu item to predeﬁned por:on measurements and iden:ﬁes restaurants with out‐of‐line por:ons Decision‐Support Systems (DSS) • Support unstructured and semistructured decisions • Model‐driven DSS – Earliest DSS were heavily model‐driven – e.g., voyage‐es:ma:ng DSS (Chapter 2) • Data‐driven DSS – Some contemporary DSS are data‐driven – Use OLAP and data mining to analyze large pools of data – e.g., business intelligence applica:ons (Chapter 6) Produced by Brian D. Janz, Ph.D. 6 Components of a DSS • Database – Used for query and analysis – Current or historical data from number of applica:ons or groups – May be small database or large data warehouse • User interface – Ofen a web interface • Sofware system – With models, data mining, other analy:cal tools Decision Support Models • Model: – Abstract representa:on that illustrates components or rela:onships of phenomenon; may be physical, mathema:cal, or verbal model – Sta:s:cal models – Op:miza:on models – Forecas:ng models – Sensi:vity analysis models Produced by Brian D. Janz, Ph.D. 7 Sensi:vity Analysis Systems for Decision Support • Using spreadsheet pivot tables to support decision making • Where do most customers come from? • Where are average purchases higher? • What :me of day do people buy? • What kinds of ads work best? – Records of online transac:ons can be analyzed using Excel – Pivot table: • Categorizes and summarizes data very quickly • Displays two or more dimensions of data in a convenient format Produced by Brian D. Janz, Ph.D. 8 Sample List of Transac:ons for Online Management Training A Pivot Table that Determines Regional Distribu:on of Customers Produced by Brian D. Janz, Ph.D. 9 Pivot Table that Examines Customer Regional Distribu:on and Adver:sing Source Systems for Decision Support • Data visualiza:on tools: – Help users see paLerns and rela:onships in large amounts of data that would be diﬃcult to discern if data were presented as tradi:onal lists of text • Geographic informa:on systems (GIS): – Category of DSS that use data visualiza:on technology to analyze and display data in form of digi:zed maps – Used for decisions that require knowledge about geographic distribu:on of people or other resources, e.g.: • Helping local governments calculate emergency response :mes to natural disasters • Help retail chains iden:fy proﬁtable new store loca:ons Produced by Brian D. Janz, Ph.D. 10 GIS‐Based Decision Support Web‐based Customer Decision‐Support Systems (CDSS) • Support decision‐making process for exis:ng or poten:al customers • Use web informa:on resources and capabili:es for interac:vity and personaliza:on to help users select products and services – e.g., search engines, intelligent agents, online catalogs, web directories, newsgroup discussions, other tools • Automobile companies that use CDSS to allow Web site visitors to conﬁgure desired car • Financial services companies with Web‐based asset‐ management tools for customers Produced by Brian D. Janz, Ph.D. 11 Group Decision Support Systems (GDSS) • Interac:ve system to facilitate solu:on of unstructured problems by group of decision makers • Hardware – computer and networking hardware, overhead projectors, display screens • GDSS sofware collects, documents, ranks, edits and stores par:cipant ideas, responses • May require facilitator and staﬀ • Enables increasing mee:ng size and increasing produc:vity • Promotes collabora:ve atmosphere, guaranteeing anonymity • Follow structured methods for organizing and evalua:ng ideas and preserving mee:ng results Execu:ve Support Systems (ESS) • Designed to help execu:ves focus on important performance indicators • Balanced scorecard method: – Measures outcomes on four dimensions: • Financial • Business process • Customer • Learning & growth – Key performance indicators (KPIs) measure each dimension • In developing an ESS, ﬁrst concern is for senior execu:ves and consultants to develop scorecard and then to automate ﬂow of informa:on for each KPI Produced by Brian D. Janz, Ph.D. 12 Role of ESS in the Firm • Used by both execu:ves and subordinates: – “Everyone’s Support System” • Drill‐down capability: Ability to move from summary informa:on to ﬁner levels of detail • Integrate data from diﬀerent func:onal systems for a ﬁrm‐wide view • Incorporate external data, e.g. stock market news, compe::ve informa:on, industry trends, legisla:ve ac:on • Include tools for modeling and analysis – Primarily for status, comparison informa:on about performance Business Value of Execu:ve Support Systems • Enables execu:ve to review more data in less :me with greater clarity than paper‐based systems – Needed ac:ons iden:ﬁed and carried out earlier • Improves management performance • Increases upper management’s span of control • Increases execu:ves’ ability to monitor ac:vi:es of lower units repor:ng to them – Also enables decision making to be decentralized and take place at lower opera:ng levels Produced by Brian D. Janz, Ph.D. 13 Part 3: Avoiding the PiNalls of Human Decision Making Decision Making: Experts vs. Novices • Experts can beLer see the big picture – Forests vs. trees • Experts possess knowledge vs. counter‐ knowledge – The importance of knowing context • Experts know their limita:ons – Focusing on our limita:ons can help us avoid problems Produced by Brian D. Janz, Ph.D. 14 Humans: Limited “Informa:on Processors” • Nobel Laureate Herb Simon (with a liLle help from Jim March) says that we display “bounded ra:onality.” • Human memory – Short‐term vs. long‐term memory • The “Magical Number 7” – Plus or minus 2… • Chunking – Experts can handle large chunks Humans: “Naïve Sta:s:cians” • Insensi:vity to sample sizes – Anecdotes vs. popula:ons • Sta:s:cal vs. prac:cal signiﬁcance • Misconcep:ons of chance – Assuming dependence where this none – Feeling overconﬁdent in groups – Coincidences and degrees of separa:on • Ignoring regression to the mean – Too much credit for excep:onal events Produced by Brian D. Janz, Ph.D. 15 More Naïve Sta:s:cs: The Erroneous Search for Causality • The Japanese eat very liLle fat and suﬀer fewer heart aLacks than the Bri:sh or Americans • The French eat a lot of fat and also suﬀer fewer heart aLacks than the Bri:sh or Americans • The Japanese drink very liLle red wine and suﬀer fewer heart aLacks than the Bri:sh or Americans • The Italians drink a lot of red wine and also suﬀer fewer heart aLacks than the Bri:sh or Americans Obvious Conclusion: Eat & drink what you like. It's speaking English that kills you. Produced by Brian D. Janz, Ph.D. 16 Mistaking Correla:on for Causality • In order for X to cause Y: – Correla:on: X and Y must be correlated • Much more diﬃcult when :me lags exist – Precedence: X must occur before Y – Eliminate other plausible explana:ons: A, B, and C can’t cause Y Humans: Limited Informa:on Processors (con:nued) • Cogni:ve Heuris:cs: “Rules of thumb” or shortcuts that we unconsciously employ • Lead to biased decisions • Lead to less‐than‐op:mal decisions Produced by Brian D. Janz, Ph.D. 17 Common Heuris:cs and Biases • The “Availability” heuris:c – Familiarity – Ease of recall • Other availability issues: – Recency eﬀects – Primacy eﬀects – Conﬁrmatory bias Other Common Heuris:cs and Biases • The “Representa:veness” heuris:c – Insensi:vity to base rates – Insensi:vity to sample sizes – The “framing” a.k.a. “anchoring and adjustment” problem Produced by Brian D. Janz, Ph.D. 18 The BoLom Line • Understanding the types of decisions we are being asked to make can help us make beLer decisions • Knowing our limita:ons as human decision makers can help us avoid decision piNalls • Decision support technologies can help us make beLer decisions Produced by Brian D. Janz, Ph.D. 19 ...
View Full Document
This note was uploaded on 02/24/2011 for the course MIS 7650 taught by Professor Janz during the Spring '11 term at U. Memphis.
- Spring '11